C'est la Z

StuyCS in Boston - connecting generations

Wed, 06 May 2026 16:04:16 -0400

Devorah and I are spending the week in Boston, well techincally Cambridge. We came up last year around this time and had a big StuyCS meet up hosted at Formlabs. I was talking to my friends in the Stuyvesant HS Alumni Association figuring there were probably a lot of StuyCS grads in the Boston area and we agreed that we should do a meetup there. Then a couple of grads offered the space at Formlabs and we were off to the races.

It was a great event with around 50 attendees but there were a couple of issues with the trip. First, as all true New Yorkers, we have relatives in the area and we weren't able to see anyone as the trip was so short.

The next problem was for me personally - usually, when I organize these StuyCS meetups we get a good turnout and then at the event, I feel like I'm a parent hosting a Bar Mitzvah - I don't get to spend any time with anyone. I get a quick hello and how are you doing but then I have to move to the next person. I'd like some quality time with my students at these events.

Another problem we discovered was the introvert problem. Groups of attendees would find the couple of peopel they knew and spend the whole evening with them. Now, it's great when these are people that haven't seen each other in a while but I really am trying to facilitate them expanding their networks. Also, and very related is the fact that many introverts won't be comfortable even coming out to a big event like the ones I'm throwing even though they're actually pretty low key and comprised of people similar to those in the high school classes they enjoyed back in the day.

The solution, or attempt at solution? Devorah and I came up for a full week. We spent the first couple of days with family. With her aunt and cousin, and on my side, I recently discovered I have a second cousin once removed in the area and was able to connect with him and his family and his father - my direct cousin whom I haven't seen in over 50 years (family falling out that is now happily mended).

Then we scheduled a couple of small scale dinners one of 7 people and one of 8. I've been getting updates from my networ via emails and have been asking who's interested in small group gatherings. I made a random selection which fortunately, also had some (but not all) of the former students I wanted to have catch up time with along wit holder and younger Stuy grads.

The first restaurant was at Silk Road a Uygher restaurant near our hotel. Starting with Devorah and me from the '80s we had grads through the generations up to one just graduating college now.

The following night we went to an Afghani place - The Helmand with a new group, again ranging from my earliest students to one who attended Stuy after I had left.

I really enjoyed both dinners - got to catch up on a deeper level with some former students that I really care about and got to know some new people a little better.

I'm really hoping they enjoyed it as well and also left with new friends. I'll find out for sure when I start asking them about it.

Also, both restaurants were excellent.

Today was our "Stuy Free" day but Devorah and I were able to meet up with some CS Ed friends. Fellow retired CS teachers Alfred Thompson and Neil Plotnick along with Charlie Meyer of Pickcode joined us at Mamaleh's - a Jewish deli that I can also recommend.

So, I'm hoping that those small group dinners hit the mark. If so, the plan is to start arranging them in NYC when I'm back and then in San Fran when I'm there visiting next.

In the meantime, I'm excited for the big meetup at Formlabs tomorrow.

Stan Teitel, in memoriam

Tue, 21 Apr 2026 18:45:26 -0400

Earlier today, I was informed that Stan Teitel, former principal of Stuyvesant high school had passed away at age 77.

Stan started his career as a teacher in 1971 joining Stuy in the Chem/Physics department in 1983 - my senior year, though our paths never crossed. He became the Chem/Phys AP in 1997 and then Principal in September 2000, right before 9/11. He retired in August 2012 after the DOE, in my opinion did him dirty but we'll get to that in a bit.

Stan and I butted heads numerous times and as a young teacher trying to build CS from scratch there were plenty of itmes when I disagreed with his decisions but wether I was right or I was wrong (and we'll get to that as well) Stan loved Stuy, Loved the kids, and on a personal level, always had his teacher's backs. In an era where Principals were all too frequently business leaders Stuy was lucky to have someone who was a teacher first and who really cared.

On a professional level, as I said, we butted heads.

When I created what was to be Stuy's intro course and I was trying to award students graduation credit, at first he wouldn't agree to it. When I tried to award students credit for their senior tech classes by taking my advanced electives, I also hit a roadblock.

There were other instances but these two are enough to make the point.

As I got older and less dumb, however, I started to realize the impossible position Stan was trying to manage a school with 1000 kids per grade, hundreds of teachers, and loads of internal personalities and politics. The truth is that he indeed was really supportive of what I was trying to build. Stan wanted to support CS but at the same time, he had to balance that with an aging tech / industrial arts department full of tenured teachers who wouldn't let go of the past.

He originally didn't allow credit for my intro class but when a tech teacher retired, he let me replace him with a CS teacher since we couldn't easily find a new drafting teacher anyway. For that teacher's students, they got credit for CS instead of half of drafting. It seemed a little arbitrary and silly but looking back, I'm pretty sure he realized that what he did opened the flood gates and that the student and parent demand would ultimately allow him to make CS a requirement instead of half of drafting while letting him keep the peace within the school.

Something similar happened with the senior electives. I didn't realize it at the time, but Stan was a whole lot more supportive of CS than I realized.

By the time Stan retired from Stuy, Intro CS was a required class which replaced half of drafting and students could take senior CS electives and they'd count for the student's required senior tech credit.

On the personal side, as I said, he always had his teachers backs. When my mom died we were in a precarious child care situation as she was our child care. Stan just pulled me aside and told me to set up the schedule however I needed. I set it up so my last period was preiod 7 instead of 8 so in case I had to leave early to get the kids I would be able to. I never had to but it was nice to know that my principal had my back.

There was another instance, one where Stan never got credit and in fact got flack. The Board of Ed had just made some new demands on teachers. I forget what it was but it was going to end up in something like a 4 or 5 extra hours of work for each teacher.

Stan set something up so that we just had to spend a little extra time during one meeting - maybe an extra hour at most to get the job done. He got a lot of flak for that because teachers thought he was imposing extra work. I only knew that in fact he was saving us because I knew what the DOE directive was through my wife who was an AP at another school.

Stan also tried some new initiatives that either did't take or got more pushback than they deserved. He pushed creating a financial literacy course - something many students needed then and still need today. He also tried to create more balance and equity within Stuy's Advanced Placement classes. He worked on craeting an environment where students didn't overload themselves and also where as many students as possible had a chance to to take what they wanted or needed.

In 2012, Stan retired right before the start of the school year. Word was that he was forced into retirment by the DOE. There was an issue with cheating on regents exams the prior June. Of course neither Stan nor any adult had anything to do with the incident and in fact Stan thoroughly investigated the incident and in fact was able to identify the culprits. After the fact, I read the entire DOE report on the entire thing and they absolutely railroaded both Stan and one of his assistant principals. We all knew it at the time but there was nothing anyone could do. Stan entered retirement after a 10 year tenure as principal and his assistant principal was sent to a "rubber room" where they spend two years until they were exonerated.

So today I was saddened to learn that Stan had passed.

He spent close to thirty years at Stuy as a teacher, AP, and Principal. He was the principal that established computer science as a requirement and was the last, maybe only Stuy principal to spent a significant amount of time at Stuy as a teacher first.

It took until I got older to appreciate the challenges Stan faced and just how much he supported me and my work but I always appreciated how much he cared for the school, the kids and the teachers. They don't seem to make principals like that anymore.

Send Lawyers, VCs, and Money

Mon, 16 Mar 2026 07:53:51 -0400

So, one of the things I've been doing since retirement is working for and with my alumni community. We've held focussed events like an AI panel and a blockchain talk and have had meetups in New York and beyond - specifically Boston and San Francisco.

Last Thursday we focussed on entrepreneurship.

The idea wasn't mine. Stuy grad, Jae Zhou, a lawyer at Morrison Forester (MOFO) approached me with the idea about a year ago. She, at MOFO had already been putting on an event and she thought it would be great for the Stuy Alumni community.

It involved a role play on negotiating a term sheet. A term sheet is a non-binding agreement between multiple parties, specifically in this case, an agreement between an investor (a venture capitalist) and a founder describing the terms under which the VC will invest capital in the founder's company.

Yes, it entails the amount of the investment, ownership terms for the investor and the basics but there are also a lot of subtleties like the possibility of changing the founders vetting terms or things like board composition.

The event was to be a role play on stage so we had to come up with a VC and a founder.

Fortunatly, we've got plenty. In addition to Jae and her colleague Jesse, we invited Dan Borok of nvp capital and Sim Blaustein of 1754 Ventures as our VCs. By having two we were able to role play two term sheet scenarios. Our founder was Tim Novikoff. Tim didn't go to Stuy but he taught with me for a few years and is a good guy and good friend. Coincidentally, just as we were setting up this event, Tim was in the middle of funding deals for his current startup, SuperTeacher.

The way it worked is that each VC presented a term sheet to Tim. One was a SAFE sheet - Simple Agreement for Future Equity - I'm told the current rage in startup investing. The other was a more traditional equity round. One of the percieved advantages of a SAFE offer is that it can typically be closed much more quickly than a traditional round but hidden in the details can be some high costs. Of course, to set up an educational role play, both the term sheets were loaded with some poison pills.

Dan went first and discussed his term sheet, the SAFE offer with Tim. As he tried to entice Tim to sign his deal, both Dan and Tim, and in fact, the rest of the panel broke the fourth wall to share real life stories, and to comment on the good and bad. Afterwards Sim and Tim discussed the other term sheet.

As it turned out, all the panelists had great comic chops and were both engaging and amusing while giving us, the audience some really great information and advice.

They pointed out red flags in both styles of offers, discussed strageties that a founder might use to help leverage a better deal, and even talked about little discussed things like how often a funder just walks away from an offer when a founder pushes back.

At the end, it was question and answer time with the audience - mostly Stuy grads but there were at least 15 current students there as well.

A lot of great information was shared. One big one was when a legal team was critical and when not so much (and this was brought up and emphasized by the non-lawyers and agreed on by all) another was how important it was for a founder to have stashed away a decent amount of cash - to keep things running while trying to raise.

That last point relates to the age old question from over anxious youngsters - why not just drop out and start my company?

This was addressed by the panel as they talked about the typical unicorns - Gates, Zuckerberg, Bezos etc. - they all came from wealthy families - there was always a safety net.

Jesse, from MOFO made a great point - in the NBA, Lebron James knew he was going to be a star right out of high school but part of the reason he knew was because everyone else knew as well. If your startup isn't Lebron - if it's not getting they hype and interest from inventors, then maybe stay in school.

Super event. I'm greatful to Jae for suggesting it and all the participants, Jae, Jesse, Dan, Sim, and Tim for executing such a unique and educational experience for our community.

More on why teaching CS matters

Thu, 26 Feb 2026 13:12:57 -0400

I wanted to follow up on my earlier post on why teaching CS still matters.

In that post, I cited a couple of former students. Not only former students but particularly high achieving ones. One friend commented on that - the idea that my thesis may makes sense for those high achievers but not everyone.

I maintain that CS, if taught right is still important for all students.

It's about thinking and problem solving and that exploring programming can help build those thinking and problem solving skills.

Let's look at a couple of other situations to illustrate the point.

A number of years ago I was talking to yet another student - yes, another super high achiever. This one was a math whiz. We're talking IMO level. Some time after working with a number of other top young math minds from other countries we were talking about his experiences.

He noted, that in his experience, in the USA, the primary prolem solving appraoch that was taught was algebra based. This was his and his friend's experience in New York and also what he saw from other top young American mathematicians. In contrast, he noticed that the mathematicians he got to know from other countries were more likely to take other approaches to solving problems - a geometrical one for example.

Kids across the world learn math but aparently, in different locations, with different foci, the kids end up approaching problems differently - thinking differently.

A number of years later, my chairman - math chair, went to an Eastern European country for a few weeks to learn about how they taught high school math. Her takeaways was that they don't do nearly as much as we do in terms of topics but they dive much deeper into what they do and not as much of a focus on Algebra, which we treast as a major piece pretty much every high school class.

This supports my student's experience.

Now, I'm not arguing that one way is better than the other, though my gut says that doing less to more depth is probably better than our current high school math progression, it's that, yes, we're trying to teach math, but we're really trying to teach them to think.

The calculator didn't make math education obsolete but people are quick to note that the calculator to math isn't the same as generative AI to CS. It isn't but there have been computer tools available to do everything through high school math and well beyond and none of those made high school math obsolete.

I'll finish this post with one more thought. Generative AI won't make CS Education any less important but it should force us to re-examine what we do and how we're doing it.

We want to focus on building thinkers but we can look at the past for an example of just that done wrong.

High school geometry.

It's called geometry but, at least in my opinion, it really isn't about geometry at all. It's a course in deductive reasoning, problem solvin and logic using Euclidean Geometry as the platform to explore said reasoning.

Unfortuantely, in most implementations that part is lost. It first became a course in the basics of Euclidean Geometry and then, in many cases devolved into a course in algebraic aspects of Euclidean Geometry.

Missed opportunity.

I know that in some pockets, CS teachers will do better. I'm confident that the home grown classes at Stuy will stand the test of time and I know some other teachers at other schools who, if given the opportunity will create great experiences for their students.

The question is, what will the gorillas in the room do - the College Board, code.org and the outher outsized influencers.

SIGCSE 2026 part 3 - Sessons and BOFs

Tue, 24 Feb 2026 15:25:26 -0400

Still in St. Louis but we're now up to sessions and BOFs.

On Thursday, I attended two sponsor sessions on AI. One by GitHub and the other Google. In retrospect, they were both really the same session - on how their developers and other professionals are using AI. The big difference? Do you want your AI co-pilot flavored or Gemini/Antigravity?

The key points from the GitHub session was that when using AI, context is everything - it's easy to say "build me a minecraft code" because the term "minecraft code" brings all the context. On the other hand, asking AI to build something new is much more challenging and requires that the engineer actually has a strong tech / CS background.

Google had a similar point of view while explicitly saying that they expect their engineers to be able to leverage AI.

An interesting point is that both seemed to indicate that there would still be a big need for CS grads but that they'd have to be able to use AI. This is also in line with Titus Winters keynote. The trouble is, companies are seriously slowing down their hiring of entry level engineers.

Mixed messages here at best and in my opinion, the industry is going to be ina bind whe all those mid and upper level engineers retire and there won't be anyone to replace them.

I went to both of these sessions to get a better feel to of what the workplace expected so they were worthwhile if redundant. They also made clear that the companies are yet again talking out of both sides of their mouths - we need tech grads but we're not hiring them.

I lost a ton of respect for Google back when they laid off a huge number of employees not because they were losing money but because they were making less money than before. While their talk was interesting, it didn't do anything to improve my image of the company that used to have a motto of "don't be evil."

I also attended a session by the ACM Gen AI task force. In it, they demo'd 6 tools and techniques educators used in their classes. Fun and interesting session. You can check out the six demo'd tool along with a bunch of others here: https://acm-education-genai-task-force.github.io/approaches.html. Another winner of the session. Some interesting ideas being shared by our colleagues.

On Friday, I attended a session on Career Paths in Tech.

One talk was on how students who partake in extracurricular activities like hackathons are more likely to get internships. No real surprise but the author shared some interesting data. One aspect not covered that I was interested in though, was the challenges of student involvement in said activities at commuter schools.

The second talk was was on TicTok's role as a source for information on internships. The only interesting point here was that the info on TicTok was mostly on job listings and salary but not on specifics. I left feeling this was a negative result, that is, don't use TicTok, it isn't really helpful.

The third talk was on the "open source resume" by Code Day. Basically the author curated internships for students at institutions where companies don't normally recruit. They formed hiring agreements with the companies "if the students show x,y,z you'll interview them." Of course, this led to success. I say of course becuase I saw this at Hunter an CUNY in general - they have some great students but the "top" companies won't look at them. If you can get those company in the door, they'll find some great students.

Actually, I have more to say about this talk and Code Day's program but I'll save it for a post of its own.

The other talk was on visualizing pathways between CS courses and careers. The author had built a tool to visually show students the relationship between any given CS class in a school's sequence and what they'll need for the workforce.

It was an interesting talk and interesting tool but I wonder about how accurate the maps are. An algorithms course taught one way could prepare students incredibly well for industry but teach it another way, not so much. Still a great attempt to show students which courses are indeed valuable for any particular career path.

I also attended a talk on Pedagogy in Theory of Compution and Algorithms but was disappointed. The talk was fine but it was really just professors discussing generic teaching techniques - nothing specific to algo or theory.

Saturday consisted of two sessions.

One was on AI tools for middle schoolers. I enjoyed the session but have to do a deeper dive on the tools - are they just surface tools that have the kids playing with AI or are they exposing the "how?"

The final session was "It seemed like a good idea at the time." My favorite SIGCSE session. Presenters share an idea that they had for their classes that went horribly awry and well, hilarity ensues.

Great session with some great discussion at the end and a great forum where educators who aren't afraid of trying something new can share.

I can't do justice to the sessions in a blog post but if you go to SIGCSE in future years, I highly recommend this session.

Finally we have the BOFs - birds of a feather sessions - basically group discussions focussed on a topic.

I attended three.

First was What do you mean by "learn how to use AI." Great small group discussions. Even more so because I was in a group with Briana Morrison and Corey Bart - two friends that I had not been able to catch up with in a long time and two really sharp people who's opinions I value. There were also a bunch of other great people in my group. It was a treat for me to hear there thoughts and to share ideas.

Second was "Sticky Analogies" run by Dan Garcia among others. This was me rewarding myself. After going to sessions all day that I thought would be good for me and good for the few projects I'm working on, I decided to treat myself to a fun session. It was. A bunch of educators sharing analogies we used in classes.

The final BOF I attended was titled "All Scripture is Useful for Teaching: CS Education through a Christian Worldview."

Why would I, a Jewish Athiest with an extremely strong Jewish identity and, if I do say so myself, one who's pretty well versed in my religion as well as culture attend this session?

Two reasons. One positive and one negative.

The negative - I was put off by the way it was advertised. While most affinity groups - LGBTQ+, Black CS Educators, etc. have explicitly inclusive langauge in their session descriptions, that is "this session is for members of the LGBTQ+ community, allies, and those interested in learning more" or something similar, this session seemed to advertise as being only for Christians. That was offputting.

The positive reason was that I was interested in how someone who was a devout Christian of any denomination would approach CS education.

So, how did it go? Well, other than three things, the session wasn't specifically Christian. They organizers at the end mentioned that attendees, if they wanted could join a Christian CS Educators email list and also had Cross stickers that attendees could take for their badges if they wanted to easily identify each other. That was two and both were done in a non offensive way.

The third was when they described what the BOF was all about, they basically shared a list of questions that were mostly based off of Amy Ko's keynote two days prior. They said that they wanted to focus on the love and that most of what Ko spoke about was consistent with their worldview and wanted to focus on those parts. That little bit - mentioning that parts were inconsistent with their worldview was the third thing.

Take out those three remarks and you'd have a session of people talking about Amy Ko's keynote.

While the parts that were inconsistent with their worldview are things I believe in I respect that they were setting up the BOF to be inclusive and to focus on shared positive values.

The session then went to two small group sessions,

If I'm being honest, my first small group was a disappointment. We went around, as per instructions, each introducing ourselves. We were then supposed to get to the group questions - how do you show love for your students, your students for each other, and so on.

I introduced myself but then the next person spent the rest of the session basically explaining how I as an atheist am clearly wrong and even gave me a card at the end if I wanted to "have a conversation." Now, I could have taken the bait and argued since obviously I disagreed with his stance but that wasn't what I was there for.

It was a shame - nobody else (including me) got to share on the actual BOF topic.

My second subgroup was much better. I won't get into details but I left that group with a better understanding of how a devout Christian might approach aspects of CS education or education in general. I would have loved to explore more - were these teachers in religious schools or secular schools for example.

In any event, I really think that while there can be some value to this BOF, they really should have more inclusive language in the description.

So, that was SIGCSE 2026 for me.

It's now Tuesday evening. One week after my arrival in St. Louis and three days after the conference ended.

Terrific overall conference but at this point, I'm really hoping to get home soon,

Why CS Still Matters

Tue, 24 Feb 2026 07:36:57 -0400

Still stuck in St. Louis and still haven't finished my SIGCSE series but there have been a few threads in the various CS Education groups about the big changes in code.org and how AI might be making learning CS the way we've been teaching it obsolete.

Rubbish and here's a couple of stories as to why.

Let's go back to the start of this century. 9/11 had just placed America into a state of shock. I was home and I heard from a former student - one who graduated in '96. I'll call him Steve since I can't check to see if he's okay with me revealing his identity until later.

Steve knew that Stuy was right near the Twin Towers so he wanted to check to see if I was okay after the traumatic events. I also asked how he was doing. It turned out he was in Texas. Since graduating college he had joined the Blackstone group. He was working on the Enron thing. If you don't remember or are too young to know, check out the link. This was a big, complicated mess. Enron owned a huge number of companies and the debt and ownership relationships were nuts.

My guy Steve was the only junior guy on the team. The rest? Experienced business people with MBAs and tons of experience and similarly experienced people with doctorates in Economics or similar. I forget what Steve's major was in college but he both studied CS with me at Stuyvesant and also knew a bunch more from self study.

Now, forgive me but my memory is a little fuzzy. I forget if what he told me was what he already did or if it unfolded over the next few weeks and then I got the update but the gist is the same. I'll go with the assumption that the events I'm going to share here had already happened.

I asked him how they were going to deal with the situation. Steve said that after a considerable amount of time, all the senior team members were stumped on any quick resolution an figured they'd be down in Texas for months trying to figure things out and even then, they weren't sure.

My guy had different plans. He went back to his hotel room, took out his laptop, wrote up some code and a short time later, well, maybe not as short as now since computers and specifically laptops were slower back then, problem solved and Steve was written up in some Wall Street publication as the next "Wall Street Wiz Kid!"

The thing is, Steve looked at the problem through the lens of a computer scientist.

Now, you might say - but now with AI, anyone could just ask for the solution. Nonsense - back then, all those senior people could have come up with a solution and then grabbed a programmer to implement it

The key was that Steve could think and problem solve from a computer scientists point of view and that gave him the edge. All the AI in the world wouldn't have helped that Blackstone team had Steve - the guy who could think and problem solve like a computer scientist was not on the team.

Fast forward another ten or so years. I was talking to another former student. He was a lawyer. Attended Harvard Law and was now at a powerful law firm.

Randomly during out chat, he said, "You know, the most important class I ever took to prepare me was your intro CS class." I was a little surprised. That was the 10th grade class that used Racket and NetLogo as implementation languages but I always thought the class was more about thinking and problem solving than about just programming.

He said that both now professionally and during law school, he had an edge. Everyone in his classes and his firm could think like a lawyer but he could also approach problems like a computer scientist.

That's the win.

Over the years I've heard this again and again from my students who haven't gone into tech. I've also heard from the ones in tech how the value of that and other classes I've taught were not just about the language and the vocation of programming but of thinking and problem solving.

Of course I didn't reach everyone in this way - I've also had students who just feel that "you taught me Java" but that's okay.

The importance of teaching CS if we teach it right is the thinking and the problem solving. Programming can continue to be a tool to develop these skills in young minds and AI can and will be incorporated in the process. The kids who's teachers and schools get the mix right will be at a big advantage to those that don't.

So, don't let the naysayers diminish the value of our field. We've got to advocate for why CS is just as important as it ever was if not more so.

Code.org will do whatever Code.org does. So will the College Board and CSTA. They might make great choices for our field or they might make horrible ones. As teachers, it's our job to continue to advocate for what's right for our kids.

Sigcse 2026 part 2 - keynotes

Sat, 21 Feb 2026 13:31:09 -0400

I was going to wait to write this up until I got home. The conference ended on Saturday and it's now Monday afternoon and we're still in St. Louis due to the east coast blizzard. Right now we're booked on a flight home for Wednesday morning but I think there's more snow expected then so we'll see.

As to the keynotes, well two keynotes and one panel, they were all solid and thought or maybe better, discussion provoking.

Opening Keynote - Love, Learning, and Computing Education

Opening was Amy Ko who gave a talk titled "Love, Learning, and Computing Education."

I very much enjoyed the talk and judging by the fact that it was referenced in the second day panel as well as by the lead of one of the BOFs I attended, the talk at least got people thinking and talking and that's the point.

Much of the talk focused on Ko's journey citing a few specific examples of teachers who were able to take the time to nurture Ko's love of learning through computing. These stories also told of the teachers creating a safe, accepting, and supporting space for Ko so we've got love going on here in multiple ways.

I say "were able to" since many times the system works against teachers giving the type of support Ko received. In one example, a teacher worked with Ko outside of the class and the implication is that a good deal of time was volunteered. This sometimes just isn't possible.

For most of my career, I had 34 kids per class, 5 classes per day, and frequently 2 or more preps. All this and almost no time to do prep work and student assessment. Much easier to make connections in a private school with say 15 kids in a class.

This is not to say that it can't be done. I was indeed able to connect with many of my students as evident by the fact that so many have chosen to remain in touch with me post graduation. I also tried to create a safe and accepting environment wherever I've taught and again, based on former student feedback, I think overall I've done a good job,

I also am pretty sure that there were some kids I was never able to connect with and some where I couldn't give them the support that they needed.

Ko also touched on the purpose of computing education - is the goal to help our students flourish in general or to prepare them for careers at companies like Google?

This is an interesting one and it's probably something more relevant to college educators.

Do we teach the subject or the student?

Well, in the early grades, it's all about the student. If you ask an elementary school teacher what they teach they'll say something like "first grade." Over the years, it shifts. If you ask the same question in high school or beyond, you might here "math" or something similar - the subjects, not the student.

Of course, that's an over simplification but it was an important reminder that it's really the student - the person that's important.

Even then, though, it's complicated.

When I retired I got a note from a former student, they thanked me but at the end of the note, they said that the first thing they were able to do after getting that big tech job that they felt I helped them get, they were finally able to move their family into a decent apartment for the first time.

Brought me to tears.

The truth is, while we should be working to be a "force for good" in the development of our students, the practical can be important as well.

Great keynote. Glad it got people thinking and talking.

Friday Panel

Friday was a panel discussion of the SIGCSE award winners. Richard E. Ladner and Paul Tymann were on stage and the moderator also had some answers from Phillip Guo who wasn't present and who's paper on Python Tutor won the "test of time" award

To be honest, I enjoyed the panel but there wasn't a lot that I found thought provoking.

The moderator asked how CS education has evolved and both live panelists talked about the availability and ubiquity of computers. Phillip Guo's answer, provided in writing ahead of time focussed on how little the way we teach CS has changed which has both good points and bad ones.

Paul Tymann commented on how we need to increase the number of CS PhDs and graduates and I'm not sure I agree here. CS has been the hot field for a while now and enrollments have been through the roof. Maybe a little contraction would be good. That's not to say that we shouldn't continue working to broaden participation but maybe we don't need the overall number to keep climbing - we need it just to be more representative of our populations.

The one answer that I really liked was when the panelists were asked what they considered their greatest successes.

I made a note "for me - my students success" and waited to hear what the panelists would say.

Paul Tymann went first - he said something like this "there are all the papers I've published, grants I've won, … but my real success has been my studnets."

Yes!!! He got it right!!

Richard Ladner agreed.

It was a great way to wrap up the main part of the panel.

During the audience question part, Tymann also had a great response noting that we don't teach students to think about the consequences of what they're doing and that we must look at the consequences of the curriculum.

Closing Saturday - CS and Software Engineering Education post AI

The closing keynote was given by Titus Winters.

The thrust of this talk was that AI has changed the playing field for teaching CS.

One point Winters made early was that the theory and practice of computing has diverged. Also that we teach too much on the theoretical side.

Winters also went to say that we should move more towards planning and validation. Things like Software Architecture and Architecting, testing, planning, verification - all things that require problem solving and thinking, all things needed by tech professionals, and all things that aren't usually taught in a typical CS program.

He also spoke on how programming is a craft but we grade it my matching specifications.

Winters questioned why we teach Binary Search trees, how we teach Hash Tables, and ether we really have to spend so much time on leetcode problems.

I mostly agree.

Having an algorithms class where everything is done via proof or students just implement tough algorithms and data structures doesn't really prepare them for their futures, particularly as most CS majors are looking to become tech professionals and not academics.

How much better would it be if we taught our students to have an understanding of runtime, how different real world situations can change performance, how we can improve performance with smart choices, and when we might not want to due to other factors.

I used to teach a lesson where students started by writing code to find the mode of a data set. It was always an n^2 solution. The lesson would then lead them to a linear solution. We never spoke about big Oh but the kids got a taste of run time, refining data structures and algorithms, and when it matters and when it doesn't

On the other hand, winters said that CS0 should be "vibe coding 101" and I'm not sure I agree with that.

Overall this was another strong talk. I think Winters is partially correct but not entirely but it doesn't really matter. His talk is hopefully a launch pad for some hard discussions that as a field we really need to have.

So those were the keynotes. Amy Ko focusing on the human side and Titus Winters on the technical with a little of both in between.

If SIGCSE decides to make these available online, I'd recommend you watch all of them.

Next up - sessions.

Sigcse 2026 1

Fri, 20 Feb 2026 09:01:11 -0400

Time for another SIGCSE - the big Computer Science Education conference. This time in St. Louis.

St. Louis Arch

This is something like my 10th SIGCSE and the fourth I've attended with my daughter who's in her first year as a grad student, pursuing her PhD in CS with a focus on education.

At this point, I think she knows more people here than I do.

Actually, that was something I was thinking about. While I have my share of friends here at the conference, I know far fewer people. Back when I started, I knew most of the teams from Google, GitHub, TEALS, Microsoft, all of which had large representation. Also, there were more people from that CS4All efforts that I interacted with.

Between all of those people not being here anymore and retirements, I feel somewhat less connected.

I guess that's just the circle of life - I'm on the retirement phase while youngsters like my daughter are on the upswing.

A goal of mine, however is to be a co-author on a paper here with her one of these days and I'm still enjoying the conference very much.

As usual, I'll write about the conference sessions, keynotes, and BOFs but since the conference is still going on as I write, I'll focus on the conference and St. Louis today.

First off, St. Louis doesn't seem to do healthy food :-). The food's been great at the conference but not too many healthy choices. That's not a complaint mind you - I'm really enjoying the food and it's just a few days - like on a vacation. Just an observation.

Now, on the food - the logistics have been impeccable. Best SIGCSE ever. At every break and at the reception they've had sufficient stations and they keep them filled. No lines at all. I've been really impressed. They also have had coffee, tea, and hot chocolate (with marshmallows and sprinkles if you like) available before the start of each day.

Well done SIGCSE.

We did a couple of touristy things before the conference started - we went to the zoo and up the Arch (the picture is from my foggy run early in the morning) - both cool but the highlight was visiting the Cathedral. It was really magnificient. Right up there with all the other great houses of worship, even the Sagrada Familia which we just visited a few weeks ago.

St. Louis Arch

We also spent some time with a friend of Devorah's who warned us about safety. Don't go north or south too many blocks from the hotel and on and on. We figured, whatever - we're New Yorkers - we'll do our usual - be aware of your surroundings, walk confident, etc.

At the conference though, for the first time in my memory, they showed a slide about safety - don't wear your badge outside, walk in groups, take ride shares.

I haven't felt unsafe, but maybe the part of St. Louis we're in isn't so safe.

I wanted to also hit the City Museum and the Blues museum but the City Museum was closed on my one tourist day and the Blues museum closed to early.

I'll have more to say about the conference overall but since we're mid day two I'll hold off on that until the end.

Next couple of posts though will focus on the sessions, keynotes, and BOFS.

Unerasable - an effort to prevent the erasure of history

Fri, 13 Feb 2026 16:36:09 -0400

History is a funny thing.

They say it's written by the victors but even when the "good guys" win, the history taught is often, let's say, incomplete.

Back in school we'd learn about all sorts of feel good American equality stuff but how many learned about Black Wall Street and the Tusla Race Massacre?

I've knew about the Green Book and Sundown Towns years before the "Green Book" movie and was shocked when I did learn about the towns, how recently they've existed and how many existed up up north.

I didn't know, though until recently that the Green Book was modelled after The Jewish Vacation Guide - a similar publication but to combat antisemitism and published a decade or so earlier.

It's bad enough that we've been taught a false and rosy history of our country but it's getting worse.

Under the current administration, history, particulary the history of marginalized groups is being erased. Most recently, the LGBTQ+ flag from Stonewall but the erasure is widespread and rather frightening.

A friend of mine is part of a group leveraging tech to try to save history from erasure. In this case, specifically Black history.

The project is Unerasable.

The project is leveraging various tech and technologies and they're looking for help.

I'm sharing what my friend sent me below and if you're interested, drop me a line and I can connect you (or you can sign up on the site).

I wanted to share a project that started from a very personal place.

After reading about the removal of slavery exhibits from national parks and other public spaces, I was heartbroken — and angry. As a father, I realized that on our trip to Philadelphia next week, I couldn't be sure what of our history would still be there to experience. So I started Unerasable with some community builders I work with.

Unerasable is a decentralized public archive that uses 3D scanning, blockchain, and immersive tech to document and preserve historical spaces and exhibits before they’re removed or altered. We just launched our V1 site here and had our first team meeting this month — and now I’m reaching out to you all because we need technical collaborators.

Here’s how you can help:

3D + Photogrammetry – We need folks who can advise on or contribute to high-quality, accessible capture (from phone LiDAR to pro gear).

Blockchain – We're looking for tech sponsors or devs to help us log provenance data (timestamped capture info, version history) in a decentralized, durable way.

XR / Immersive – We're designing a lightweight web experience where users can return to erased spaces. If you’ve worked in AR/VR or WebXR, we’d love your input or help building that layer.

This project is open-source in spirit, urgent in mission, and deeply personal. If you’re interested in helping, collaborating, or connecting us to people in your networks — I’d love to talk.

Even with all the bad going on right now, it's heartening that there are good people trying to leverage tech for good.

Teachers Observing Teachers

Tue, 10 Feb 2026 08:38:37 -0400

Finally we get to teachers observing teachers - what Mark was writing about to begin with.

These observations could be a one off, for a few days, or even through a full course.

This could also be a team teaching situation but to be honest, I think that's a different beast. If I'm team teaching a course with someone, true, I can learn from them if and when we do joint preperation and things like grading and yes, I can learn something by being in the room when they're leading some activity but it's not the same since at that time, my job isn't to observe, my job would be to do something else to enhance the lesson - direct student support perhaps. A school won't be paying me to just observe another teacher except under special circumstances. If they're paying for two bodies in the room, they'll expect two bodies worth of work, whatever that means.

So, let's stick with straigt observations.

First, there are scheduling difficulties. A typical traditional High school teacher teaches 5 out of 8 periods a day. The school day might extend further but then the teacher would be coming in before hours or staying after.

Of the three non teaching periods, one is their prep - when they make lesson plans, grade assignments etc.. One is lunch which contractually is "duty free," and even though one can observe during lunch, another teacher might not want a teacher eating in their room given the rodent and bug problem prevelant in many schools. The third period is the teachers "professional period." In some schools this is treated like a semi-free period, like the prep, but in many school's you get your professional assignment - maybe you have lunch or yard duty, maybe you're doing small group tutoring. If you're lucky, maybe your principal will allow you to use that period to observe.

So, finding time to observe can be tough. If you have the free period, you're still limited - is the teacher you want to observe teaching the class you want to sit in on during your limited free time? What's more, do you also have time to talk either before the lesson or debrief after?

Not easy.

The one thing that works, scheduling wise, is that generally when setting up a school's program, you try to spread all the sections of a class across the day. That is, you won't have multiples of one class during the same period until pretty much every other period has at least one section going.

In college, it theoretically should be easier - lecturers and professors schedules are generally more flexible but I'm sure there's still a laundry list of challenes that make observing hard.

When I was scheduling CS at Stuy, I did try where I could to allow for teacher observations, particularly if a teacher was new to a subject. I tried to make sure that the new teacher could sit in on an experienced teacher's section and if possible, do it in a way that didn't give them something like 4 in a row or have them running all over the building. I wasn't always able, but I tried.

Personally, I only managed large scale observations three times in my career. When I first started at Stuy, I was teaching a section of Linear Algebra. Having never taught it before, I sat in on a colleague's class earlier in the day. He also shared his old lesson plans with me but we never conferenced about the class. That was more about material and pacing though than about instruction.

I also audited a History of New York class but that was because I wanted to take it and never had the chance back when I was a student.

Both of these cases though were relatively early in my career when the professional period was mostly a second prep which made things easier.

I also observed quite a bit in my first year teaching but back then the DOE had a new teacher mentoring program where the teacher, me, would teach one period less as did my mentor. This gave me more time to both conference and observe which, combined with the supportive math department and faculty in general at Seward Park High School was what allowed me to survive as a first year teacher.

I also did the video observations and peer observations back when the my school at the time (Seward) allowed them.

Since then, it was mostly one offs (not counting when I officially observed as the CS program coordinator).

Now, in terms of the actual observations - for those to be valuable the observer must be intentional. You can sit in on a class and just be another student trying to observe content and that's fine if that's what you're looking for but to develop as a pedagogue you've got to look for more.

Of course, it's easy if you and the teacher you're observing specify what you're looking at ahead of time. Early on, I observed a series of lessons a friend taught because he was really good at setting up group experiences and using paper cut manipulatives - this was in a math class. I wasn't particularly good at that (or anything back then) and he invited me in.

Other times, you really have to pay attention. Maybe things are obvious like an ongoing procedure a teacher uses. Sometimes it's subtle like how a teacher might move around a classroom. If they move towards a student who's starting to answer a question and then away. Was that random or for a purpose. Which way they face while writing on the board (yes that can be intentional). If they wrote out a definition on the board and had the students copy it - was that because they wanted the students to engage in a particular way - by transcribing a definition or did they just not have time to make up a slide with it before class?

What about jokes? Are they spontaneous or scripted and if spontenious what will the teacher do when they teach the same class again later in the day.

Observing a teacher can be incredibly valuable but you'll only get maximum value if you pick up on the subtleties. Maybe a teacher does something great and they don't even realize it!!! This is why talking shop around the observation is also hugely valuable.

I'd like to think that at StuyCS we had a culture where teachers were comfortable popping in on each others classes and learning from each other but even so, the timing and other burdens make teachers observing each other a far rarer occurence than it should be.

Rain in Spain

Fri, 06 Feb 2026 08:52:12 -0400

Cervantes monument

Taking time out from my observation blog posts to write up a trip report.

For the past few winters, Devorah and I have been trying to spend a few days somewhere warmer. Not necessarily warm, but warmer.

We've done San Diego twice then Charleston and last year Savannah. We were thinking of New Orleans this time around but we're going there for the CSTA conference in July so started looking around. It turned out that air fairs to Spain, specifically to Madrid or Barcelona were ridiculously inexpensive and hotels were also pretty affordable. Going to Europe was actually going to be a good site cheaper than travelling domestically so we booked the flights.

Five days in Barcelona and five in Madrid with a day trip to Toledo.

Got tons of great recs from former students and friends and this post will be the highlights. Partly to share and partly so that I have a memory.

Besides the price, Spain checked the boxes. Barcelona was supposed to be low 40s going into the upper 50s most days and Madrid around 5 degrees cooler. According to my research, each usually gets 4 days of rain in January.

Unfortunately, we got all of them :-(.

It threatened rain every day we were there and rained most of them. Fortunately, the rain really didn't hinder our trip and we still had a great time.

Barcelona

We stayed at the Oriente Atiram hotel on Las Ramblas which is in a historic building and has a beautiful atrium.

After checking in we spent most of the first afternoon wandering around in the rain. Over the next few days we toured La Sagrada Familia and Park Guell - two of the architect Antonio Gaudi's most famous works and walked by a number of his others. Both were as magnificent as advertised and well worth the visit.

Park Guell

Sagrada Familia

Two of the architectural highlights though weren't by Gaudi. One was Sant Pau Recine Modernista - a complex that was formerly a hospital. Amazing site and really interesting both from the architecture and the history of the hospital and hospitals in general. The other was the Palau de la Musica Catalana - an amazing concert hall. We visited but unfortunately, nothing was playing that interested us during our time in Barcelona. Both of these sites were designed by Lluís Domènech i Montaner.

Concert hall

Sant Pau Recine Modernista

We also visited the Picasso museum, Miro foundation, a couple of smaller museums and even found a small museum in what was an old Jewish Synagogue.

Food was also dynomite. We hit one place for paella but mostly did Tapas. I was a little concerned with dining since they eat much later than we do. I thought we'd either have to totally shift our schedule or we'd just get bad tourist food. It actually worked out fine. We for the most part ate a little later but still earlier than most locals but as long as we were off the main tourist drags and squares we seemed to do fine.

In fact, one particular place - probably our best meal in Barcelona, our early dining worked for us. It was Bodega Biarritsz where they had a fixed tapas meal - 45 euros for 2 people or 55 for two with an extra course of local cured meats and cheese. I looked to see if they took resrvations and their site said no and if you didn't want to wait, come off hours, specifically between 2:00pm and 6:30pm - score!!!!!

Toledo

We did a day trip to Toledo from Madrid. There was enough in Madrid so we could have spent plenty more days without venturing out but Toledo had some Jewish stuff and we always like exploring our heritage when travelling even if we're technically of Ashkenazi decent and Spanish Jews would be Sophardic.

Synagogue

We visited one Synagogue/museum from the 1400s and another that was first a synagogue and then a Catholic church that dates back to 1205. That older synagogue interestingly enough was also designed by Muslims.

We also hit few other interesting Jewish historic sites and some Catholic ones as well not to mention a quick stop in the El Greco museum.

I found it interesting that on the one hand, prior to the trip I saw a fair amount of materials talking about Toledo as a center for the three major Abrahamic religions with an implied veneer of equality. It was nice to see that once there, the history wasn't whitewashed with references to Jews living under Islamic law as Dhimmis - second class citizens and the expulsion of the Jews from Spain in 1492 by Spain's Catholic royalty.

Our Toledo visit was our rainiest day but I 100% recommend. If we get back to the region, we'll probably return to Toledo for another day since there are still a few things I want to see.

Madrid

Madid was also pretty terrific. We skipped the Prado - Madrid's largest and most famous art museum and decided to hit the other two of Madrid's big three - the Reina Sofía and the Museo Thyssen-Bornemisza.

We did something similar when we went to Paris for a couple of days on a trip last year - skipped the Louvre but hit the Musee d'Orsay and a couple of others.

The collections and exhibits at the Riena Sofia and the Thyssen-Bornemisra were more in line in what we're currently enjoying.

I think we made the right decision. Devorah really wanted to see Picasso's Guernica so we hit that first at the Riena Sofia.

Guernica

It wasn't too crowded so we could get both up close and distance views to really take it in. I also loved that they had information on it's development and images of themes that never made it into the composition.

The physical viewing was a stark contrast to when we were at the Rijksmuseum in Amsterdam last year and saw "The Night Watch." It wa covered in glass and you couldn't really get up to it (though a year before, when I was there with Batya you could).

Overall the museum was great but I think I liked the Thyssen-Bornemisra even better. The main exhibit was set up chronologically starting with rooms full of fourteenth century mostly religous works. As you go through the museum you move through time, later visiting a room full of Hudson River School paintings, also impressionists, and up to more modern works.

We also visited the Royal Palace which was impressive and various other sites around town.

Of course, lots of great food.

Churros

tapas

Just as we found one particularly terrific restaurant in Barcelona we also did the same in Madrid. We originally set out for a place that looked good on Google maps but when we got to the address there was no restaurant to be found.

We ended up a few blocks down at what maps to this place - En Busca del Tiempo though the menu online doesn't match at all what we saw and what we had.

We were given Spanish only menus but muddled through. Ordered three dishes - one tapa - a mushroom dish, what they had listed as a "snack" (based on Google translate) - Prawns braised in cider on rice and a meat dish. Way too much food but each dish was sublime.

We had to go back the following night because we wanted to try their paella. Our waited recognized us and took good care of us. His English was far better than our Spanish so we managed to have a nice chat and at the end we were comped an extra. Definately going back next time we're in town.

So that was ou January in Spain trip.

Great time in spite of the rain.

Thanks to all my friends who helped guide us in making our intenerary.

Next time, we'll dive back in to observations - teachers observing each other.

Being Observed

Tue, 03 Feb 2026 04:03:43 +0000

Following up on my last post, let's dive in to being oberved as a model for teacher improvement and teacher evaluation. Next time we'll look at the flip side - teachers observing for improvement.

As I mentioned last time, I'm writing about New York City public schools but would be interested in hearing what things are like in other locations and other types of institutions.

Evaluation and improvement - that's the line but it's rarely the case. It's almost never really about teacher improvement and for evaluation, well, boxes are checked and contracts fulfilled but I'm not convinced that even in evaluation, they do more than that.

I recently saw someone post a comment about observaitons noting that "as long as your administration is reasonable or better, they're not really a big deal."

I think this is accurate and we'll see why soon but there are cases where observations are weaponized.

My first chairman absolutely loved me. I was a solid young teacher but he treated me like I was the greatest thing since Betty White. He even wrote up in one of my evaluations "this lesson was a perfect as a lesosn could be" which was patently wrong - it was a solid lesson and I daresay a really good "delivered by a beginner" lesson but it was nowhere near perfect.

The problem with that chair though was that there were other teachers he didn't like at all, including a couple of my mentors and friends. I remember one morning in the Math office when he commented to me on the way out, "I'm going to get Bob today." Bob being a stand in name for the real teacher's name. The implication was that he was going to go in to observe him with the plan of writing up a bad report.

This really scarred me. I taught for over thirty years and was observed countless time. I never received a bad writeup in all that time, although once, in those early years, that chair came into my classroom, saw it was "one of those days" and noped right out since he liked me, but for all my time as a teacher, I always tightened up when my supervisor came in.

Now, I say I never received a bad writeup but not all my observed lessons were winners. Early on when at Stuy, I was observed in a couple of "meh" lessons but my chairman there, Richard Rothenberg was much gentler. He honestly tried to make it about teacher improvement and just really liked watching his teachers teach. One lesson, I was just severely underestimating my classes potential and Richie gently nudged me along while really working hard to not make me feel bad about it. Another time, he felt that my lesson could be enriched with outside materials and that led to an ongoing conversation about math enrichment and context materials and I became a better teacher for it.

Still, I would always freeze up.

Getting back to why observations don't work for either evaluation or improvement, let's start with the burden on the observer.

In a small high school of about 100 kids per grade, give or take, you'll have around 17 teachers - probably more like 20. When I was teaching high school, everyone was observed 6 times a semester though if you were tenured, 4 of the 6 would be partial period rather than full.

That's over 100 observations.

That same small school has one principal and one assistant principal, maybe a second AP but they probably also have more than 20 teachers anyway so let's say 1 principal and 1 AP. There are typically 180 school days each year so that's over 1 observation a day.

Oh, did I mention that every observation has a post ob conference and the full period observations are supposed to have a pre observation conference though that seems to rarely happen.

Then, when it's all done, the supervisor has to write it all up.

Now, don't forget that the principal and AP also have other responsibilities like, oh, I don't know, running the school in general and dealing with every emergency that comes up.

On top of that, in a small school, the likelihood is that the supervisor isn't going to know much about the subject they're observing.

Is it any surprise that this model fails in terms of teacher improvement and is, in schools with "reasonable or better" administrations pretty much box checking lip service.

The other major problem with the observation model is that it's a snapshot without context. I've heard that some teachers, in schools with bad admins have an "observation" lesson ready to go and when the supervisor came in, that lesson would come out.

I actually sort of did this but not really.

NYC DOE at one point started using the Danielson Framework for observations. As a teacher evaluation tool, it's garbage. Even Danielson has repeatedly said this (and she kept saying it all the way to the bank). The Framework has dozens of areas of expertise that teachers are supposed to exhibit. Far too many to keep track of so our school decided on a subset that supervisors would look for as check boxes for observations.

Well, to start, think about how ridiculous that is - what if I teach a great lesson that emphasizes the other areas of Danielson's framework? If that happened, my lesson would be unsatisfactory. I was actually talking to a colleague once and she said she hated the framework. I asked why since what they were looking for fit this teacher like a glove. She replied - it's only a matter of time until they change it and then I'll all of a sudden be a bad teacher.

In any event, I had a great chairman and we'd have an unspoken game we'd play when I was observed. We both knew my lesson would be fine so my chair would look for one or two Danielson points to "ding" me on and I'd try to cover them all while hiding one or two so they weren't obvious.

It was all in good fun.

One year, one of the categories was redirecting a student question to another student. During the post ob, my chair pointed out that I didn't cover that. Now I forget what I did, but I did something convoluted to hide it but specifically put in an instance of redirecting a student question to another question. I pointed it out. I won that round. I list others but again, this was a game we played.

The truth is, we also spoke about the lesson in general and that was valuable. We did the Danielson nonsense because the DOE required it but then we just talked shop.

And that actually brings me back to the beginning. The entire model is horribly flawed so is mostly just a check off in a decent school and unfortunately, a weapon in a bad one but for all the money spent and wasted on Danielson and refining how teachers are observed, my first chairman gave me guidance that I always tried to follow.

He understood that all sorts of lessons can be good and can be bad. One of my colleagues was big on cooperative learning so that's how he framed most lessons. That didn't work so well for me so while I might incorporate some "cooperative learning" concepts, by lessons would look very different. He said that when observing you just have to ask "is learning happening." And then as the observer you have to figure out where and why. If you can't you discuss that with the teacher.

Simple but it's always served me well.

It was reinforced very late in my career. I was talking to my current principal and they were talking about a certain legendary teacher. They commented "I can never bring any DOE officials into his class because we all know he's one of the best teachers in the school but he doesn't do things the way they expect them to be done."

I'll finish off with a couple of observation alternatives that existed when I started teaching but seemed to fall to the wayside. They may still exist in other schools but I just don't know.

One was the video observation. For one of my three each semester, I was allowed to video tape a lesson (yes, it was tape back then), watch it, then watch it again with my supervisor for the official observation.

That took more time but it was actually pretty valuable if you could get over the cringe inducing moments when rewatching the lesson for the official part.

The other was that we could do a peer observation for one of each three. I'd first conference with a colleague, we'd then observe each other, conference again with each other and then finally all meet with the supervisor for an official debriefing.

This was also really good but took much more time and coordination was tough.

So that's being observed in NYC public schools. Like I said at the top, I'd love to hear what's done elsewhere and if it's any better for either improvement or evaluation.

Next up, we'll talk about teachers observing other teachers for improvement - the thing that Mark's initial post was really about that got me started on this rabbit hole.

Improving Instruction - College and K12

Wed, 21 Jan 2026 04:03:43 +0000

Back in for the evening. Not in Barcelona anymore - spending a few days now in Madrid. Sadly, I got a message from a fellow CS Educator who lives in Barcelona the day we left so if any reader out there is in Madrid - I'm here a couple more days - let me know.

For the post at hand, I read this post the other day by Mark Guzdial and it got me thinking about a number of things. The difference in educator development between K12 and college, how teachers are and aren't improved, and specifically, observations as that was a core part of Mark's post.

Since I'm just in from a long day of walking and touristing and there's a HUGE amount to cover, I think I'll just lay out how observations, in my experience, are and arn't used and then dig deeper in future posts, probably when I'm back in New York.

As a public school teacher in New York, the primary method of being evaluated is observation - you're observed by your supervisor. In my case, it was usually my department chair but in a smaller school it might be the principal or any assistant principal.

Traditionally, assistant principals and principals taught for many years before moving into administration but under the Bloomberg Administration, they fast tracked non educators into the positions so you might be observed by someone who just had a couple of years in the classroom. Hardly ideal.

Note that all of this is the principal observing the teacher whereas Mark was talking about an instructor observing another instructor to see what they could glean. I'll talk about that and the relationships between the two in a future post.

Also, I said this was the method by which teachers were evaluated - not improved. The observation plan was supposed to be about teacher improvement but it rarely is. Again, more on this later.

When I started as an untenured, unappointed teacher, I was observed 3 times a semester - 6 a year. Once appointed, which meant that I was officially assigned to a school and was no longer considered term to term, it was once a semester. Finally, once tenured it was once a year.

That changed over time and in my later years it was back to three times a semester or six a year.

As a CS teacher, I was never observed by someone who knew my subject area but still some observations were helpful. Most though were just checking the boxes and not necessarily for bad reasons but again, that's for a later post.

Now note that none of this involves a teacher observing another teacher though that could be done in a limited capacity early in my career. Except for that, it was all top down.

Let's contrast that to college. I can only speak to how things worked at Hunter but I think at least all of CUNY's the same. Once each semester I was observed by a fellow full time faculty member. Each department had it's own way of doing this. Curriculum and Teaching, where I did the observing, had a lengthy form that, to be honest, was pretty worthless. CS, where I was observed, had a much shorter form with checkoffs as well as room for written responses but it was similarly worthless.

Basically, a full time faculty member, who wasn't trained as a teacher, would see me teach one lesson and then write it up. Most saw the lesson, we talked about it then wrote it up. One teacher didn't like some aspects of what I was doing, I found it annoying because the observer was a professor with a reputaton as a bad teacher but it didn't matter because it wasn't going to affect me overall anyway. In any event, it wasn't really about improving instruction, it was really about a cursory check on instructurs.

Neither in K12 nor in college was there any official way of observing other instructors.

So that's the framework on how observations were used over my career.

In future posts I'll dig into:

The variations on observations in K12
Teachers observing or sitting in on other teachers
Why it's tricky to do
Why it's tricky to get something out of. But I think this is enough for tonight.

Funding for CS Educational tools

Tue, 20 Jan 2026 04:03:43 +0000

It's 10:00am and I'm sitting in my hotel room in Barcelona.

For the past few years, Devorah and I have taken a trip to someplace warmer (but not necessarily warm) in late January or early February. Something we could never do before retiring and this year, we decided in Barcelona and Madrid because airfare and hotel prices were just so low - cheaper than the things we were first considering doing in our own country.

Normally we'd be out and about but we're going on a walking tour at 11:00 and it's a rainy day so we're just staying in until then.

I also realize that I haven't blogged in a couple of months. It's not that I haven't been thinking about writing - I've been programming a bit more, playing with AI and other tools and a number of times, I've thought that maybe I should blog about it but indeed the frequency of CS Ed blogging is really down and I guess blogging in general so that takes away some motivation.

This morning though, I read Alfred Thompsons piece on funding CS Education tools so with time to kill figured I'd put in some thoughts. Alfred talks about University funded and public funded efforts like Alice and also private tools with a free tier but he left out commercial endeavors specifically created for the eduation market.

Remember Codecademy? They were one of the darlings during the Internet 2.0 boom and everyone can (and should) code movement and was one of the companies Mayor Michael Bloomberg highlighted as he pushed Tech in NY during his administration.

They were interesting because they marketed to individuals not to schools and not to educators at schools. They had a free tier and pushed individual learners but ultimately, I think they pivoted to corporate training. An interesting criticism that their users frequently brought up was that since you learned in a sandbox even after you finished their courses, you couldn't do anything real. You could theoretically write something in Python let's say but had no way of running it outside of the Codecademy environment.

Another was Codesters.com which was a new product path for Tutor Associates. Codesters was and is an online Python IDE learning environment complete with lessons. They had an innovative combination of drag and drop and text based programming that I liked and I saw reproduced recently by another project that I saw demoed at a recent CSTA conference. In contrast to Codecademy, Codesters marketed to schools and teachers with the intent on teachers using it with their classes.

Both of these contrast with something like AppInventor which I don't believe started as an educational endeavor. I could be wrong but I seem to recall it was a fast prototyping tool and then shifted to the ed space. They also contrast AppInventor and MakeCode in that Codesters came out of a small business and Codecademy was a startup while Google and Microsoft - both tech giants.

Going with the privately developed CS Education products I want to highlight two more. Pickode and CodeHS. Both developed specifically as tools to be used for learning programming. Full disclosure - I know the founders of both companies and consider all of them friends. Independent of that, I like both of their products.

I also want to contrast both of them with another startup project that was briefly the darling of the CSEd community - Repl.it.

Repl.it was and is an online development environment. They provided a substantial free tier and environments to program in almost any language. They also integrated with Github and had a variety of teacher support tools so managing classes was pretty easy.

Overnight it seemed there was mass adoption of Repl.it in the CS Ed world. They appeared to want to support education and they provided some great tools.

It didn't last though.

Abruptly, mid semester, Repl.it decided they were going to pivot their business and they shut down their educational support tools. I can't tell you how many teachers were left out in the lurch.

A perfect example of the danger of using commercial software. This wasn't even a free tier going away it was a product going away because the company no longer wants to support it.

In contrast to Repl.it which was a generic online IDE looking for a market, and found CS education until something better came along, Pickcode and CodeHS were specifically designed for education so were safer landing spots for teachers and their students. CodeHS, more established, bigger, and with more offferings and Pickcode, newer, smaller, and able to offer a more boutique type service where teachers who were customers could really help shape the product.

This classroom side edtech startup ecosystem makes another player to add to the ones Alfred talked about - university based research, and big tech industry projects.

Both CodeHS and Pickcode offer free tiers and there's no guarantee they'll be there forever but since online leaning IDE is their core business, one runs less of a risk of the company pivoting like what happened with Repl.it

Moving forward, I'm not sure what the edtech startup ecosystem will look like for CS in the near to mid future - I haven't given it much thought yet but one would think that even as CS education starts to deal with AI, there will still be enough of a need and market for new startups and or new products from existing players.

Okay, so that's it for my first blog post in a couple of months and first one in 2026. We'll see if I can get back into a semi regular writing groove.

Ray Kurzweil, AI, Intelligence, and Education

Tue, 18 Nov 2025 19:27:11 -0400

Over a month since my last post. Wow.

Anyway, last week was full of Artificial Intelligence events. On Wednesday, I attended a talk by Ray Kurzweil, hosted by Trinity High School - a posh private school on the Upper West Side, The Academy for Teachers, and Teach for America. Thursday was a StuyCS event which was really about early stage startups but much of the discussion revolved around AI and Friday was a meeting of CUNY CS educators where the topic of the day was, of course, AI - I was there to pitch, well, Pitchfest.

All three were great but I wanted to write about the Kurzweil talk.

The talk was supposed to be about AI and the future of education but he really didn't share any deep thoughts on the subject. He did, however bring up some topics that made me think about a number of issues.

A big part of the talk involved Kurzweil talking about what he described as our biological brain and our other, external brain - that being what we have access to via our phones or other computing devices. The example he gave multiple times was that we might be thinking of an actor and forgot who they were but we can easily use our phone to look them up which, in a way, increases our intelligence over us, say 30 years ago.

Kurzweil predicts that by some time next decade we'll have achieved Artificial General Intelligence and in fact our biological brain will be directly connected to our "external brain." Then, if we're asked about that actor we'll just be able to give the answer instantly. We won't even know if the answer came from the biological brain or the other one.

It will have made us smarter.

Or did it?

This was the big question I was left with.

I was catching up with an old high school buddy and was relaying this to him and he hit the nail on the head. He said "or at least we'll have more knowledge."

Does knowing more stuff actually make us smarter?

It's an interesting question.

Years ago, a few of us at Stuy were talking about two particular students. Both very smart by any reasonable measure and in fact both were individual standouts. There was a big difference between them though. One student knew a lot - had a lot of information. Had a great memory, read tons of books, could produce formula after formula on demand and so forth. The other didn't have nearly as much information at the ready but was stronger at figuring things out. The first student would just know the theorem, the second would derive it.

Who was "smarter?"

We mostly felt the one who was the stronger problem solver was "smarter" than the one with more knowledge.

So how does it play out with AI?

Will all that knowledge make us smarter?

We'll "know" all the formulas but will we know when to use them? Will knowing all the chords make me a better composer? Knowledge of color theory a better artist?

It also raises the question of how the information will exist and be "communicated" to our brain. Right now most of the information we interface with is textual - word based descriptions. Will textual knowledge translate to non textual concepts?

How does this play against things like Synesthesia where some people might "see" music as colors or "smell" scents for shapes.

I don't have any answers here but these are questions that go well beyond the already interesting and challenging questions of students using LLMs.

Of course, those LLM questions came up during the second part of the talk. That was something of a reverse panel where three guests - one, a representative from Math for America, one from the Academy for Teachers, and one form Trinity asked Kurzweil questions.

Another question that came up during that panel part made me chuckle. It was raised by the Trinity representative - I think she was head of school. She went on about he strong belief in equity and diversity and ended with concerns about AI - could it be an equalizer.

I chuckled because the person who was asking the question hold a position that in spite of statements to the contrary is very much about maintaining separation between the "elite" and everyone else. I mean, a high school that has an annual tuition of over $60,000 isn't really working for the common man. The person she was asking, Ray Kurzweil, on the other hand was a product of New York City public schools. Martin Van Buren high school specifically.

Chuckles aside, it's an important question and I didn't agree with the answer. Kurzweil addressed the audience noting that they all had phones and that having a phone was not a big deal financially. As I looked around, I noted that the audience was mostly white, mostly well dressed, and apparently largely affluent.

Sadly, the digital divide is still real and I don't see things getting better with AI.

In any event, I really enjoyed the evening and it did leave me thinking about things.

I know that we're all thinking about AI as educators. I wonder if others are also thinking about the types of issues I was left thinking about after Kurzweil's talk.

Gifted Education - Acceleration or Enrichment

Fri, 10 Oct 2025 08:28:53 -0400

Last week my StuyCS family community had a discussion about Zohran Mamdani's plans for gifted education. For people outside of New York, Mamdani is the Democratic candidate for mayor and there was a recent article outlining his plans or rather, his positions on gifted ed - always a hot button topic for Stuy alums. It also turns out to be a hot button topic in education in general in New York.

Long story short, he plans to phase out admission to gifted programs in kindergarten while leaving the 3rd grade starting point. He also plans to leave the SHSAT - the entrance exam used for Stuy, Science, Tech and a few other schools in place while bolstering programs that help raise awareness and help prep middle school students from underrepresented groups for the exam.

Overall, I approve of his plans but something struck me in the article it mentioned in passing that the gifted programs would use a model of acceleration, a path that I would question as I think enrichment is generally a better way to go.

Acceleration basically means you go through the material faster. When Abe Baumel was principal of Stuy - from Fall '84 through sometime in the early '90s, he only allowed honors sections be created if they were accelerated - his stance was that "all Stuy classes are honors classes." This meant that the math honors program had students finishing geometry and alg2/trig in one three semesters rather than four. This did allow those students to take a class in Linear Algebra and Number Theory.

After Baumel retired, the math department shifted the honors program to an enrichment model.

Looking at gifted or honors programs not only at Stuy but also in the earlier grades, it's easy to see some of the problems with the acceleration model but other issues are subtler.

To start with, once you're in an accelerated program, you're off track - topic for topic you're not at the same place as the rest of the school. This means students can't be placed in to the gifted program if they're identified later unless the gifted and regular programs sync up at some point later on.

It also means that a student who has to leave the gifted program for the regular program will either be placed in a class where they've already covered all the material or in a class that's more advanced and the student is missing something.

Far from ideal.

Another issue is that kids are sometimes accelerated too far too fast and while they can get by the standardized tests, they don't really grok the material. I saw this a number of times at Stuyvesant. A kid would come from a school that had an overall weak program but they'd have a small number of high achieving students. They'd push them through the regents classes at an accelerated rate. I'd then get them at Stuy in a more advanced math class and even though this kid got an A in their prior classes and a ninety-something on the regents, they just didn't really have a strong foundation in the material. They studied it too superficially and just learned how to pass the test.

I've also seen this phenomenon when kids take Physics too early.

Relating to this, but sticking with math education, a number of friends and colleagues over the years have had a chance to study education systems in other countries and a frequent comment is that some other countries cover far less in their math courses but what is covered is covered more deeply. Every single person I've spoken to who's observed this noted that the kids educated in that way seem to end up much stronger mathematically overall.

I've even seen this with my own kids. When they were really young, I tried to introduce them to Algebra. First it was too early and they just didn't get it. I quickly dropped the effort. Later, I tried again and they were able to do the algebra but they didn't really get it, they just seemed to be following the form. I dropped the effort again. Finally, I approached the topic a third time and this time, they were at the right developmental place and mastered the topic with no difficulty in no time.

So, acceleration is, let us say, problematic.

Enrichment is better. What's enrichment? Going more deeply into the subject you'd normally be studying anyway, looking at interesting applications of the subject, or covering tangential material that won't be covered in future required classes.

In a math class, maybe doing more derivations or proofs or as another example, when I taught linear algebra, I'd relate it to computer graphics applications. Nowadays, I'd probably look at something that relates to AI or ML but back then, graphics were all the rage.

In my CS honors classes at Hunter, I'd cover aspects of software engineering or interesting approaches that they would see neither in the regular CS sections of my class nor in future required CS classes.

Actually, many of the classes I designed at Stuy were like this - what are enrichment topics that these kids could use that they won't see in college. If I'm just teaching them stuff they'll get again in a couple of years when they'll probably be more ready for it, then what's the point.

With enrichment, kids can enter and leave gifted programs at any point in time. They also serve to reinforce the core topics they're supposed to be learning rather than glossing over materials in a race to get as far ahead as possible.

If Zohran Mamdani wins (and he currently has my vote), I hope his education people go further than just look at gifted programs at a superficial level and that they look at how they're implemented as well.

Dual Credit Or AP

Mon, 15 Sep 2025 14:50:11 -0400

I saw this release from NYSED the other day.

New York State wants to create a standard set of policies and procedures for schools implementing Dual Credit programs.

A Dual Credit class is a college class taken by high school students. The high school offering the class does so in partnership with some college or university. A student who passes the class receives both high school and college credit. They in fact, get a transcript from the partner college with the course or courses they've taken.

The partnerships and models were all handled independently and have taken on different forms. Over the years, I've seen these courses offered in the high schools taught by HS teachers and a the colleges, taught by college instructors. I've also seen it where the classes are entirely made up of HS students and also when they're integrated into classes with college students.

It's probably good that the state is standardizing dual credit but I'm not convinced that it will actually improve anything. It may in fact make things worse in some cases.

So, let's dive in by first comparing dual credit with the other big college credit for high school plan - the AP program offered by the college board. As I'm sure you know, for AP, a school offers an AP course after having a syllabus (and maybe instructor now) approved by the college board and they take a single high stakes exam near the end of the school year. Colleges then decide based on that single score if they want to award credit, placement or neither.

Let's look at AP first:

Pros of AP

It's huge and it exists.
There's no real cost to the school othan the single exam fee which students might pay or in some cases, districts or government funds pay.
It's national so for the most part, the AP Calculus curriculum anywhere in the country will, at the least common denominator level, be the same though some schools might have more rigorous programs.
Similarly, it means a wealth or resources are available for most AP classes.

Cons of AP

A single test on a single day is not a good assessment.
APCS Principles and possibly other classes are not college level (not necessarily a bad thing but we're talking college classes here).
Curriculum and many course decisions made by an outside organization - the College Board. They are neither a K12 school nor a college.
Test credit seems to be awarded less and less although blanket elective credit seems relatively freely given.
Instruction quality can be uneven across schools since the assessment is a single test on a single day.

It's worth noting that the College Board has been branching out to more non college level courses recently for whatever that's worth but since, as I just mentioned above, we're talking specifically about college courses and college credit, that's neither here nor there.

Pros of dual credit

Assessments can include anything and everything and the student should receive a grade based on the entirety of their performance and not on a single high stakes exam.
Transcript credit is harder for a college to ignore than an AP score.

Cons of dual credit

Class quality is uneven. Some dual credit courses are rigorous college courses but, as was noted in a news expose a few years ago, there are dual credit schools where the course work was clearly not at a college level (link).
In my opinion, HS teachers are typically better instructors than college instructors and in many partnerships, they aren't allowed to teach the dual credit courses.

Maybe pro, maybe con for dual credit

If the class is taught on the college campus, it can expose high school students to the college experience and taking a college level class while in high school.

Now, I've never been a fan of the College Board. I feel they have too much influence over American education and I don't agree with a number of their decisions in creating at least their computer science offerings. I've always believed that a HS teacher who knows their stuff could design a better experience for their students without the college board.

I've also always felt that the transcript credit from dual credit programs is safer and stronger. I still believe that but it's not necessarily a great thing.

Also, while I haven't changed my opinion on the College Board I will say that more and more teachers are being poorly prepared overall and just want a canned curriculum (something I find abhorrent). Back when I started a "curriculum" was a short line item list of topics. Maybe it had matching text book page numbers on the side. That was it - you were expected to be able to build out from there. Not so much anymore. I'm sad about this because the students suffer when it's one size fits none but I can't do anything about it anymore. Even when I could do something it was only in my teeny tiny corner of the world.

On the plus side for the College Board, or at least APCS-A is that when I taught at Hunter, I placed every student who had taken APCS-A into CSCI13500 - basically the equivalent or a stronger APCS-A with more OOP stuff but in C++ and in one semester, not a year. They all did splendidly regardless of AP score. My feelings was that one year of active programming made the difference. On the other hand, I didn't give any placement for students who took APCS-Principles and that was the right decision as some of them came in with some knowledge, but many would have been better off by taking nothing at all.

So, in my experience, APCS-A, in spite of my College Board misgivings consistently did right by the students but APCS-Principles, not so much.

What about dual credit. Well, while it all sounded superior to AP on paper, in practice, that wasn't the case. Over my first couple of years at Hunter. I met, both in my classes and just on campus, a number of students from PTECH, the school mentioned in the article I linked in the con dual credit list. In my estimation, at least in the classes that can be objectively evaluated (CS and Math) the kids learned little to nothing in their "college" courses.

It got worse. I then started getting kids who had taken a dual credit CS class so I had to place them in a more advanced section. Most of the classes were self contained - that is, only HS kids taking the classes either on the college campus or the high school campus. The level of preparation was uneven at best but strongly weighed as poorly prepared. Some of this could be institutional differences but after looking into things, no, it's just that the high school kids weren't getting the real college treatment.

This was disappointing.

So, now here comes NYSED to save the day. Requiring partnership agreements and presumably putting some state requirements on them are probably good.

The data reporting and college course standards are where I'm concerned, particularly since they can't track kids after they complete the dual credit courses.

If they're tracking pass/fail and they probably will then that will encourage the dual credit programs to pass everyone and I really don't see how they can enforce the class standards.

On the other hand, if they come down too heavy, creative programs will likely be stifled.

I guess we'll just have to wait and see.

In the meantime, we've got the two options of AP and Dual Credit. Are there any others? What does your school do and which do you think is better?

Will they?

I don't know.

pros and cons of each

my experience with AP students at Hunter

my experience with dual credit at Hunter

CS Ed Podcast, connecting with students, and a little AI

Sun, 07 Sep 2025 14:38:00 -0400

I listened to the latest episode of Kristen Stephen-Martinez's CS Ed Podcast. The episode focused on Professor Emeritus, William G. Griswold, of UCSD's practice of requiring students meet with him outside of class early during the semester. These meetings weren't about class material but rather to foster a better connection and to increase student engagement.

Good episode. Take a listen.

The class in question was an upper division software engineering class - mostly majors but a significant percent of non-majors. Due to class size, individual meetings were impractical so students met in small groups.

One interesting thing that came up, but it's really not a surprise was that in spite of this being an upper division class, one third of the students had never spoken to a CS professor prior to these required meetings. Another interesting part of the discussion involved AI but I'll get to that later.

I love this practice.

Clearly Professor Griswold cares about his students and understands the importance of making connections. I also love the small group aspect of these meetings as it makes the meetings less intimidating and also helps develop connections from student to student as well as student to instructor. It also didn't seem to be a huge extra expense in time. By scheduling these meetings early, they can essentially fill in for office hours which, for many instructors are sparsely attended, particularly early in the semester.

Unfortunately, I'm not convinced that too many college instructors would implement such a practice because. I'll keep my reasons to myself so as to not offend anyone.

When I taught at Hunter, I would encourage students to come to office hours but due to my specific situation, I didn't feel a need to use office hours to build a connection. I was running a special honors program and would be running special events throughout the semester with the kids so we all had plenty of opportunities to get to know each other.

When I started teaching at Stuy, around 1993, I tried a similar practice, but in high school, it just wasn't practical. Instead of around 200 students, the number Griswold cited, I had about 150 but I couldn't do group meetings since I was teaching regents math and the discussions would frequently have a component about the students work. High School teachers also just don't have the time or space. I taught 5 out of 8 periods and had a professional period where I was tasked to do something by the school. That left 40 minutes for lunch and 40 for prep - not enough to even get grading and lesson planning done let alone meeting students. Couldn't really do it after school either. Besides me no longer being on the clock, can't ask students to arbitrarily stay.

So, while I love the practice, in high school, it's largely a no go unless the school actually sets up a system where these connections can be made. Back in the day, we had homeroom - 10 minutes every day with the same homeroom teacher. Not the same as connecting with all your instructors but it did, if the teacher and class wanted, allow for connections within that cohort. Unfortunately, at least at Stuy, daily homeroom is no more. Other schools have "advisory" periods which can function similarly but I think they're too few and far between. The human side of education has consistently been marginalized for more test prep and teaching for jobs.

That human side brings me to another point of the podcast - the rise if genAI, both as a resource for students on their own and genAI TAs and tutors.

In the podcast it was noted that students usually won't go to a professor until after they've exhausted other resources including genAI. With AI tutors and TAs it's even worse - with them, institutions are effectively encouraging students to avoid human interactions.

I mentioned this in my last post - how the big difference at Stuy was that kids are so much more self absorbed in their phones and other devices and not interacting with each other. That's also a theme I keep hearing across education - that kids don't know how to interact with each other nor with adults. This is why things like advisory or regular homeroom in high school and practices like Prof. Griswold's are so important - so much more important than people realize.

I've long maintained that connections and relationships are critical in education so thanks to William G. Griswold for your practice and for sharing it and thanks to Kristen Stephens-Martinez for providing the platform.

Looking at the start of school for 2025

Mon, 01 Sep 2025 19:34:21 -0400

Today is Labor Day. That means that tomorrow NYC public school teachers return to work. I mean, not me - I'm still happily retired but still I have many friends going back. Kids come back on Thursday.

Of course, I don't doubt that I'll have a "going back to school" dream tonight.

I'm not going to make any predictions here but there are a couple of big tech related elephants in the room that will be interesting to keep tabs on over the course of the school year.

The first are New York's "Distraction Free Learning" policies. Specifically, a cell phone ban. While students will be able to bring their phone's to school, they won't be able to use them during the school day on school grounds.

It's amazing how embedded cell phones have become in our lives. Twenty Five years ago, I didn't have one, many people didn't and smart phones certainly weren't a thing. I got my first cell phone shortly after 9/11 - a cheap pay by the minute plan. Having been essentially around the corner during 9/11 and since we were going to be relocated to a different school, we thought it made sense that I have one.

Of course now, everyone has smartphones and the kids are on them all the time.

I taught for a couple of weeks at Stuy last March and people asked me if things had changed. Well, the students in my classes were largely the same, and I say that in a good way and the overall feel seemed similar in the school but the big difference I saw was that when kids weren't in class, they had their nose buried in their phones.

Back in the day, when kids were free they'd hang out in groups along the halls. They still do. The difference though was that back in the day, a couple might be doing homework, a couple playing Magic the Gathering, and the rest talking with each other.

Now silence. They're all alone in their phones but sitting together.

A friend told me that fire drills were even worse. Back in the day, the challenge was keeping them quiet and stopping them from sneaking off to Fake Terry's. Now just row after row of phone zombies.

Funny thing is that in class during my two week stay, I didn't find the phones to be a major problem. A couple of kids went to them but since I was a sub, I didn't stress about it.

At other schools, the problems might be reversed.

I don't know how exactly how Stuy or other schools are going to be handling the ban - I've heard that many schools are issuing signal blocking pouches to the kids but I don't know how well that'll actually work but we'll find out soon enough.

I am interested in seeing though if this new cell phone ban is implemented seriously and if it has a positive effect on students.

The other is of course AI for both CS and non CS classes.

While AI has become entrenched as a coding assistant and general assistant, the landscape has overall become more complicated. For college students looking to enter tech, some companies want candidates to use AI in their interview process and others forbid it. When writing up the CSTA conference a month or so ago, I talked about how a founder commented on how students were writing better through the use of AI but on the other hand, the majority of teachers I've spoken to talk about how AI has made their writing worse.

It will be interesting to see how schools and colleges continue to navigate this landscape.

I was thinking that if I were back at Stuy, I don't think it would be that big of a deal within my classes. I mean, other than setting policy. I was thinking this due to the large amount of lab time in the courses I designed and taught. Even with over 30 students per class, we always spent sufficient time with them working on projects during class so that I knew what my students were doing and what they could do. That plus other in class assessments I don't think I'd have to worry too much about AI use beyond whatever the policy was.

At Hunter, or in college in general, I'm thinking it would be much harder. At Hunter, I only met with my students a couple of times a week and it was all class - not a whole lot of project time. The kids did all of that outside of class. That means it would be a whole lot harder for me to assess where they're at.

Like I said up top - not doing predictions here but I think the cell phone policy and AI are going to be interesting to keep tabs on over the coming year.

I'm not sure what this school year will bring for me but I'm hoping that an interesting opportunity or two will pop up so that I'll be able to observe some of this first hand.

Rossini by the sea

Sat, 23 Aug 2025 12:20:42 -0400

We basically finished up our trip in Pesaro, a resort town on the Adriatic Sea. Well, there plus one final day in Rome before our flight home.

Why Pesaro? Seaside resorts aren't usually our speed. Well, Pesaro is the birthplace of composer Gioachino Rossini and hosts an annual Rossini festival. Now, I like Opera but I wouldn't travel across an ocean to see some Rossini so how did we end up there? It's Natan's fault. This year was the first year since graduating Michigan when he's been able to take a real vacation. We decided we wanted to do a family thing and since he's had the fewest opportunities to travel gave him the choice.

He, being a Rossini fan, decided he'd love to attend the festival. We built the trip around that. Unfortunately, as the planning progressed it turned out we weren't able to do a whole family thing since Batya had to be up in Illinois to start grad school concurrent with the festival. Still, a great trip.

As Rossini's birthplace, Pesaro has a Rossini museum and you can tour his birthplace. Both very interesting even if you're not big on Rossini or opera.

Beyond that we didn't do too much touring.

We did go down to the Adriatic where we found this sculpture:

Globe

It reminded me of the World Trade Center Sphere crushed on 9/11 that was on display outside of Stuy for a while until it was relocated to a permanent home in Battery Park. Any similarity though was just coincidence as the Globe in Pesaro was created and installed years prior.

Also, in spite of my persistent hip issues, I did get a run in early by the sea:

Sunrise by the sea

And another highlight was yet another Church - the Cattedrale di Santa Maria Assunta. A thirteenth century church built on top of a sixth century Basilica which itself was built atop a fourth century Roman Basilica. The cool thing was that part of the floor is exposed to show the extensive mosaic work on the earlier floors.

Old Mosaics

For the festival, I saw two operas, Zelmira and L'Italiana in Algeri (The Italian in Algeria).

I enjoyed both but Zelmira, being a production in the round (in a modern space) detracted from the performance. If you want a deep dive into both productions and the festival in general, check out Natan's blog.

L'Italiana in Algeri was staged in the Teatro Rossini, an 1818 old school opera house.

Teatro Rossini

Really fun opera and production and great performances and it was a trip to see it in an old school venue. For details, again I refer you to Natan's blog.

The downside of the Pesaro part of the trip is that we were well into holiday season so even more restaurants were closed. Combine that with Italians eating later (most restaurants didn't open until 7:00 or 7:30) and the operas starting at 7:00 or 8:00 made dinners a challenge. That said, we did have some good food including a terrific no frills seafood place for one of our lunches.

From Pesaro we were back to Rome for a day before our flight. There we checked out two more sites.

First was the Vittoriano and the Museum of the Risorgimento within. THe Vittoriano is a monument to Vittorio Emanuele II, the first king of unified Italy and is really impressive inside and out. The museum was really cool telling the tale of Italy's unification.

The Vittoriano

Our final official stop was the Capitoline museum, right next door although we had to walk down the hill and right back up. Of course, every time I read "Capitoline Museum" I hear Samuel L. Jackson saying "what's in your wallet?"

On the Capitline Hill, the museum (or collection of museums) is an archaeological museum with artifacts from ancient (Greek) times through the Renaissance. Highligths include Marcus Aurelius on a horse, a bust of Medusa by Bernini. and much much more.

Capitoline Museum

After that, we had our last dinner (last supper?) and then the next day, back home.

Great trip, now to plan the next one.

Bologna Bound

Mon, 11 Aug 2025 04:56:37 -0400

Next three days were in Bologna.

Highlight was getting dinner with former student and friend Paul who is now living in Modena and working for the Ferrari Formula 1 racing team - how cool is that. I also realized over dinner that I've known Paul now for over 20 years - man, I'm getting old.

Bologna's usually a few degrees cooler than Rome but overall the mercury kept rising so outside wasn't particularly comfortable.

On the plus side, all the streets in the center city have porticos and this helped a lot. There was almost always a shady side of the street where it felt much cooler. I was still dripping with sweat all day but the porticos definitely helped.

Porticos

For me, other than the food (which I could also say for Rome), the highlight was probably the Basilica di Santo Stefano. It's what I'd call a Frankenbuilding, that is a building made from multiple buildings. I always found those pretty neat. Seems that there are a lot of bookstores that are Frankenbuildings - Powell's in Portland, City Lights in San Fran, and E. Shavers in Savanah. Even the old and now closed main branch of Barnes and Nobles on fifth avenue in New York was two buildigns smooshed together.

In this case it started as one church and over the years, additional churches were stitched (built) on. Really fascinating.

From the outsider

One church

And another

And even another

Not done yet

There was also a little museum.

Excellent!!!

We also visited the Basilica di San Petronio which boasts the worlds largest sundial that's basically a line with markings running from the entrance to the altar area.

The Basilica

Inside

Sundial

This was in the city's main square. Also on the square is the Salaborsa Library , under which there are some neat ruins.

Under the library

More ruins

An Neptune.

Neptune

We were told by a local that Maserati trident logo was taken from this statue.

A short walk away was one of the original buildings from the University of Bologna - the oldest continuously operating university in the world where we saw an original operating theater and a room where Rossini premiered one of his works.

operating

To round out our trip, we also visited their Jewish museum, their museum of music which had interesting instruments, old scores, and more as well as the Museo Davia Bargellini which was a cool little free museum featuring "industrial art."

Super hot but a good few days.

Now off to the Rossini festival in Pesaro.

Rome in 3 days

Wed, 06 Aug 2025 02:00:56 -0400

Time for a trip report.

Been in Bologna for a few days now but before that, we started in Rome. Now, I was correctly informed that Italy in August isn't the smartest move - it's hot - 97 degrees out as I'm writing this post. Not pleasant outside. Still, Rome was pretty awesome even if I did fade in the afternoons.

Why Italy? It was Natan's choice. Having graduated right before Covid, Natan hasn't had a chance for a real vacation for years. We thought we'd do a family thing. He's a big opera fan and there's a Rossini festival in Pesaro Italy starting on August 10th so that was the target. A few days in Rome, a few in Bologna, down to Pesaro for three days and the festival, then back to Rome then home.

As usual we did a lot of walking. All around Rome.

Here are some of the places and things we visited.

Trajan and Titus's Baths
Papal basilica if Saint Mary Major
Moses Fountain
The Colloseum / Forum / Paletine Hill
The Great Synagogue
Castel Saint Angelo
Saint Andre Della Valle
The Palazzo Farnese
Fountain of the Four Rivers
Saint Agnese in Agone
Largo de torre argentina (where they went all stabby on Julius Caesar)
The Pantheon
Capuchin Crypt and Museum
Galleria Borghese
Spanish Steps
The Trevi Fountain

Here are some highlights:

Maybe the best bang for the buck was Castel Saint Angelo. It's a little off the main tourist path so it wasn't that crowded but it was pretty amazing. Originally built by Hadrian for his mausoleum it was later used as Papal residences as well as for a prison and a fortress. Inside it's got it's own architecture and artifacts and there are also museum sections. The result is you get ancient roman history and art, Catholic history and art, military stuff, and more. All with some great views.

We easily spent a morning there.

Castel Saint Angelo

A trap door

Some nice views

Saint Michael at the top

We only ended up going to Castel Saint Angelo due to Natan. As I mentioned, he's really into opera. Last year we saw Tosca, an opera by Puccini. Tosca is an opera in three acts and each act is at a real location in Rome. The third act takes place at Castel Saint Angelo where Tosca flings herself off the parapet. Act 2 takes place at the Palazzo Farnese, now the French Embassy and the first at Saint Andre Della Valle, a nearby church.

The other unexpected highlight was the Villa Bourghese.

The Villa Bourghese

Great collection of artwork in an amazing space.

One of the rooms

A "Last Supper" that we prefer due to the cute doggie:

The Last Supper

A cute doggie sculpture:

Cute

Even the ceilings.

Ceilings

Also highly recommended.

Of course, in our wanderings we saw so much more:

Stabby stabby

Moses

I won't bore you with more pictures but the Jewish Museum and Synagogue were also really interesting. Learned about the Jewish population in early Rome, through the expulsion, discrimination under Papal rule, periods of acceptance and more. Obviously the other big sites were also pretty awesome.

On our last day, we also visited the Capuchin museum and crypt. Fascinating. In addition to the museum with a history of the order the crypts are decorated with the bones of the friars (more info here). Really interesting.

One of the crypts

The Churches were also all amazingly ornate. I appreciated the beauty but also questioned the excess having seen these extravagent houses of worship which were only matched by the palaces of the ultra rich and powerful like the The Palazzo Farnese.

Finally, lots of good food and much gelato - a necessity in 90+ degree heat.

So that's the Rome report. Next up Bologna and tomorrow we leave for Pesaro.

CSTA 2025 7 - the good, the bad, and the Cleveland

Wed, 23 Jul 2025 08:31:44 -0400

Lets start off with Cleveland and then we'll get to my suggestions for the CSTA.

Cleveland

I never had a chance to explore Cleveland before. I was actually in the city once before, for a junior national fencing competition but never saw anything other than the competition venue and the hotel.

This time, I didn't have much more of an opportunity. Normally, I like heading to a conference a day or two early or stay a day or two later to check out the town but this time, schedules didn't allow it. I drove in the day before the conference and had to leave before the conference ended. That basically gave me Tuesday morning - a half day to explore.

Started the morning with a run by lake Erie.

Afterwards, Devorah and I got breakfast at a local diner and then wandered the city, or at least the downtown. We used Adventurelab as a guide. Adventurelab is an offshoot of Geocaching where people hide little boxes all around the world. People find them using a GPS, or now the app and discover cool places and things. Adventurelab is essentially the same thing without the little boxes. People write up the experiences to share local treasures with visitors.

This took us past the Erie Street Cemetery - Cleveland's oldest existing cemetery, The Hermit Club - "the nation's oldest private club devoted to the preforming arts," the Old Stone Church, and other local treasures. Along the way we also passed a great Lebanses bakery, Clevo Books - a bookstore that specializes in foreign books translated into English, and other local treasures like the Arcade:

On the way to the Old Stone Church, w passed the Soldiers' and Sailor's Monument - a civil war monument. Outside, we started talking to a volunteer who was doing some gardening. He gave us a rundown on the monument and it's history as well as some other local history including showing us where Charles F. Brush lived, an inventor who worked on the arc lamp.

We got to talking architecture a bit and mentioned that we had walked past the Cleveland Bankers Trust Building. He asked if we had gone inside, now a supermarket. We hadn't but later in the day we did. Magnificent.

All in all a pretty cool whirlwind tour. That plus the CSTA's reception at the Rock and Roll Hall of Fame let me leave feeling that I got a reasonable taste of Cleveland.

Now, on to the good and bad from the conference and my suggestions for the future.

CSTA the Good and the Bad

First off, great overall conference. I personally would have liked to see more sessions with more advanced HS content but I also understand that the majority of conference goers don't really need that.

I'm going to mostly talk about a couple of negative points here and a couple of small positives since pretty much all the posts up until now have been glowing. Just remember as you read on - it was a great conference and just because this post highlights one or two things I didn't think were terrific, it was still a great conference and I'm 100% going to be back next year.

I liked how they mixed in regional meetups in the schedule, making it easier for people to get together although I also appreciate the New York meetup after the conference which also facilitated that but with the chapter picking up the tab for drinks and snacks :-) .

One really simple thing I'd suggest is to sell additional tickets to the opening reception. I know this doesnt' affect a lot of people but it also should be pretty easy. My wife usually comes with me to conferences and I know a handful of other people who also bring their partners. Our partners play tourist by day but they also know some of the other conference attendees. TL;DR. they'd like to attend that opening reception but there' no mechanism to do so. To be honest, we're probably talking about so few people that the door police could probably just look the other way but why not just charge whatever the cost is for these few extra people.

They do have +1 tickets available for the "after hours" but those were expensive. I'm not complaining. Last year, in Las Vegas, a +1 ticket was either $125 or $150 which seemed like a lot and pool parties aren't my thing. I was told by someone on the inside though, that the cost was what the venue wanted so I'm not complaining about the CSTA here.

This time, I opted to go to the event since I wanted to check out the R&R Hall of Fame but Devorah opted not to since a regular day ticket was only $40 vs the $125 for the +1 ticket.

My other big suggestion has to do with the keynotes and the Impact (nee equity) fellows.

This year, I found all the keynotes lacking, mostly because they were all really short and this was exacerbated by having pseudo panels. Losing a good chunk of the keynote slot to one or more impact fellows didn't do it for me. The keynotes are supposed to be special - but idea inspirational talks and this year they were mostly teasers for later sessions. Loosing that chunk of time makes a difference.

I'll also say that the Impact Fellows talks really didn't do anything for me. Some of their stories were more compelling than others but we were in a room of hundreds of teachers and my guess is the story or the message from one of the Impact fellows isn't any more or less powerful or relevant than anyone else in the room.

I guess, and I'm sure this isn't a popular opinion, that I don't get the Impact fellow thing. Sure, it's a nice reward to a small handful of CS teachers but I'm not seeing the impact. I can't look over the bios and statements of all the fellows since the CSTA website seems to be having issues but I've read their articles as they've come out and seen the bios. Some have come in having been real leaders with real experience while others are still wet behind the ears. The value of the content I've seen is also mixed. Some great and some not so great but in any event, I'm not seeing a lot special, or rather I see as much or more value from everyday CS teachers who aren't fellows on the interwebs. Actually, back when blogging was more popular I'd say that there was more good content being shared and discussed by the community at large although the community was admittedly smaller.

So anyway, I'm not seeing it. Great reward but to me it takes away from all the other teachers who aren't Impact fellows who are doing the same or more. In any event, if you want to highlight the Impact scholars, make an impact scholars session where they can share their stories and leave the keynote slots to the keynote speakers.

Just my opinion.

Another small thing that I liked was that email beforehand telling us that coffee/tea/pastries would or wouldn't be available in the morning before the start of the first session. This is a small thing but it's a nice touch. Before I weened myself off caffeine, I needed my cup of coffee in the morning and hated it when I spent far too much for overpriced hotel or convention center coffee only to find it there available at the conference. On the other hand, knowing it was there was a really nice touch.

So, that was CSTA2025. I got a couple of other side blog post ideas from the conference so maybe I'll write those up, probably after we get back from our trip to Italy which we're embarking on soon.

If you were at CSTA2025 I hope you enjoyed it as much as I did and if you weren't I hope you get a chance to go next year in New Orleans.

CSTA 2025 6 - Exhibit hall and receptions

Sat, 19 Jul 2025 15:29:30 -0400

Exhibit Hall

Let's start with the exhibit hall. First things I noticed that while Amazon, Google, and Microsoft were all listed as sponsors, none of them had booths in the exhibition space. This was a big change from past years. I'm not sure what to make of this since all three have either products or programs relevant to K12. Maybe nothing going on here but I think this was worth noting.

So, who was there?

A lot of robotics and physical computing players - Raspberry Pi, Lego, Ozobot, and more. A lot of tables with electronics on them. Since I'm not teaching and not doing anything with hardware, I didn't speak to anyone at these tables.

There were also a lot of organizations. The expexcted host tables for CSTA and also the expected code.org, The College Board, and the other usual suspects. Cleveland State had a booth - that's where I followed up on the session on teacher cert in Cleveland as did the University of Nebraska and Carnegie Mellon. Also SXSW edu, U Teach CS, and a bunch of others.

Another well represented area were online IDE providers. I spent a fair amount of time with my friends at Pickcode and at CodeHS but there were others in the room including Kira Learning, Juicemind and a handful of others that I didn't check out.

With fewer representatives there were still a couple of straight up curriculum providers, autograders, AI related vendors and the like. Interestingly enough, only one, if I'm remembering correctly book vendor. Maybe that's because K12 CS is in some ways not well enough defined to have lots of books sellers pushing products or in the case if the high school offerings - frequently APCSP, APCSA, too well defined with teachers relying on full online offerings and maybe eschewing traditional dead tree resources.

While I probably spent most of my exhibit hall booth time catching up with old friends - specifically at the CMU booth, Pickcode and CodeHS, I also spent some time at the Cleveland State booth and with U Teach CS where I was able to find out a bit more about how things were going on in their respective states in terms of teacher preparation - long story short - while individual organizations and institutions are fighting the good fight, overall I haven't found a state that, in my opinion, is doing it right. No surprise there.

Since I did spend a good amount of time looking at online IDE booths, I guess I should say more.

Now, I've known the CodeHS people for along time and liked them for a long time In the past couple of years, got to know Pickcode and like that very much as well. I also see both as filling important roles. CodeHS has been around longer and is more fleshed out. They provide full curricula and I guess what I'd call end to end tools for HS teachers. That's probably why I never used them personally, okay, the real reason is that hardly ever was in a situation where my classes used an online IDE, but beyond that, what they've built is terrific for new and less experienced teachers but by the time they came to the game, I had my tools, techniques, and experiences worked out. I briefly used repl.it (back when they were cool) for a summer online experience but I liked that because it was bare bones - just the IDE, perfect for me. That said, I've frequently recommended newer teachers check out CodeHS because of everything they provide.

Pickcode on the other hand is newer with a leaner experience. That's my personal preference albeit a subjective one. Makes it easier for a person like me to do my own thing and just use the tool where I want to. While I've been recommending CodeHS to some people for their complete curricula, tutorials and more, I've been recommending Pickcode to others - for people who want a simpler interface and product and a more agile company that's still young enough to be able to tune their product to a single teacher's needs, Think of it like a boutique product in terms of service but not price (this isn't to say that CodeHS isn't responsive but that's just an advantage of a smaller company vs a midsize one). I do want to note that Pickcode also does have learning resources as well as the IDE and a language that's transitional from Scratch to Python but I haven't personally played with that.

I also spent a little time with Juicemind but so far I've found there web site less inviting. On the other hand, my friends at Siena use them for some programs and they wouldn't use them if there wasn't something there.

There was also a booth that I didn't check out that had IDE on the sign but was also touting autograding.

The other player that, while I didn't spend time with them at CSTA, I did at SIGCSE was Kira Learning. That's Andrew Ng's new outfit. They're touting an all in one learning system with AI grader/assistant.

To be honest, I wasn't impressed and I'm saying this as a guy that really loved Andrew Ng's ML class when it first hit the scene. I mean, it seemed fine but nothing special or amazing. Maybe I'm missing something but the most exciting "new" thing I saw from the demo and from playing with it was that it had Andrew Ng's name attached. Seemed like when a Broadway show counts on a big name to sell tickets rather than the show itself and has the big star push the show to even greater heights.

So that was it for vendors for me.

Hallway

Also big at CSTA was the hallway track. I've always said that CSTA is more of a people conference than a session one.

As I mentioned, I had a chance to spend time with a friend from CMU, the Pickcode team and reconnect with CodeHS and also had a chance to catch up with a number of New York and former NY CS educators throughout the conference (in no small part thanks to Tom O'Connell for organizing the NY Meetup during the conference). I even managed to steal a few minutes with Jake Baskind although I wasn't able to catch up with my code.org friends as well as a few other people who I think were at the conference but never ran into.

I also managed to meet some great new educators, or rather, great educators that were new to me. In some cases they're very experienced. Now, given that I'm your typical socially awkward, introverted tech geek, that says something about the vibe at the conference that even I can end up making new connections :-).

Receptions

Last up, the two receptions. First one was opening night. I love the way that they've been setting it up in recent years - bringing the food in to the exhibit hall. Let's people mix and mingle and get an early look at all the vendors and suppliers. Food was good and stations were well layed out so that there was never much of a line for either food or drink. Kudos, particularly when thinking back to my first or second CSTA when the reception was, well, let's just say, the less said the better. The CSTA organizing teams have really upped their game in recent years.

The other one was the evening at the Rock and Roll Hall of Fame. I skipped last year's "after hours" event. That was an outdoor pool party during a heat wave in Las Vegas in July. No thanks. Not my scene. I might have gone anyway if +1 tickets were reasonably priced since my wife Devorah was along for the trip. Trouble was that the price for an extra ticket was something like $125 or $150. I don't blame CSTA for this - I was told by someone with inside knowledge that this was the price that the venue set but I decided to pass.

This time, +1 tickets were also expensive ($125) so Devorah went to the hall during one of the days when I was in the conference (ticket price was something like $40) but since I wanted to check it out, I went.

Really fun time. There were around 1300 people registered for the conference but the Hall never seemed overly crowded. Probably because throughout the evening people switched up from hanging and chatting at tables near the food and touring the museum. Spent most of the time with a fellow NY CS educator. Thoroughly enjoyed the museum and the evening overall.

That's it for the actual conference but there's going to be one more post in the series - things I liked and things I think the CSTA should change along with a bit on actually visiting Cleveland.

CSTA 2025 5 - Day 3 part 2

Thu, 17 Jul 2025 17:39:56 -0400

Let's talk about the final two sessions I attended.

The Future of Computer Science Education in an Age of AI

This was a panel. Pat Yongpradit and BT Twarek co-moderated a panel consisting of Julie York, Maya Israel, Allen Antoine, and Jacob Shulman.

It was a good mix. Julie York is a High School teacher, Maya Israel, a researcher, Allen Antoine is the Director of CS Ed Strategy at UT Austin's Texas Advanced Computing Scenter, and Jacob Shulman is a tech entrepreneur as the founder of Jippity and long time private tutor.

Pat and BT managed the panel well - good questions and didn't make the mistake of having every panelist answer every question.

Some answers missed the mark for me like the question of how to fit AI into a crowded curriculum. There could be multiple full sessions on this one question so it was no surprise that the answer had to be pretty generic - basically view it as a tool and something to use across everything. I would have liked an answer that differentiated learning CS vs using CS and learning CS in a CS class vs a non CS class but time couldn't allow for that. The answer was probably the best that could be given under the constraints but I would have rather a question that was more focused which would have allowed a more directly actionable answer.

On the other hand, I felt the question and answer to how AI can make CS more accessible was more helpful. The idea of using AI based tools that could provide very specific immediate feedback and also how those tools can be used for self learning. While I liked this answer, provided by Maya Israel, it also left me concerned but we'll talk about that in a bit.

Allen Antoine made a good point on how AI still can instill a sense of wonder in kids which can be used to drive interest and broaden participation but I'd caution that the shine eventually wears off. He also expressed concerns about communities getting left behind which is a very real concern. Every technology that has been touted as a great equalizer has never been one. The haves get the best of the tech and the have nots get the cost cutting budget saving tech foisted upon them. AI in the best sense can uplift communities by implementing tools like what Maya mentioned and I wrote on above or they can leave them further behind if those communities get bad AI instructors to replace good teachers.

Jacob Shulman made a couple of interesting points that I didn't agree with but that's okay, as I said, this was a nicely formed panel - that means differences in backgrounds and differences of opinions.

Jacob's product is jippity.pro- it's an online larning IDE that integrates AI. It's basically a vibe coding platform. Looks interesting but currently I'm not yet sold on vibe coding as either a practical or educational tool.

Jacob talked about the value of code reading and how you can learn that way - the implication that you can get some vibe coded solution and figure it out from there. Now, I do agreee that code reading is important and we should do more of it with our classes. You can indeed learn from reading other peoples code but it's not a replacement for all the other ways we learn to code.

I mean, think about it, how often does a kid not get something in math and then they read a solution or better, step through it as instructed by the teacher, TA, or professor and they leave thinking that they get it only to realize on the next exam that they really don't.

Reading or just being dumped into existing code isn't going to do it except for a small subset of kids. I've seen this type of "just dump them into the code" work for tech pros starting at a company even when they're working on a platform and in a language new to them but I've just as often seen it fail unless the new hire got other supports and these are people who while being new hires have extensive CS knowledge.

Actually, if I'm right on this point, that's a potential danger for those self-learning tools.

Jacob also talked about the explosion in students verbal skills and that this is what he's heard from English teachers, presumably due to students using LLMs which are raising their level of discourse.

I've seen the exact opposite. English and other teachers lamenting that the kids aren't learning how to either write or communicate at all but rather they're just parroting what they get from chatGPT.

Now, I could be wrong, Jacob could be wrong or the answer could be somewhere in between. Wish we had time for a discussion but in any event, allowing for these disparate views make for a good panel.

Next up were questions. I wanted to ask one but didn't get a chance. Don't know if I would have been called on regardless but the question answer period ended when after a couple of good questions, one participant took the opportunity to share their individual views on teaching AI rather than asking a question.

So, after that, Pat and BT wrapped up the panel by asking each member to give one piece of advice on a risk with respect to AI.

Julie commented on copyright issues, Allen, ethical concerns and Jacob on the dangers of taking extreme positions on either side - always AI or never AI - all great points. Pat mentioned the concern that kids will only learn AI through prompt engineering and not really about AI - a concern I share as well.

Now, back to my question - I wanted to ask about putting up guardrails to make sure we don't lose student - teacher interaction and student - student interaction. A huge part of teaching is all about connection and relationships with one's classes and a huge part of learning is interacting with one's classmates and one's teachers. If a kid is always going to a chatbot what will happen to their human to human skills. If a teacher just uses AI assignments and autograders how well can they get to know their kids?

I was disappointed not getting to ask about this. Not because I expected a full and useful answer - the question's too big for a quick response but rather, my feeling that very few, if any attendees were thinking about this.

So, I was psyched when both BT and Maya's answers on risk spoke directly to this question even though it was unasked. BT commented on how we're teaching kids to be citizens and that has to be taught even when all the tech is stripped away and Maya hammered in a concern for mental health and an over reliance on chatbots. After that, everyone was thinking about it.

Really great thought provoking panel so thanks to all the panelists and moderators.

Impacts & Insights from the 2024 Teacher Landscape Survey

Last up was a discussion of the 2024 CSTA teacher landascape survey. The facilitators were BT Twarek so I had the pleasure of attending a BT run session back to back, Lisa Novohatski, Sonia Koshy, and Shaina Glass.

Great presentation and discussion. I won't go into any details on the survey as it will be published soon enough but a couple of points I found interesting. One was that it's really challenging to create a meaningful survey like this. The CSTA wants to get as many CS teachers as possible but they only have contact info for CSTA members. The facilitators also pointed out that even with the efforts they took they ended up withe HUGE number of false surveys (bots I'm presuming). It's also hard getting insightful answers. They can ask how much a teacher might emphasize a topic or issue but what does that mean for an individual teacher? I really don't envy the survey writers at all.

It turns out it's also really hard interpreting the answers. For example, if one is at a school with a single CS class that happens to be a requirement, diversity might not be an issue for you at all, or maybe if you're school isn't diverse it's a huge issue but you can't do anything about it.

Challenges about.

In any event, another interesting and well run session and keep a lookout for the survey once CSTA releases it.

So, that's it for the sessions I attended. Still have to talk about the exhibit hall, receptions, hallway track and Cleveland. That's all up next.

CSTA 2025 part 4 - day 3, part 1

Thu, 17 Jul 2025 14:54:44 -0400

Thursday, day 3. This was my last conference day since I had to leave bright and early on Friday. Four more sessions. As I started to write them up I thought things were getting a little long so this post will just cover the two morning sessions and the next one, the other two.

Strype

First up "Strype: Frame-Based Editing for Python." A session run by Neil Brown, creator of Strype with Miles Berry and Colleen Lewis acting basically as TAs during the hands on session.

Truth be told, I wasn't going to go to this session, I was going to use the time to visit the exhibit hall. Since there's no way I'll be teaching using Strype any time in the near future, the session wasn't really on my radar. I noticed Miles was listed as a speaker and I've known him online so I figured I'd try to introduce myself in person before the session then I'd hop.

I did end up chatting with Miles which was great but then I decided to stay for the session anyway. Glad I did. As a retired teacher I'm not going to use Strype but the it was a well run and enjoyable session. Neil Brown spent the session walking us through a small hands on project using Strype.

For those of you do don't know, I didn't, Stype is a "frame-based approach to Python." You can check it out here.

I'd describe it as a tool halfway between drag and drop and a text based Python site. It reminded me of a site I used a number of years ago - codesters.com. I also took a few minutes during the session to check to see if codesters was still live and if it had changed.

Strype provides a clean, simplified interface. A section for imports, one for function definitions and one for the rest of the code. It's also fairly constrained. You can't just click on an area and start typing. You have to make sure you're input marker is at the correct location, hit the "hot key" for the thing you want to create (assignment, loop, function call etc.) and then you type the details.

This might sound restrictive and it is, but I think in a good way. My description isn't doing it justice so you should check the Strype site. The environment is constrained enough to help keep beginners on track. Keep them from being overwhelmed and lost but free enough so that I think they'll feel like they're doing traditional coding (which they really are).

Strype also has access to graphics and I'm not sure what else.

One issue that came up was that I was using an Android tablet with keyboard and there were some weird interface issues. Apparently the development effort right now are for the mainstream browsers. I hope though that they can lend some attention to Android since that's what chromebooks run and lots of schools and students use chromebooks.

Fun, well run session and a nice coding platform. If you're looking for a beginner Python platform with some structure, I recommend checking it out.

Insights from Industry: Projects to Bridge the Gap

Second was "Insights from Industry: Projects to Bridge the Gap" run by Joseph Maxwell.

Another good session.

Joseph works at Hyland - a tech company and there, he runs a tech outreach program. Joseph specifically focused on a summer program where High School students "interned" at Hyland - building a tech product while learning industry standard software engineering techniques. I put interned in quotes because the students aren't working on Hyland's codebase. Rather they're building their own projects - something that hopefully will be both useful and complete-able by the end of the program.

After going through the programs five-ish year history, Joseph shared lessons learned.

Teams of 4 or 5 worked best. This is something I learned long ago at Stuy. More than 5, or even 5 and the group is too unwieldy. Three or fewer, not enough to really develop good SWE skills.

For the program, kids had to come in with some experience. Joseph described it as "intermediate experience." I'd say that probably means at least a year of programming but not necessarily things like data structures. In some of the early renditions of the program, they accepted kids with zero experience and the program was more like an intro programming boot camp or class.

They also learned that in person was important for a variety of reasons - team building, having mentors available, and being able to insert group diversions were among those reasons. Being in person though would make setting up a similar program in a rural location or a in fact any location without many tech companies.

It was also important for the students and company to have shared ownership of the project.

The program covered full stack development and Joseph shared a number of resources. Here's a link to his presentation page which includes links to slides, resources, and some of the student projects.

The whole gist of the talk was about giving students a real world experience and introduce them to professional sensibilities but in a safe, relatively low stakes, fun environment.

I love the program but unfortunately there were some severe limitations. Joseph and Hyland are doing this, I'm guessing, to give back since I don't recall Joseph mention any external funding. They are also only working with a small handful of students. That's great for those students but really limits the scope.

It's really a challenge to have a company dedicate the resources needed to support student programs. I ran a program for Google back in 2010. Basically had to take a conference room offline for a month to host the kids. When I was told how much it was "costing" them, I was shocked. You can also think about how hard it is for small companies to run effective college intern programs and those are for kids with much more experience. Having companies dedicate the amount of time and resources needed to support high school students on a summer or even after school project is a big ask.

So, it's a great program but unfortunately can only affect a few kids at a time.

On the flip side, a number of schools do run courses where students do make full stack web apps. Some are more effective than others.

I think my SoftDev class at Stuyvesant worked particularly well but there were a lot of factors that went into that.

The kids all had 2 solid years under their belts. An intro class plus a college CS1 (programming) and CS2 (data structures) so they've written a lot of code in multiple languages.
They knew each other from the prior courses so community was already built.
From those courses they were already comfortable with some of the industry practices and tools.
From all my graduates I had a lot of industry connections that I could tap into and since we were an in school class, they didn't have to act as project mentors. My asks were much more manageable.
As a year long course we have lots of time and can even do things like have groups work in each other's code bases.
I'm a teacher with a strong CS background and professional experience (albeit from the stone age) so I could handle teaching all the technologies and techniques and could directly support the students.

I know other schools and teachers who've done web app courses and many are good or great courses but very few can effectively cover all of the SWE techniques they might want to. The kids by and large just don't have the keyboard time and experience under their belts before starting the course. While at Stuy they have an intro course more rigorous with much more programming than APCS-P and the a superset of the old APCS-AB, also with lots of programming. In many cases, the kids at some of the other schools I know with webdev courses, the kids might only have APCSP or not even. Still often great courses but some things just take time and require experience.

So, what can a teacher do? One thought would be to create something in partnership with a college. College students as mentors would cost less, or maybe even something could be set up as a dual credit course and those college students could be employed by the college as TAs and then act as mentors. True, college students won't have the same SWE experienced as a pro but a third year could certainly have enough experience to be a valuable mentor.

Tech employees could still help support high school students but it would be easier to make a more limited ask - not directly mentoring a full project like a PM but maybe less intense general mentoring.

Another strong session and a great program.

Next up, the rest of the sessions I attended.

CSTA 2025 3 - day 2

Wed, 16 Jul 2025 08:49:46 -0400

Four sessions up for today.

First one "Certified to Rock: Cleveland's Journey in CS Teacher Prep." Basically a discussion of the pathways for teachers to become CS certified in Cleveland. It wasn't a crowded session and most of the attendees were affiliated with Cleveland's certification programs in one way or another. Then there were a handful of people like me from assorted locations who wanted to hear more. I mention this not as a criticism but to wonder about the composition of conference attendees. It would make sense that most teachers in Ohio would already know this info and it probably wouldn't be of interest to teachers in other states. That would leave people in higher ed, advocacy, or government who might be working on certification or teacher training in their own states. The fact that this wasn't a super crowded session as compared to the direct teaching related sessions is probably a good thing.

The speakers took us through their journey starting in 2014 citing the same reasons as most other states for pushing CS education. They ended up with three pathways. An endorsement which is composed of 6 courses, a PD model and what they called advanced training. I was left wondering if both the PD model, which I think is woefully insufficient is still used now that the endorsement exists and wasn't clear as to what the "advanced training" entailed.

As to the 6 course endorsement. It sounded similar, at least in spirit to what I designed for Hunter College's Advanced Certificate program. They said that programs had to cover both CS content and pedagogy but didn't share details. Their program, like mine was all remote - something that I don't think is ideal for teacher prep but when you have to accommodate teachers all across the state from one location, there really isn't any other choice. I also left feeling that just like in NY, some Ohio institution could offer a really awesome 6 course sequence while another could create 6 joke courses, focus on their state's content exam, and call it a day.

The session also featured teachers who have completed certification sharing some of their work.

All in all, an interesting session for me. I had a few questions afterwards but I was able to find out some more information as the speakers had a booth in the exhibit hall.

For a future conference, an interesting session might be to have four representatives from four different states compare and contrast requirements and offerings.

Next up was "Charting the Course: Building Data Literacy and Data Science Learning Progressions" run by Tiffany Davis, Chris Orban, Jake Koressel, and Mahmoud Harding.

For my money this was the session of the day.

The lead question was "What should students know about data science?"

Some of the things that were brought up was how that in some locations data literacy is already in math and science standards. Also challenges like that in Ohio you can only teach a course in data science if you're a math teacher, other teachers can only integrate data science into other classes.

Also, as part of the discussion, Jake Koressel spoke on data science, data literacy and the CSTA standards, currently under revision.

What made the session so compelling was how it was run. During part of the session, run by Mahmoud Harding, there were cards on each table with various aspects of data science and data literacy. At the tables we discussed what aspects were most important and why.

This lead to a discussion of what was important for all students vs what was important for a student using data science in a career.

The discussion was rich and valuable. At one point, at my table, we were discussing fundamental skills. One person at the table was pushing story telling - something that is indeed important since at the end of the day, it's the story that wins support. I pushed back that having a data BS detector was more fundamental since story telling can be quite convincing yet devoid of facts.

I can't speak to all tables but the table I was at had healthy respectful disagreement, compromise and resolution.

Most importantly, I think most of the attendees left the room thinking about the issues.

Another highlight of the session was that after knowing him online for years, I was finally able to meet Chris Orban in person. Same for another couple of people at the session.

Third session wasn't really for me. The title was "Faith-based and community partnerships to broaden stem access for Youth." I was curious about this session for two reasons.

One, if it was an inappropriate session (it wasn't). I was turned off by the listing of a BOF at SIGCSE. It was titled "All Scripture is Useful for Teaching: CS Education through a Christian Worldview." It was a BOF specifically advertised "for Christians at SIGCSE…" Now, I wasn't actually against the session and considered going to it. I was turned off because right prior to SIGCSE there was a big todo about some sessions being rejected because they weren't inclusive or the language in their description wasn't inclusive. Using the language "for Christians at SIGCSE" to describe who the session was for is certainly not inclusive.

Anyway, that was SIGCSE. Part of me was wondering if this session would be similar.

On the other hand, I figured it would more likely be something interesting and useful, something like "hey, you don't have an after school program, here's how you can partner with your Church, Synagogue, or Mosque to do some cool stuff for your kids and your community."

So, the session was more on the later but was really more of a report on the partnerships that a group at Texas State set up. I didn't feel that I walked away with any actionable advice.

For the last session of the day, I attended "Handy 'back pocket code' and code explanations for APCSA" run by Colleen Lewis and Maria Camarena.

Since I'm not teaching APCS and won't be in the future, I knew that there would be nothing useful for me personally at this session. I also knew that Colleen runs a great session so it would be fun. Also, Colleen made a throw away remark that I thought was really interesting and will likely spawn its own blog post after I'm done with the CSTA posts.

The gist was using the idea of literate programming - that is code surrounded by extensive commenting. Here's an example: link. What was interesting was that different teachers, some who've used the idea and some who didn't had very different takes on how, when, and why use them. Some thought that the annotations could be too busy but felt that annotated snippets could be a great resource for review and reference. Others liked them later in the course when the annotations were more targeted. Still others liked the idea from the get go.

I loved it. I'm a big proponent for teachers having a large tool kit of techniques and pulling them out and using them, customized, as needed. This was just that.

As I said, nothing for me personally as a retired teacher but fun, informative, and I'm betting pretty useful for attendees still teaching CS.

So, that was day 2.

Still to come - day 3, the exhibit hall and hallway track, the reception and party, and a bit on Cleveland.

CSTA 2025 Part 2- Opening day

Tue, 15 Jul 2025 12:33:51 -0400

With the keynotes out of the way, let's get to some sessions. First up, Tuesday - day 1. The morning was set for workshops. I didn't attend any of those but I did go to two afternoon sessions. One was alright but, at least for me, missed the mark and the other was really strong.

The first one was titled "AI Ready: Future-Forward CS Education." The talk was on how Gwinett County Public schools have been implementing AI tools and teaching AI.

The session left me a little flat. The talk as very much at an overview level. They talked about the fact that all schools offer some manner of CS and also gave some examples of AI related activities. Two that I recall are one, a Tic-Tac-Toe game that they cover in the earlier grades (K-5) and another is a discussion of the idea that in the NBA AI is used to determine player risk in contract negotiations.

The problem is, that they really didn't get to the meat - how do they implement the tools and how do they teach the AI. Also, how do they prepare the teachers.

With the Tic-Tac-Toe example, it sounded like they used a premade site where they "trained" a program to play Tic-Tac-Toe. It's unclear how much AI is being taught and what understanding is really being developed. I remember a few years ago, at SIGCSE, Google showed off an AI teaching platform that had students train a pizza pr not pizza classifier. They gave examples to the computer of pizza and not pizza and eventually the program could identify pizza. The problem was that the students were given no understanding of what was going on. The most a kid would take away was the fact that it was possible for someone to write this thing that could classify pizzas.

Same for this example, at least as shared in the session. It was clear that some manner of program can "learn" to play Tic-Tac-Toe effectively via repeated input of examples of good and bad moves but nothing under the hood is being exposed. Now, I'm not saying that the schools don't go further, but I left the session not knowing anything more about implementing an AI strategy than when I entered the session.

The NBA example was also lacking. It was just a discussion, apparently because using the NBA as a real world example can be motivating but to me it wasn't really even AI but rather data science and again, no exploration of the how.

Overall, nothing wrong with the session but, like some of the keynotes, was really more of an experience report and at CSTA, if I were an active teacher, given the title, I'd want something more actionable.

Fortunately, the second session I attended, "Partnering with AI: Enhancing Coding and Debugging Skills in the Classroom" did a much better job scratching the itch. I almost didn't go to it. There was another session "Beyond OOP" which was an ML session run by John Chapin. John ran a great session on teaching AI last year so I figured this one would also be great. In the end, I decided to go to the other session figuring I can always reach out to John for his materials thus getting to see a new presenter and presentation while also getting info on the session I skipped.

The session was run by Katie O'Shea of CodeHS. It showed explicitly how AI might be used in a programming class, included a discussion of the limitations and challenges while also providing ideas on how to takes these ideas to the next level.

The talk covered strategies and difficulties in debugging and also discussed abilities and limitations of using AI as a coding assistant.

The meat of the talk was on using AI coding tools to develop code. Code that invariably has errors and then have the class dive in using debugging strategies already discussed and normalized. The speaker emphasized the PRIMM pproach (Predict, Run, Investigate, Modify, Make).

O'Shea also explicitly mentioned using bool or v0.dev to generate the code. Not that those were the only options but so that the audience had something specific they could use to get started on their own. The audience also had a chance to actually go through the process live.

At the end, O'Shea discussed some of the limitations and issues with this approach and also other possibilities for using AI in the CS class - to do a code review, to present other coding styles or solutions and more, while acknowleding that these other possibilities also have their own challenges. For instance, while an AI can share an alternate approach to students, it could also share coding techniques and constructs that students may not know or that a teacher might not want to introduce at the present time.

Wonderful session.

Given the time constraints, there were of course things that couldn't be covered. For instance, how do we know that the AI solution will have one or more errors and how do we know they'll be the types of errors we'd be looking for.

Still, I really enjoyed the session. The audience was given an overview of the problem - teaching debugging, given specific tools and techniques they could use, and even walked through the process.

Slam dunk.

I'll write about Wednesday's sessions next time up.

CSTA 2025 Part 1 - overview and keynotes

Sun, 13 Jul 2025 19:14:10 -0400

Just got back from CSTA 2025. Well, not directly. Last Monday Devorah and I got in the car and drove to Cleveland for the festivities. Unfortunately, we couldn't stay for the duration - I had to miss Friday so we could drive down to Lancaster to meet the kids for our annual corn maze weekend. Drove back to NY today.

That means it's time for my annual recap of the Computer Science Teachers Association's annual conference.

Last year was in Vegas, this year Cleveland - a much better destination as far as I'm concerned but I'll talk about the city in the last post in the series.

The TL;DR is that if you're a K12 CS teacher, you should really try to attend. The conference has just gotten better year after year and there's something for everyone - from CS newbie to industry pro turned teacher, from first year to 20+ year vet, from K all the way up to 12.

That said, to me, it's always been more of a people conference than a session conference. While I've always found some good sessions, it's been more about the people. Maybe that's because I probably don't fit the profile of the majority of attendees but that's fine, I still get and got value out of the sessions but get so much more out of the conversations and the connections (and the fact that an introverted, a-social, socially awkward computer geek like me can walk away from CSTA with new connections is a testament to how open and friendly to conference is).

For this post, I'll talk about the keynotes. Usually a highlight but to be honest, they just didn't do it for me this year. I think that was mostly due to the format. I also wasn't able to attend the closing keynote - probably the one I was looking forward to the most, so I can't comment on that one. I also want to emphasize that it was a great conference so even though I'm nitpicking on what I felt was the weakest part of the conference here, the whole show was a big win.

All the keynotes I attended followed the same basic format. The MC hyped the audience and thanked the sponsors, Various announcements were shared and we also got a flash talk given by CSTA Impact Fellows, formerly known as CSTA Equity Fellows. This was followed by the actual keynote address. The end result was that much of each sixty minute keynote block had elapsed before the keynote even started. Two of the three keynotes I attended were panels and the result was a super short keynote that, to be honest, felt lacking.

The first keynote started with representatives form CSForDetroit, Cleveland State, and CodeCrew talk about their accomplishments followed by a minipanel made up of educators who worked with each of these organizations. Given the short amount of time, we ended up with three short flash talks by the first group and a super mini panel. Overall, I felt it was something of a cheer leading talk but I neither felt inspired nor that I really learned anything. There also wasn't any time for audience questions.

Overall, it was fine, nice stories about three organizations working in CS Ed but for a keynote, I was hoping for more.

The day 2 keynote was also a panel, or rather three speakers, Maya Israel, Tia Madkins, and Deborah Fields each giving a flash talk followed by a mini panel. The title was "Theory and Practice: Innovative CS Teaching Strategies." Again, the talks were fine but I didn't leave with any teaching strategies. Maya Israel talked about barriers facing students, learner variability, accessibility and more but in a very course overview manner. Tia Madkins talked about Justice Oriented Teaching and Learning and Deborah Fields talked about bias.

All of this was fine but none had the time to really dive into their subjects. The mini-panel similarly felt short and rushed.

At the end, the MC said we could learn more by attending any of the three speakers sessions later in the conference.

Basically, the keynote was really an advertisement for the three speaker's full sessions. This left me cold. I'd much rather have heard from one of the three and actually walked away a deeper dive into the theory and maybe something concreted in terms of practice.

Thursday's keynote was a solo by Jaleesa Trapp: "Fostering a Generation of Creative and Ethical Thinkers in the Age of AI." Trapp talked about creativity and justice, and the fact the one size doesn't fit all for CS Education, though that's true for all education. She also described an environmental justice project that grew into a far greater environmental justice effort. There was a brief mention of facial recognition and the related bias but I didn't see any AI connection in the talk. Again, it was fine, but the same talk could have been given five years ago before the AI boom. Nothing wrong with the talk, it was engaging but I didn't see any strong AI link. It also felt, somewhat like a teaser for another later session.

As I said, I had to leave before the last keynote.

So all were fine but nothing particularly inspiring. I think CSTA would be better served giving more of the keynote block to the actual keynote. If they want to highlight the Impact Fellows, either extend the block length or do those together in a block of their own.

So, that's my take on the keynotes. Fine but not the highlights of the conference. We'll get to those in the next few posts.

Teaching CS in the Age of AI - my two cents

Tue, 01 Jul 2025 07:50:50 -0400

Saw this article in the New York Times yesterday. It's been making its rounds. There was some discussion in some of the CS Ed related online groups, my friend Alfred Thompson posted his take here, and when my daughter, a professional SWE at Meta for close to a decade came over for dinner yesterday, she brought it up.

I haven't posted in a while so I figured I'd add my two cents.

The gist of the article is that with recent AI advancements, CS Education has to change.AI tools could make programming a less needed or at least a less in demand skill, students can lean on AI tools to complete class assignments, and tech companies are hiring less and looking for more and more experience in entry level.

Looking at the industry sign, things look bleak. It's hard out there. The NY Times article says that "many students say they send out 100 to 200 applications" which sounds crazy but really, even in the good times, many applicants, particularly if you didn't go to CMU or MIT which only accepts students they deem to be premade for success and then the school's reputations carry them to jobs, you were already sending out that many resumes.

It sucked then and it sucks now.

It's a broken system but it's been broken for a long time.

Of course, I'm not arguing that jobs are less plentiful now, they are but I suspect that things will correct in time. Alfred seems to share my beliefs here, or at least questions if this downturn will remain.

AI is great for taking what's already in the corpus of existing code, mashing it together, and spitting it back. The result is that it can save a lot of time for a seasoned programmer who can prompt the AI effectively and then take the results, debug it, and then adapt it to their needs .

Someone with no programming experience can't do that. I'm also skeptical of AI's ability to generate new and innovative solutions.

A friend of mine recently described relying on AI for code is like relying on a drunk grad students. That got a chuckle but there's some truth there. Hiring a drunk grad student as an engineer and blindly relying on them is probably not a good idea but having a talented drunk grad student on staff can, if used correctly make your company or team more productive.

On the industry side, another point that my daughter brought up is that companies still want mid level and upper level software engineers. If they're not hiring any entry level people, how will those mid level and upper level engineers ever come to be? Colleges have enough challenges making entry level engineers.

My take is that at some point, tech companies will realize that they still need entry level software engineers. They might need fewer or they might need them in different capacities, but they'll still be needed. Meanwhile, they'll refine how they use AI - tools to assist their software engineers, tools to allow their non-programmers to get some programming done in non critical areas, and to support other business efforts.

Sure, it won't be like the days when every two bit code school could place a "grad" with 3 months of Ruby on Rails experience into a super high paying job but kids with a strong background who are passionate about the field will still be in demand.

On the teaching side, sure, there will be changes. The NYT article talks about changes CMU has experimented with. They allowed AI in intro courses but then realized that students didn't really understand or learn what they were supposed to learn.

This is going to be a tricky thing to adapt to but nothing new. It's not really different than a student in a math class going to a teacher or professor for help, having the teacher go through a problem or proof and the student honestly thinking they understood things. Later, when they got home or on the exam, they realized they could follow the teacher but they didn't really understand the solution.

Now, some instructors understand this gap and work to teach students at a deeper level, some don't. AI is the same but we don't have self-aware AIs that can make the decision to try to do better.

It's also like when calculators came in. Teachers adapted, things were lost, but also things were gained.

It's going to take probably a few years but if they're allowed to, teachers and professors will figure it out. I saw a video the other day from a writing professor. He was talking about the process he's been going through to adapt to AI. The same will happen in CS.

Well, not universally. Some places will stick to the old ways. I mean, many places or at least CS profs still do things like they did 20 or 30 years ago which was bad then and is bad now. Some will follow the mainstream - what's dictated, say, in K12 by the College Board. I'm anticipating that will be mediocre at best.

Still, others will actually come up with wonderful ideas and they'll implement them.

It'll be a mixed bag but it won't be different than other innovations or other subject areas. It just seems a little scarier now because the changes have come so rapidly and from all sides - the students, the schools, and the world the students are graduating into.

Cool store that I can't support. Maybe ignorance, maybe antisemitism.

Sat, 07 Jun 2025 13:06:59 -0400

Warning: This post is political and involves my views on antisemitism and the Gaza situation.

So, in my travels over the past year or so, I've been trying to hit independent book stores or otherwise cool local places. I mean Devorah always wants to go to the local yarn store to get something either spun or dyed in the area but I don't have a similar hobby. I figured that I could visit a local book store. In San Francisco last October I got, among other books "Taking Greenwich Village" which was terrific and this past year in Savannah I picked up a trio of books including "Lovers in Auschwitz" which I haven't gotten to yet. In Philly, I picked up Stephen Ambrose's "Nothing Like it in the World" on the building of the transcontinental railroad.

Well, during my travels, I discovered a really cool store but had to leave almost immediately due to what I consider antisemitic posters on the wall. I'm not at this time going to share the location or name of the store since I plan to contact them via email and hopefully be able to have a dialog, I mean, there's so much antisemitic messaging out there now and most of the people proudly sharing the messaging are, or at least I'd like to believe just ignorant of the facts and realities. I probably should have tried to open the dialog when I was there but at the time it was raining, Devorah had already exited the store and I didn't want to leave her in the rain.

Hopefully, I can reach these people and they'll see the harm they're doing because it's a really cool store - they've do tech workshops like hosting a website or working a bluetooth/wifi scanner. Have electronics supplies, tech zines (notably, Julia Evan's zines which are terrific) and lots of cool maker and off grid type stuff.

Unfortunately, many people who call themselves liberals and claim to be for the persecuted are super comfortable promoting antisemitism, they just don't see it. I have friends on Facebook who readily share posts with Hamas propoganda. Stuff like that falsehood about 14,000 kids will starve by next week, things like the fraudulent Hamas published casualty numbers, or remember when Israel was accused of bombing civilian sites only for really small corrections being published days later about Hamas tunnels and facilities under or right next to the sites in question. There are horrible tragedies happening in this war but the whole situation is a lot more nuanced than propoganda releases like "Israel's killing children."

Now, let me be clear, I have no love for the current Israeli government. Netanyahu is a criminal - he's Israel's trump and it's absolutely fine to criticized him and his government's actions but when you blindly post Hamas propoganda and at the same time ignore Arab on Arab atrocities not to mention little things like the constant barrage of rockets into Israel or the fact that Oct 7 was a Hamas attack and there are still hostages being held, all you're doing is promoting antisemitism which is becoming a real issue. We seem to be reading about attacks on Jews weekly and you all are fine with it and that's a problem.

Now, in the particular instance of this store. There were a couple of signs supporting Palestine. One was "Ceasefire now" and the other, "From the River to the Sea." Nothing about the hostages, nothing about rockets being fired into Israel, nothing about the Arab on Arab atrocities in places like Yemen. Just the antisemitic soundbytes of Hamas.

Now, I'm all for a ceasefire, but given that the sign seems to be made by the usual suspects, it's coming from people who are demanding that Israel cease defending itself. No mention on releasing the hostages, no mention of stopping the rockets, in fact, the ceasefire now camp is also the intifada everywhere camp and the Oct 7 everywhere camp so indeed when one sees that same "ceasefire now" messaging, it's almost always antisemitic and it's code for "the Jews have to stop defending themselves so we can slaughter them." If you really want a ceasefire you should be upfront with demanding that Hamas disarm and the hostages be released.

The other sign is even more problematic. "From the River to the Sea" is short in this case for "From the River to the Sea, Palestine will be free." If you question the people sharing this message it means that Palestine will exist from the Jordan to the Mediterranean Sea. Note that that's where Israel is now.

They're chanting for the destruction of Israel. They claim it's "their land" ignoring that over 50% of the Jews in Israel are indigenous to the region and the rest are genetically connected, their ancestors having gotten kicked out during the diaspora. Supporting "From the River to the Sea" means you don't support a two state solution and you're calling for the destruction of Israel and guess what? That's antisemitic. You're saying that Jews do not have the right to self determination in their ancestral homeland. It's antisemitic because you want to replace Israel with yet another Arab/Muslim state while denying one specific people the right to have their own state. This is why I keep coming back to the truth that Antizionism is Antisemitism.

I could go on and debunk other falsehoods but I don't want to go on for too long.

Israeli's are colonizers - no, they're not. There is only one nation in the world that speaks Hebrew and it's in the geographic location that Jews came from. That's not colonization. How many Islamic countries are there or English speaking countries? You can't be colonizers without any colonies.
Genocide - There's no genocide. Right now there's an ugly war going on but the populations in Gaza and the West bank continue to rise as does the Arab Israeli population. How can you have a genocide with rising populations?
Apartheid - Again, I call BS. All Israeli's pulled out of Gaza two decades ago. Hamas could have built a paradise instead they built tunnels and weapons of war. In Israel Arab Israeli's have full rights and representation in government so no apartheid there. Is there racism? Maybe but that's a different conversation and American's certainly don't hold the high ground there. West Bank? Well, that gets complicated and very nuanced so we can debate that one.
The Nakba? That's another one I call into question. Funny how those who keep talking about Arabs being expelled never talk about the hundreds of thousands of Jews expelled at that time from Arab countries or the 50,000 or so Arabs that remained in Israel who have now grown to become over 2 million Israeli citizens with full rights so were clearly not expelled.

Back to "From the River to the Sea" some smartass always says "well, Israeli's used that slogan." Okay, so they did, and it was a scummy thing back then too.

To close off, posting things like "From the River to the Sea" is at the end of the day calling for the destruction of Israel, it's antisemitic and should never be accepted.

Call out Netanyahu. Call out specific Israeli policies and actions you don't like but if you don't at the same time call out Hamas, call out Arab on Arab atrocities like in Yemen then you're just being antisemitic.

I hope the store in question thinks about what they're promoting and removes those signs and then I'll be able to shop there and otherwise support them.

CUNY Pitchfest 2025

Thu, 29 May 2025 12:07:53 -0400

Although retired, I'm trying to keep some skin in the game. An opportunity to do that came up back in mid march when my friend Aankit Patel reached out to me. I've known Aankit since he ran CSforAll for the NYC Department of Education but now he's the University Dean for Tech and Computer and Info Sciences at CUNY. Aankit told me he had some extra funds to run an event and asked if I wanted to organize and run a hackathon for CUNY students.

I said no.

Too many moving parts and not enough time to put one together from mid March to end of May. I also figured that students wouldn't have time or interest so close to or right after finals.

Instead, I pitched, well, Pitchfest.

I took a page from Cristina Dolan's book. Cristina created a College and High School tech event "Dream it. Code it. Win it." that ran for a couple of years around a decade ago. The idea was that student teams submitted programs that they had already written to the event. At DiCiWi, winners were announced ahead of time and the event was an awards ceremony where there was a guest speaker, some project videos were shown, awards were given and a reception followed.

Pitchfest borrowed the "enter a project you already wrote" part but the event itself was very different.

On the surface, teams and individuals submitted projects that they had already completed - from a hackathon, a course like a capstone, or just a personal project. At Pitchfest, accepted groups were set up at different tables and over a series of rounds, showed off their work to tech professionals who then provided feedback.

To be honest, the actual pitches and project specific feedback was secondary. The real intent of the evening was to get the CUNY students talking with both a bunch of tech professionals as well as each other. To, yes, get feedback, but also inspiration and general guidance and to build their network. As one of our tech volunteers said, the goal wasn't really to improve and build the projects but rather to improve and build the participants.

Of course, in order to make this all happen, I had to come up with sufficient tech professionals so that each project would have at least one pro with them for each round. Since we had 45 participants split into 18 teams, it meant I needed at least 18 tech pros circulating.

Fortunately, and this is something I'm always appreciative of, my former students - my StuyCS family and my Hunter Daedalus family came through as usual and big time. We had around 25 volunteers plus a bunch more who said they could come if needed but had tight schedules.

We had a great range - recent grads to people with over ten and even over 20 years of experience. Software engineers, product managers, entrepreneurs, investors, people at small startups and people at tech giants. It was a terrific mix.

It's also worth noting that there was an extra benefit to having both my Hunter College students as some of the volunteers as well as a few of my StuyCS volunteers who had also attended a CUNY college - those students were all living proof that you can do amazingly well coming from New York's public schools and a few of my volunteers noted that they felt that their being CUNY products made an even bigger positive impact on the participants.

The evening went great - we had four rounds at the tables followed by three short demos and then open mixing and networking.

I'm just starting to get feedback from the professionals and the participants but initial reports are extremely positive. The participants got a lot out of the event and made new connections. One of my former students just sent me an email to tell me that they invited a couple of the participating teams to their offices to meet with their engineering teams.

Overall I think the event was a big win.

The thing that's sticking with me now the next day is my former student's comment about building the participants not the projects. This event was promoted as being about the projects but it was really about networking, building relationships, interacting with people in the tech community and in fact learning that you belong in that community. It's something I've long focused on when I set up programs and events.

We ran this one because Aankit had some extra funds. To do it again and moving forward we're going to have to find some company willing to sponsor. Hopefully that won't be too big of a challenge or ask.

Better teachers - CS teachers without CS or Math teachers with math

Fri, 02 May 2025 15:12:41 -0400

This article made its rounds this past week.

It cites a couple of papers claiming that CS teachers are delivering better instruction and might be better qualified than teachers in other subject areas. This is in spite of the fact that CS teachers generally have a generally weak background in CS as well as in CS related pedagogy compared to teachers of other more well established subjects.

The papers forming the basis for the article were written by Paul Bruno a professor of Education Policy, Organization and Leadership

From the article, a key measure in determining the results in the papers related to the number of students taking AP exams. Those results indicated that "When computer science instructors had an additional year of experience teaching that subject, their students were more likely to take the advanced placement exam, and they also scored slightly higher."

After making that statement, the article quotes from Bruno:

“Even when veteran teachers lack formal experience with or specialized training in computer science, they may have other attributes that make them effective instructors,” Bruno said. “Staffing these courses with veteran teachers may be a reasonable way for schools to expand their computer science curricula, when recruiting faculty who have specialized training in this subject is challenging.”

Unfortunately, I don't have access to Bruno's papers but I find this last quote troublesome and take issue with the article. I do want to emphasize however that I'm just talking about the article on the papers, I can't comment on what the papers actually did or didn't say.

The thing is comparing high school CS, and in particular AP scores is an apples and oranges comparison. You just can't meaningfully do it.

Are we talking about APCS Principles or APCS-A? APCS-A is a college level course. APCS-P is a high school course that can be handled by a typical 10th grader.

I doubt a teacher without a decent CS background is doing a great job in APCS-A but regardless, I know students who've passed the APCS-A exam by using canned curricula - something that, to my knowledge isn't done in other subjects like Calculus, AP-Bio or APUSH.

If APCS-P is the class in question, then it's really apples and oranges. AP-Bio - college level - you take it after Bio and Chemistry. AP-Calculus - college level - you take it after you've taken all the other high school math classes.

APCS-P - no prerequisites and as I mentioned in my last post, Over my 8 years at Hunter College, I've seen far too many students who've gotten a 5 in APCS-P who really didn't learn anything (however, all these students were very bright and hard working and generally did very well in my CS0 class and then beyond).

I'm pretty sure you can't get a 3, let alone a 5 in AP Calculus without actually knowing any Calculus.

So how can you compare? A teacher can't do well teaching Calculus unless they know Calculus but APCS-P is set up so that a student can indeed get a passing exam grade without really learning anything.

I'm sure that last sentence will ruffle some feathers but that's what I've seen looking over 6 cohorts of around 25 students each year. I'll go further and say that while i didn't track the numbers, there were numerous applicants to my program with passing APCS-AP scores who struggled with classes like Geometry.

You can't compare an APCS-P score with any other AP score.

APCS-P is also promoted as an AP you can take without any prerequisites and there's been a HUGE CS push over the last decade. Compare that to AP-Bio where only students interested in pursuing Bio typically take the class. AP Calculus is more regularly taken but it's seen as "the next" math class so it's also different - can't take that until you finish all the other math classes.

I'd love to see Bruno's actual work - he may very well account for all of these things but K12 computer science is already plagued with too much "you don't need to know CS to teach CS" so while withholding judgment on the source material I'm giving the article two thumbs down.

2024 Calc - over 550,000 APush 250000 AP Euro 83000 Bio 260,000 APCSA 98,136

APCSP 175,261

How many roads for intro CS

Tue, 22 Apr 2025 17:33:21 -0400

How many different intro CS courses should we have? A question that resurfaces periodically. This time due to an article in which C++ creator Bjarne Stroustrup posits that different audiences would benefit from a different intro treatment for CS. Perennial CS Blogger and friend Alfred Thompson shared his thoughts on the subject here.

The thought is that someone who's going into academic CS needs one thing, a software engineer another, a non CS person who could benefit from some programming within their field something else.

A person looking towards a PhD in CS might need more theory than say a Software Engineer, although, I agree with Alfred in that Software Engineers benefit from more in terms of data structures and algorithms than say what a coding bootcamp provides.

On the other side, a non CS major could benefit from a more end user programming course that leans heavily on libraries along with a data analysis or statistical bent.

Seems to make sense.

Stroustrup was talking about college offerings but Alfred, in his post brings it to the high school level and notes that while a college may or may not have the resources to offer multiple intro CS classes, high schools likely won't. He also talks about how APCS Principles could or maybe even should be the single intro course.

I agree with Alfred - high school's don't have the resources to offer multiple pathways and I'll go further and say that in High School, I think a single intro course, which should be required is preferable so long as its well designed and well taught.

In college a student might already know their desired path so should be able to chose a CS class that matches that path. There should also be a more general CS0 for kids who haven't decided yet.

High School isn't about picking a major, it's about, among other things, creating a well educated populace. As such, students should be introduced to a variety of fields and CS should be one of them. Now, some readers will be adamantly against this and say that CS shouldn't be required. Well, as long as they think math, biology, history, English and the rest shouldn't be required then I guess that's a fair position. Otherwise, why is CS a less important field?

Now, of course, the quality of the course becomes important. It shouldn't be a deep dive programming course that only a computer scientist would love, that is APCS-A or similar. It should have a programming component and should introduce students into various CS fields and how a computer scientist thinks. It should also show how CS interacts and is useful in other subject areas and how CS and technology are important in society. All that good stuff.

Can this be done? Sure - Stuy's one semester intro course does a pretty good job at it and their full year version does an excellent one.

I don't think, however, that APCS Principles should be the answer. It likely will be the answer because it has the College Board and all the other big players behind it and as the saying goes "nobody ever got fired for hiring IBM" but it shouldn't be.

The course, whatever it is also has to be taught by a very knowledgeable teacher and frequently this isn't the case.

Now, why don't I think it should be APCS Principles, let me tell you what I've observed over the past decade.

What I've observed can be attributed to the course, the teacher, or maybe both.

I ran an honors CS program at Hunter College and every year I admitted a new cohort. From my third year on I accepted about 50 kids per cohort. For each incoming class, I placed them based on their APCS experience. If a kid took APCS-A, regardless of class grade or AP score, I placed them in to my CS1 class - similar material but in C++ and they all did great. The rest went in to my CS0. The majority of the kids who went to CS0 had taken APCS Principles in high school and almost all had gotten a 5 on the exam.

The truth of the matter is that the majority of my students who took and passed APCS-P would have been better off with nothing. A small number each year were decently prepared, either from their APCS-P class or elsewhere - they had some fundamental programming experience, understood some of the "big issues" and had some general exposure to CS. The majority of those kids who had APCS-P, however didn't and in some cases had acquired bad habits and faulty knowledge.

Was this the fault of the course or the teacher? Maybe a little of both. I actually don't want to blame the teacher per se here - they're frequently placed in a rough situation - they don't know the subject and are trying to do right by their students, but their preparation - PD from the NYC DOE, Code.org, and others has been woefully insufficient.

Now, a well prepared, knowledgeable CS teacher can turn APCS-P into a good course - that was evident when I saw the few well prepared kids - each year coming from the same schools with the same teachers. The Stuy intro course could in fact be easily adapted to be a superset of APCS-P. On the other hand, APCS-P can be a wasted year in spite of a kid "earning" a 5.

So what are we left with? While we've come a long way in terms of CS offerings there's still a long way to go. We're currently not doing a good job across our states to find and prepare the number of qualified CS teachers that we need. We're doing a better job with offerings but offerings without great teachers isn't the answer.

We need to keep pushing on promoting the good rigorous teacher preparation programs out there and call out the ones that are fluff. We've got to keep pushing for a required CS class in the high school. It should be a single general course for everyone but, finally getting back to the original issue, it should prepare students so that they can take a follow up CS class appropriate to their interests when in college.

Vibe coding doesn't jibe

Mon, 07 Apr 2025 15:09:29 -0400

The truth is, I don't use those modern AI tools much. I'm not doing much coding and when I do, I've been using Clojure or Elisp - not so mainstream so as to have all sorts of AI support. For writing? I prefer do to it myself. Anything else? Well, to be honest, there's not too much going on in my life where I feel like AI will improve things. Maybe if I were writing important business emails or I was cranking out tons of boilerplate code for a company but as a retiree just working on personal projects, I just don't see the need.

In any event, this newfangled Vibe Coding thing caught my attention.

The idea is that you have AI generate a programmed application from repeated prompts. I'm not going to say you're programming or coding via AI because you're not but more on that in a bit.

You prompt an AI code generator to write something. If it works, great, if not, restate the prompt or give the next prompt to address the issue. Whatever the AI comes up with, accept it. If the solution works, you're done, if not, rinse and repeat.

Apparently the Vibe Coding label and hype got a big push when Andrej Karpathy tweeted this:

There’s a new kind of coding I call “vibe coding”, where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It’s possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good.

Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like “decrease the padding on the sidebar by half” because I’m too lazy to find it. I “Accept All” always, I don’t read the diffs anymore.

When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I’d have to really read through it for a while.

Sometimes the LLMs can’t fix a bug so I just work around it or ask for random changes until it goes away. It’s not too bad for throwaway weekend projects, but still quite amusing.

I’m building a project or webapp, but it’s not really coding – I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.

Now, in the past couple of weeks, I've seen a number of examples of vibe coding, testimonials about it and, to me the one I really want to address, people floating that it's something we should be teaching, at times, as a valid coding technique.

Reading the start of Karpathy's tweet, it sounds like he's pretty excited but reading the end we get "It's not too bad for throwaway weekend projects" and "it's not real coding." It appears some people are forgetting those parts.

The first time I saw vibe coding - prior to my knowing the term was when watching a YouTube video. The author was showing how to do something, I forget what and then said he coded up a script to help along the way. Then he said he didn't actually code it but used AI. He then did give the disclaimer that he really didn't know how the script worked and would have been reluctant to use it if the task at hand was actually a really important one.

That raised a big point - with Vibe Coding, the human prompting the AI might have no understanding of the solution. This means they won't be able to customize it and more importantly, won't be able to audit the code to make sure something bad isn't happening.

Do we want a vibe coded banking application? How about a vibe coding insurance claims approval system. Don't even think about a vibe coded surgical assistant. How about military control stuff? I mean the WOPR was bad enough with Falken writing the code. Imagine if it were Vibe Coded.

This means that while vibe coding can be useful for a quick prototype or proof of concept, it's going to be extremely dangerous to actually rely on Vibe Coded solutions.

Of course, this is America where common sense and prudence always takes a back seat to the dollar so I wouldn't be surprised to see attempts at Vibe Coded "solutions." Particularly if it can be set up so that the customer and common man gets screwed by any issues and not the business that rolled out the problematic code.

The other thing I keep seeing with Vibe Code is that the examples people show are variations on well known themes - an ecommerce site, Tetris, or something else familiar with tons of pre-coded solutions out there. This makes sense given the way AI works. It can do a reasonable job at generating variations on a theme - remix the existing stuff. Let's see about Vibe coding something brand new.

In any event, for prototyping I could see a big win but I'm pretty sure Vibe coding will be used in lots of places where it shouldn't. For now, I'm placing Vibe Coding in the same category as other AI coding assists - helpful and maybe even extremely helpful for real developers. Also helpful for non-coders at the prototyping or toy level but I'm not seeing the end of the human software developer in the near future.

But now let's look at the education side.

I've already heard the praises for Vibe Coding on the education side - you can create without knowing a line of code (or even that it is coding) and that it's empowering,

You can create without knowing a line of code, I guess, but create in a way that reminded me of something I think Mitch Resnick said at this past SIGCSE - "Do you want your kid to learn to play Guitar Hero or do you want them to learn to play guitar?" If they Vibe code a solution what are they really learning? Sure, there's a lot of mundane, nitty gritty code stuff that's annoying to teach and learn about coding but that comes along with things like problem solving, dealing with complexity, and yes, even dealing with the details.

Vibe coding, like other AI techniques can be tremendous time savers for people who already know but if they're used too much like a crutch before people know, then they'll never learn.

Look at math. Sure, we use calculators but does that mean there's no value to learning basic numeracy, number facts and operations? Sure, we don't have to do them super fast, nor memorize all the times tables but there was absolute value to learning the basics and only then going to the calculator.

We now have all sorts of mathematical problem solving tools. Does that mean there's no value in learning algebra? While there are many who would say that there are a lot of specifics in an algebra class that maybe isn't so important, anyone being honest with themselves will agree that there's also a lot of fundamental problem solving that's learned through taking algebra.

If someone just learns to Vibe Code, it's just like hiring a bad software engineer - telling them what to build, and when they come back with an error, not being able to help fix it but basically guessing at some other prompt.

So in my opinion, you're not really creating with Vibe Coding - you're asking something else to create for you and it's not empowering for a student because the student isn't growing as a result of using the tool.

Could Vibe Coding be used successfully in an educational environment? Sure. Particularly when it comes to prototyping but that's for later.

For those intro levels, let's not teach Guitar Hero, let's teach guitar.

On Slide Decks

Sat, 29 Mar 2025 11:48:52 -0400

Based on some of the feedback I got on my last post, I thought I'd dive a little deeper on using slide decks.

I had some people tell me that they too found that slide decks didn't work for them while others shared that they use them regularly. Still others shared how they specifically used them.

Now, as I said, they don't work for me. Sorry if I wasn't clear. I didn't mean that they never worked or didn't work for anybody. The important thing is that slides or complete slide decks are tools and can be used for good or for ill. What's important is that a teacher is intentional about the practices they use and use the right tool in the right way at the right time.

There are ways that slides can be horrible. If they accompany a one directional old school lecture with no active learning going on. If they convey enough information so that students zone out because they can just get the slides later but then realize that the slides aren't nearly enough on their own or when slides drive the direction and pace of the lesson as opposed to the actual class situation.

On the other hand slides can be effective - they can be a time saver if you want to bring up say, a definition or long quote as an example and indeed I've seen some slide decks that form the basis of excellent reference materials. They can also provide a focal point for both students and instructors.

Of course just about any practice can be good or bad. A few years ago before the pendulum started swinging back, discovery learning was all the rage, of course everyone ignored the difficulties in setting up a strong discovery learning experience and just told teachers they should do all discovery all the time. To them, since I was tenured and in general have an irreverent streak when it comes to superiors I'd say things like "well, in the history of mankind only two people discovery-learned calculus" or something similar.

Well, as a practice, doing a discovery lesson where the experience isn't sufficiently set up and scaffolded and the students not sufficiently prepared can be disastrous. In fact, I'd go as far as to say that a bad lecture can be better than a bad discovery based lesson.

Now to get back to slides, I should probably share where I'm coming from. I started teaching in the early 90s. The only slides you could use were old school overhead projector slides. You also probably had to hand write them. I guess professors at the college level could print them but high schools just didn't have those fancy resources. I had a couple of math colleagues who made heavy use of slides but more by using the overhead as a marker based board and wrote on the pages live.

When I started teaching CS, I was in a traditional classroom with a chalkboard. We only went in to the computer lab when the kids were programming. I could deliver instruction in the computer lab - it even had a VGA overhead projector that wasn't all that terrible but teaching in the computer lab was less than ideal - there was no room on the desks for students to work between the large computers, big CRT monitors and keyboards and the sight lines were awful. When we were dumped into the computer labs to teach full time, at least we were allowed to rearrange the room to make it better and flatscreen monitors have made a huge difference for desk space.

Looking back, our greatest tech was that graph maker thing. For you youngsters, it was like a projector screen that you pulled down in front of a chalkboard. It had a grid of holes in it. You'd go over the whole surface with a dusty eraser which would let the chalk dust through the holes. Pull up the screen and you ended up with a perforated grid on the board - instant graph. Pretty cool tech.

So, I came up as a teacher without slides.

Now, younger teachers have shared with me what they like about slides.

A recurring theme was that it meant that there was something to give a student who missed class. This can be a benefit, particularly if the slides are enough to catch the student up. I'm guessing that usually, they can help but hopefully they aren't enough. Why hopefully? Well, if a student can get the full value of the lesson by merely reading through the slides then how much value is the teacher bringing to the class. I think more often the slides can provide some information but that absent student will still have to do some extra to catch up.

So what did I do without slides? Well, back in the day, students took notes so I paired the absent student with a present one who had good notes. Often at Stuy the absent kid would take it upon them self to do this with or for a friend.

I also know that instructors often share slides with other instructors. This seems bigger at the college level. Well, we used to share lesson plans. It wasn't uncommon for a teacher to have a book of lesson plans for a semester or a folder or something similar. The first time I taught linear algebra, I borrowed the lesson plans from a colleague. These were likely far better than slides unless the slides had copious presenter notes.

I've also had people tell me that they use slides to present prompts, key questions, and things like code snippets.

Using them this way sounds great to me. I don't do that with slides but I'd have code snippets ready in an Emacs buffer and key questions written in my lesson plans. Sounds like a good use of slides but it's also something that can be accomplished without slides.

When you cut away the chaff, most of the people who told me they liked slides basically seem to be using them as visible lesson plans with time saver materials. Sounds good to me, but personally I think I'm more effective without them. At times, I like the idea of instantly bringing up a quote or a definition or a prompt but I like the pacing I get sometimes when writing it out or when I have students actually copy something down. It's a different type of engagement.

Besides, for me the slides would be too constraining. I've frequently likened teaching to rap or stand up comedy. You've got your core material (the lesson or the main lyrics or routine), you're catch phrases or go to lines (I don't get no respect, or "put your hands in the air like you just don't care"), and then the improv stuff. For me, if I had detailed slides, I know that it would keep me from the improv stuff which includes riffing off discussion and yes, at times tangents. I think my lessons would lose a lot of whatever richness would normally be there.

That's just me.

That brings up another issue in terms of "best practices." Too often the "experts" and "thought leaders" who force "the research" upon us hit us with a one size fits all. When I started, everyone had to use discovery learning (not to be confused with group work or other similar practices). A while later, discovery became the rage. More recently, although not being touted as the be all and end all, direct instruction has been making a comeback, albeit with active learning. If you didn't teach using the chosen method, you were "ineffective" regardless of how good you actually were.

The truth is, there are multiple things that go into a practice being effective. Obviously the students in a particular class. The specific topic also comes into play. All sorts of other school factors also have to be considered - class size, composition, even time of day and what else is going on within the kids schedules.

On top of all of those though is the one factor that is almost always left out - the teacher. Most teachers have their own strengths and weaknesses and also our preferences.

If an effective teacher leans towards direct instruction with slides, then all things being equal, that will likely be one of their more effective practices more often than not. Sure, there will be times to mix in other practices and there might even be classes where using a whole other practice would be better for the majority of lessons but the "all things equal" default probably should lean towards a teachers strength, providing that that strength can be a good practice.

For example, I'm not a cooperative learning guy. Sure, I've done it - I've set up the roles, prepared the materials, and employed the practice but it was never my strength and at the end of the day, other than as a change of pace, my students didn't enjoy my cooperative learning lessons as much as my other lessons.

Contrast that to one of my mentor teachers who was all cooperative all the time. Cooperative learning went over way better for him but if he tried to teach like me, it didn't take.

At the end of the day we were both well respected, well liked, and got good results but ended up leaning on different practices to get there.

Closing up, I'll also give one more reason why I don't do slide decks and that's pure laziness. Even when I teach a class over year after year, each class is a little different so I rethink my lesson plans and make new notes. Sure, the lessons are largely similar so the lesson plans are mostly the same but some details always change.

If I used slides, I'd have to revise them or remake them every time through and I'm not nearly organized enough to do this. I'm far too lazy.

So, that's it - slide decks - not for me but for others they can be effective when used well.

Sorry if my last post implied otherwise

The Daily Aim

Thu, 27 Mar 2025 15:11:51 -0400

The other day I read Kristin Stephens-Martinez's latest blog post talking about how she has started to insert a daily plan slide at the front of her class slide decks.

Stephens-Martinez, you might know, hosts the CS-Ed Podcast which I very much enjoy even if the episode lengths never quite go long enough for my morning run which is my primary podcast listening time.

Here, the blog post talked about what Stephens-Martinez described as the "teaching practice" of creating a slide and putting it at the front of the daily class slide deck. The slide would describe via a short bullet list, the plan for the day so that students would have a road map for the lesson.

Not a bad idea but reading it I started thinking about a few things - one, the difference between teaching and teacher preparation in K12 vs college and also when a practice is considered a best practice and when one should implement any practice.

For starters, the idea of using a set opening slide. It reminded me how often I hear college instructors talk about slides - can you share your class slides with me? Does the text come with slides? And so on. Some K12 teachers use slides and maybe in the current day of smartboards more teachers use them but I rarely if ever use slides and neither do most High School teachers I know. They're too restrictive. I might have some set text or some code or a diagram here or there to project so I will use occasional individual slides or similar projected content but I've never had or used a slide deck for a lesson.

Slide decks constrain things too much. For the later slides to work, I have to follow the narrow instructional path of the earlier slides and, in an active lesson, hope that the class follows the same path at the appropriate pace.

Much easier to adjust to the needs of a particular class on a particular day without a set deck.

True, slides can save time and they have their place, but sometimes typing on a screen, drawing with a paint tool or doing either directly on a white board make more sense. So too at times does having students transcribe as opposed grabbing the slides from the website later. Doing things in these varied ways engage students differently and can be utilized to tune the cadence and tempo of a class - something that educators should probably think about more than most do.

The slides/no slides difference probably comes from the history of how things were taught. Although I'm I'm heartened to see more and more professors, including Stephens-Martinez who really care about and work on their teaching, anecdotally at least , college instruction grew out of the rote lecture. While I'm sure old school HS teaching was and sometimes is like this, K12 education as a whole is also influenced a lot by the earlier grades and how things were taught there with more interactive and active practices which, in my lifetime have become relatively mainstream in HS. So, coming out of a tradition of lecture, slides make sense. Coming out of training that also includes K-8, not nearly as much. Of course this is just conjecture.

Now, to the actual practice. It struck me as something that, in general is a good idea and it also reminded me of the age old HS practice of the Aim. That is, either before class, or within the first few minutes of class, a teacher is supposed to have on the board, or possibly elicit from the class, a sentence that describes what the class period would be spent on. Typically top front and center on a HS blackboard (do they even still have those?) you'd see the word Aim: followed by that key sentence.

Very similar idea.

I was happy to see the post sharing the practice but was also a bit sad that this fundamental technique, something maybe taught to every K12 teaching in America and probably something most American educated adults saw during their high school days isn't common knowledge for college instructors. Of course, that's nothing new. Every SIGCSE I've been do has been chock-full of "new" practices that a college person came up with that I had seen in K12 years if not decades earlier.

So, the ideas are similar but subtly different. The lead slide can contain more information and as shown in the blog post, a checklist that can be updated and the slide reused. This is good. On the other hand, the slide only stays up for a short time unless explicitly returned to. On the other hand the Aim stays up at the top of the board for the entire lesson. Less information but an anchor point that survives through the entire class.

Now, I said that the Aim or in this case, slide, is generally a good idea but I'd argue not always, and not just because I could be sloppy as a teacher and would sometimes forget the aim. In fact, one time, my principal wrote it up as one of the things I could improve on. The thing is, I'm a believer that there are no best practices, just practices that can be good, bad, or even great when used in the write time in the right way with the right population. Having an Aim or intro slide is easy and you'll generally get good bang for you buck.

Sometimes, though, you don't want to give away the punchline and telling the students the plan can do just that. As a simple example, let's say you're about to teach binary search. Typically I'd want to lead the students towards the algorithm, first by having them "discover" the possible limitations and advantages of linear search and then "discover" the binary search as a solution. In this case, an Aim or Slide that describes the why and what of the lesson would spoil the lesson. Sure, you can still have a meaningful Aim but not something that gives away the plan.

There are lots of lessons where this is the case.

I'd like to think that the teachers that have come through the program I built at Hunter get this - they know all these practices and techniques form a belt full of tools to be used as appropriate. On the other hand, I've seen far too many K12 teachers who weren't given the tool belt and tools but were rather taught "Here's how to do it" and they would always have an Aim or teach via a particular practice no matter what. Of course, too few college instructors get any form of preparation in terms of teaching so they're left in an even less desirable place.

So, that's my take.

I applaud Kristin Stephens-Martinez for helping to share better teaching practices at the college level through both her blog and the CS-ed podcast. In this day and age with fewer and fewer of us CS Educators blogging it's read and hear more voices sharing good stuff.

CS For All or All for CS?

Mon, 24 Mar 2025 07:34:31 -0400

Is it CS for All or is it All for CS?

CS for All, to me has always meant exposing all students to CS for two reasons. First, so that every student gains some understanding of CS and also so that some of those students might be inspired to go on and study more CS. For that basic understanding, that should include a few things - the way a CS person thinks and problem solves, what a CS person does or might do professionally, and also specific practical skills that students can employ regardless of what they pursue in the future - that is programming.

Now, this isn't revolutionary, it's the same in other subjects. All students in the USA study Biology. If taught well, they learn a bit about what a biologist does, they learn the basics - how biology affects their lives, and perhaps some are inspired to go on to be doctors and scientists.

Of course, subjects like History and English are a little different in emphasis but you get the idea.

CS isn't special but a well educated student should have this exposure to lots of fields and CS is one of them.

Mark Guzdial wrote a great post relating to this, specifically that the emphasis should not be in the "Sense of Belonging in CS" but rather on using computing to support whatever students pursue later on. I won't rehash Mark's post as he already stated his case more eloquently than I could so just click on through and read it. Shortly after Mark posted, Alfred Thompson added his two cents. That's also worth a read.

I thought I'd add a little myself. I won't rehash what they wrote but just add another wrinkle for thought. I will say I'm 100% in agreement with Mark's position.

I had an epiphany on this around 1998 when I realized that I should be working on more than just the experiences for the top Stuy CS kids. I've always referred to it as the 80% / 20% problem. The percentages are really all wrong, the 80% should be much higher and 20% much lower (though at Stuy, historically, the 80% was somewhat lower and the 20% higher).

The idea that the majority of your students in a required CS class (the 80% or more) aren't going to go into a tech field and may or may not even take a more advanced CS class. The rest will take more classes but still may or may not go into the field. The Stuy numbers skew differently in part due to the fact that Stuy has always been considered a STEM school in spite of strong humanities and arts so the population probably already skews towards taking more CS.

Anyway, the wrinkle I'll add is the teacher. I'm a strong believer in having a separate required CS class in High School to address the 80/20 problem - the one required course for EVERYBODY and I think what we designed at Stuy has been very successful both in terms of what I've heard back from those that went into tech as well as those that went in other directions. In earlier grades, integrating CS or tech concepts could make more sense. Now, I'm not against more subject integration - that would be a great thing, I just don't see it as realistic given the current high school model.

Given that model, teacher background and preference can have a great impact on what is taught. Sure, there might be a set curriculum but teacher impact is so much more important.

I've met teachers who approach their class with the assumption that all their students are or will be subject majors. I was once speaking with a young, charismatic, popular, and effective chemistry teacher. I asked - how do you deal with the fact that the majority of your students won't do any more chemistry after your class? He said he didn't. His approach was that they were all going to be chemists. I've seen the same in other subjects.

I have a friend who's been a high school math teacher for years. Super smart but was only effective with the high end math kids. We joked that he was the only teacher we knew teaching a graduate seminar in high school geometry. If he had the right students, it worked great. If not, well….

If a teacher has this approach, then the 80% will likely get short shrift. On the other hand, if the teacher doesn't have any passion for the subject (or doesn't have sufficient knowledge as is frequently the case right now for CS teachers in K12) then the 20% get the short end of the stick whether the subject is Math, Chemistry, CS or anything else. I've even seen this with some English teachers who view High School English more like an advanced college literature course rather than a place to yes, work on literature but also a place where students learn to write and communicate.

Whatever curricula, standards, frameworks, or guidelines the powers that be push down into the schools, the implementation on the street is going to rely on the teachers and their biases and preferences will do as much if not more to influence if we're "CS for All" or "All for CS."

Sigcse 2025 part 4 - sessions and the rest

Fri, 21 Mar 2025 11:06:57 -0400

To finish up, we'll look at some of the other sessions I attended but first a couple of notes.

One thing I again noticed was the lack of pedagogy sessions. Maybe I missed something but other than Nifty Assignments, which isn't really about pedagogy and doesn't interest me much anymore I didn't see anything. This is nothing new but at least a couple of years ago we had "It seemed a like a good idea at the time" coordinated by Jim Huggins and Dan Garcia in all their pink suited glory and Colleen Lewis's "Micopedagogy" session (I hope I'm remembering the name right). Jim and Dan's session wasn't purely about pedagogy but as it happened there was a good deal of it covered.

That's a shame given that the higher ed teaching community could probably use it.

The other was a format change. The paper sessions all had an additional paper per session which cut the time per paper down by a few minutes. I think the stated goal was to keep the number of accepted papers up and to minimize the number of parallel sessions. I'm not sure if it affected the # of parallel sessions in practice - there were fewer rooms but since moving from room to room mid paper is far from ideal I wonder if there was any benefit. In any event, I'm glad the organizing team is thinking about things like this and trying things to improve the conference.

For this run through though, I think it led to some unevenness in presentation. Some sessions I went to, the speaker basically just covered their methodology barely speaking about the paper before running out of time. Another wasn't able to get to the methodology at all. I'm sure, assuming this time format remains the same, speakers will readjust so not a big deal, just something I noticed.

I attended a couple of sessions that came back to the BOF and keynote theme I wrote about earlier - reforming and possibly simplifying the CS major. One speaker talked about how at his their institution they looked at prerequisite trees for both their CS major and CS minor and found a good deal of redundancy as well as superficial information. They found, for instance, that a course might have one or two important topics for the CS student - say dot product for instance but the majority of the class had no relevance to the later CS classes. They also found redundancy in that one needed topic might be covered both in multiple prerequisite classes as well as in other required CS classes.

I'm betting this is true for most CS programs across the country and I'm really glad that there seems to be more light being shed on it.

This was followed by a session run by Richard of Regis University who looked at similar issues by proposing "10 wicked curricular issues."

There were a few sessions on pedagogical topics but I wouldn't call them about pedagogy. A couple were on live coding. These were interesting because they were back to back and while one seemed to find that the live coding did result in improved student learning, the other did not. Just shows the complexity of education. Another session talked about algorithm classes and noted that classes and texts were taught from a mathy theoretical point of view and argued for a more experiential approach with students writing programs to implement algorithms and then make use of them. I was in agreement with the speaker and left wondering how correct he was in his premise. What's going on in algorithm classes out there? Are students just learning a catalog of Data Structures and Algorithms and proving runtimes or are instructors working to develop a problem solving sense with their students while exploring key known data structures and algorithms along the way?

I attended a bunch of other sessions but for brevity I'll only share my thoughts on one more. The paper was "Why is my code slow? Efficiency bugs in student code."

The paper and talk looked at what the speaker described as "efficiency bugs." Things like superfluous computations, suboptimal data structures or algorithms, caching or failing to do so.

The speaker used as an example, using a list to check for uniqueness of data rather than using a hash table or a set. It's actually very much in line with the mode lesson I taught at Stuy a couple of weeks ago.

It seemed like the intervention was in the form of feedback on projects but it would be nice to weave these concepts so that they come up in lessons.

So, that was my SIGCSETS 2025. Pittsburgh was a great host town - lots of food options close to the convention center and lots to explore for those of us who came early or stayed late.

Next year is in St. Louis. Even though retired I'll probably head out to that one as well.

Back in the Classroom - Week 2

Tue, 18 Mar 2025 07:55:33 -0400

So last week I completed my second and final week back at Stuy teaching CS. We spent Monday finishing up the mode lesson and then spent most of the week on this topic - writing a computer program to automatically decode something "encrypted" with a Caesar Cipher. The post I just linked to was from when I first taught the lesson. I've refined and reworked it considerable since then but the gist is the same.

This time, after running the cipher, we took a detour - I asked about what things they considered when purchasing a new laptop - they gave me all sorts of features like price, weight, screen size and more.

We then compared laptops by a single feature, say price, then two - maybe price and weight. This led to an x y graph - one axis price, the other weight. Ultimately this led to us normalizing the features (making them all within the same range) and then the distance formula which we can use in 2D, 3D or any higher dimension.

When everything was done, we were able to apply the distance formula to all the rotations of an encoded sentence and for each, compare the letter frequencies of the rotated sentence with the letter frequencies expected in an English sentence. The rotation closest to the expected frequencies was always the answer, provided that we had a long enough sentence to provide sufficient data.

We didn't get to build a completed program to do the decryption in class but we build some of the code, all of the key algorithms and did end up building the scaffolding needed - a subgoal labeled Python file.

That took us through Thursday. We probably could have completed the program if we continued on Friday but I thought that doing an AMA might be more fun and more valuable. After the weekend passed I did send along a finished program in case anyone wanted it and I encouraged them to write their own.

After my last class, I spent about an hour speaking with a few Stuy seniors and then I was off. This week, back to retirement with their regularly scheduled teacher back from grand jury duty.

I hope the kids had a good two weeks and got something out of the classes. I certainly enjoyed the time I spent with them.

Overall the in class vibe was similar to what I was used to, particularly if you allow for the fact that I was only in for two weeks so couldn't develop the same level of rapport with the students. I mean, I think I did okay - I was even gifted a mini lemon pie on my last day (3/14) which my me and my family promptly devoured later that evening. On the other hand, I was told by some Stuy faculty members that some things have changed - phone use, for instance is much more rampant with more kids glued to their devices when in the halls. In my classes, I had a few kids on devices or even on the computers doing non class stuff while in class but I didn't stress about it due to the nature of my time there. Also, it didn't necessarily mean the student wasn't engaged. I had one kid who kept playing a snake type game on the computer. Once, the person who was the official sub chastised her (I was unofficial, but that's another story). After class, I apologized to her and told her I didn't mind that she was on the computer doing something that I consider a modern day version of doodling. The reason was because she was actually always engaged - participating in discussion, completing assignments and in every respect, besides the "doodling" being an active engaged student.

I also, over the course of the two weeks got to meet a number of StuyCS seniors and had great informal conversations with a number of them. I'm going to save that though for a concluding post on my time back at Stuy.

As a final note, the CS and Math teachers were amazingly warm and welcoming - it was like I had never left, even with the people I had never met before. Likewise with the teachers I ran into from other departments.

On the other hand, one particular AP made for a much less smooth experience. Not the person in charge of the CS department - she was delightful and helpful, but rather the AP Admin. When I was first asked to do this, I said okay but the AP Admin said they couldn't hire me for a partial schedule of only 3 classes (rather than 5). That's false. Then other obstacles were thrown in the way. Apparently, he thought it better that the students receive no instruction rather than make the least effort to smooth the way. Finally, when it was okayed that I come in (on a volunteer basis), he refused to let me get a temporary ID which necessitated other people being inconvenienced daily to come and sign me in.

Now, had a decent AP Admin been at the helm, I would just have been hired part time. If something came up to make that impossible, they would have appealed to the PA or Alumni association for a modest honorarium and certainly they'd have given me the ability to come in without stealing precious time from some teacher on a non teaching period to sign me in.

So, those were my two weeks back as a teacher. It was weird to have to be at a specific place and time every day for two weeks - I had gotten used to the freedom of having no set time obligations but it was a fun two weeks.

No way I'll be going back full time, but if the StuyCS team needs me again, I'd be ready and able to pinch hit.

SIGCSETS 2025 part 3 - BOFs

Wed, 12 Mar 2025 11:02:32 -0400

Once again, there were three sets of Birds of a Feather sessions. Two at the end of the day on Thursday but before the big reception and one partly over lunch on Friday.

Last year I enjoyed two but walked out o the third mid session. This year, I enjoyed two but just missed the third because I got caught up talking to an old friend.

Both of the sessions I enjoyed suffered from the usual BOF problems - since the sessions were well attended and time was short, it was tough to get deep discussions going and also keep the entire BOF together.

While both of the sessions I attended had to deal with these issues, the organizing teams of both really did a great job within the constraints of the session.

The first session was "Teaching Computing in Prison" run by Emma Hogan, Keith O'Hara, and Leo Porter. I already knew Emma and Keith but didn't realize they were running the session so it was nice to chat with them prior to the BOF starting.

Now, I've never taught in prison. The closest was in my very first class of my very first year of teaching. I taught a remedial math class and a good portion of the students in that class were recently out of prison and some would be soon to go in. I was also a little familiar with a coding bootcamp type effort but I think that focused on people recently released and not those actually still incarcerated.

The BOF was fascinating. It's amazing the restrictions both instructors in prison as well as the students have to work under. Nothing connected to the internet, frequent interruptions (alarms), students coming and going and that's the easy stuff. In many cases, can't even run their actual code - it has to be printed out and submitted to the instructor who can then, if they chose to and have the time, they can run the code (after typing it in) and test it and give feedback.

Other interesting tidbits, though I could be misremembering some of this - neither the students nor the instructors decides who gets into the class. It's assigned by some third party (the prison administration???). The prisoners in the program are all part of the same major and it isn't CS.

The whole thing was fascinating. I'm interested in learning more. A big question I have is what's the point of the CS course or the major for that matter? Pure academic exploration? It seems a poor design for job prospects although that's something that Keith brought up - the push and pull between what he taught which was purely academic, being from a small liberal arts college and the prison's desire for a more jobs oriented program.

The other BOF I attended and loved was titled "Towards a Computer Science Curriculum “Microkernel" and raised the questions of if there's a common core that could/should be taught in CS. This was very much related to some of the themes that came up during Friday's keynote and is something I've thought long and hard about for years.

Many similar themes as the keynote but we talked about them in small groups - do we need all the courses that we teach? Do we need to teach as much about sorting as we do? What's the point of teaching any particular topic?

I don't have too much to say in terms of specifics, but I'm glad this conversation is happening and happening in public. It's so hard to move institutions and even when you can steer the ship it can take years or even longer. At Hunter, there were some people who wanted to reform the CS major but all it took was a couple of "we always did it this way" people to sink any innovation.

Still, with the conversation happening maybe some places will have success.

So, those were the keynotes. Stay tuned for more on a bunch of other types of sessions.

prison

CS Microkernel BOF

Back in the classroom - week 1

Sun, 09 Mar 2025 13:55:20 -0400

As you might have heard, I'm back at Stuyvesant teaching.

Not permanent or full time - just covering for a CS teacher who's on grand jury duty. She's out for three week but since I was at SIGCSETS for the first one, I'm only covering for two. Last week and the one upcoming.

I wouldn't be covering all 5 of the teacher's classes, just 3 - periods 6, 7, and 8, the second half of Stuy's intro CS class. A class that I actually designed ages ago, of course, the class had evolved since I left Stuy about a decade ago.

The class used Python and the classes were at a nice place. They had learned up through and including lists. I could have just ploughed ahead and covered the rest of Python, starting with dictionaries but I decided better would be to do a couple of interesting applications.

My thought process was that I'm not the regular teacher and while their regular teacher brings certain things to the classroom that I don't, similarly I bring some things that she doesn't. If I did list application lessons that let me work my spin on the class, the kids would potentially be getting more out of the class than just someone teaching mundane syntax.

I decided for the first unit I'd do this lesson. It's a nice application of lists, introduces some deep runtime concepts, and is an overall nice gateway subject. I won't talk about the unit I'll start tomorrow on the off chance that one of the students happens upon this blog. I don't want to give away any surprises.

I figured I'd have fun with the gig but I did have a number of concerns.

First and foremost, I'd be "the sub." The classes were sent an email saying I'd be coming and that I had originally developed the course but who knows how many kids paid attention to the email or what that meant to them. I figured I'd get a certain level of respect - Stuy kids are good kids but still, I'd be the sub.

I also knew that by having only two weeks, I wouldn't be able to set up my culture. It wouldn't be enough time for me to really get to know my students nor for them to really get my schtick. Similarly I wouldn't have time to set new class protocols and procedures effectively.

On top of that, I had my first "grand student." Specifically the child of former students and in fact students that I've grown close to.

On the one hand, I was the sub, on the other hand, if there were expectations due to reputation, since I was only there for two weeks, I'd likely not meet them.

On the third hand, if I weren't there, they'd be going multiple weeks with no live CS instruction, anything I did get done was a bonus.

So, I figured I'd keep things simple. I'd use my usual bag of tricks - live coding with questions and predictions, Eliciting subgoals, and similar techniques that don't require deep knowledge of my students.

So, how'd it go?

Well first off, the kids haven't really changed in a decade. It was similar to when I started a class in the Fall. Students by and large well behaved, mostly quiet with a few more frequent contributors and all in all good kids and smart kids. Over the week it seemed like more kids were warming up to me as the teacher and to my take on the class. Sure, it was mid semester but the dynamic was pretty much like when I'd start any class at Stuy.

It was notably different though from starting a class at Hunter college but I attribute that to the fact that I meet with the Stuy kids every day and at Hunter it was only twice a week.

The students acted in ways I expected and made mistakes that I expected them to make. All in all it's been a fun experience for me and I hope worthwhile for them.

Now, things haven't been perfect. There have been kids spending the class period doing other work or on their phones but as long as they haven't been a distraction to the kids who've wanted to learn, I've let it slide. I figure that this is on them and with two weeks, I'm going to give everything I can to the people who want to learn rather than putting time and energy into the few who don't. Were this my actual full time class, I'd handle things differently.

I guess another thing worth commenting on is that this is the first time in years when I've had to be somewhere to do something five straight days. While I'm enjoying the teaching, I'm not enjoying the every day specific time obligation. Maybe if it were first thing in the morning it would be different but teaching between around 11:50 - 2:00 is just a bit awkward.

So week's in the books. Starting another topic that I think is pretty fun tomorrow and we'll finish off the week with an AMA.

SIGCSETS 2025 - day 2 keynote - What belongs in CS

Thu, 06 Mar 2025 19:06:41 -0400

The day 2 keynote was really a treat. After giving out the years SIGCSE awards, instead of getting a standard talk, all three recipients, Jonathan Mwaura, Mitchel Resnick and Manuel Pérez-Quiñones participated in a fireside chat with Dan Garcia. The only letdown was that since Dan was moderating the chat, we also didn't get his views into the mix.

With three independent voices, the chat pushed in a few different directions but the theme that spoke most to me was when Pérez Quiñones, commenting on what the future held for CS Education said "I hope the future is less so we can do more." The gist being that college CS curricula has become really bloated.

Of course, it's much more than that and at the same time the statement is pretty nuanced.

On the one hand, CS Curricula has become more and more packed. Course after course has been added while none have been removed. Now, that's not entirely true. I seem to recall that Compilers was a frequently required course back in my day and that's no longer the case but what about operating systems? Does every undergrad need a 3 credit OS course? What about the core sequence? Is it necessary to implement every sort? If not, how many, which ones and why? I'll return to this question when I write up some of the individual sessions.

Maybe we could take the important parts of all the current classes, ditch the cruft and distill it all down to fewer requirements. This would allow programs to more comfortably add in some of the new topics that might currently not fit - software engineering, HCI, Ethics, stuff relating CS to the humanities, and more.

Another spin on this was the question of what would or should remain in CS and what would spin off? Should their be a separate Software Engineering Major? What about AI? How about CS classes for non majors? Should all of this be in CS or should CS become a small program that just focuses on the theoretical core?

I don't thin the panelists nor the audience were or are ready to propose a hard answer here but I'm really glad the questions were raised.

Coincidentally, or maybe not, Mark Guzdial just wrote a post asking a related question, his focus being on CS for non CS majors. Alfred Thompson responded with his related thoughts on the High School level.

For the high school level, I feel that our job for all students is three fold. Give them enough of a background to as a lay person understand the fundamentals of CS. Give them some practical tools (ie. programming) that they can employ regardless of their future path, and finally provide inspiration for those who sync with CS so that they can take future courses and pursue CS (or more likely software engineering) as a field of study.

This is what we do, or try to do in most other subject areas. Everyone takes Biology. They learn a bit about the field, they learn something about the field that's in theory relevant to their lives and hopefully some will be inspired to study more advanced bio either in the upper High School grades or beyond.

I think we did and are doing a very good job with this at Stuy where I designed a course specifically for this around 25 years ago. AP CS Principles could be this but while I've seen some solid implementations, I've also seen far too many weak implementations.

College is another matter and is tricky. Should everyone take a CS0? Should it be focused on majors, non majors, or try to do both? Should CS departments run a whole bunch of intros? There is at least one college that does this - I'm blanking on the name, and it seems to work very well for them. On the other hand, this leads to all sorts of resource questions. Another option is for other departments to take over intro CS courses appropriate for their disciplines.

No easy answers but I'm super happy that the conversations are starting. Nothing will resolve itself any time soon as education moves slowly but as they say, the best time to plant a tree is 20 years ago, the second best time is now.

There were other topics addressed in the chat - CS education for giving everyone agency. Not merely using all the modern tools and advances to do the same thing better but to move closer to the type of world we want to live in.

AI of course also came up with Pérez Quiñones talking about how having copilot took away a teaching moment between him and a student - something I've seen myself with older code help tools.

I'll talk more about my thoughts about what belongs in CS and were it belongs in another post and might also share my thoughts on Mark and Alfred's posts but for now, I've got to say, this was my favorite keynote in a long time.

SIGCSE 2025 - Opening Keynote

Mon, 03 Mar 2025 19:40:54 -0400

I spent last week in Pittsburgh for The SIGCSE Technical Symposium, that's SIGCSETS to you. Lots of talks, lots of sessions and lots of people.

We drove in on Tuesday since Batya had a workshop Wednesday afternoon. Devorah and I along with a cousin who was also at the conferences spent the day playing tourist. We hit the Warhol museum and also wandered by Randyland. Even made it to Carnegie Hall without practicing.

The opening keynote was given by Cecilia Aragon from the University of Oregon, the topic "What is Human Centered AI and why does it matter." The gist being that algorithms aren't free of bias regardless and that keeping the human element in CS is both critical and beneficial.

Here are some of what I felt were highlights. Some were core parts of the talk and others might have been a throwaway line but it still resonated with me or I otherwise thought it was memorable.

Early on, Aragon questioned if human centered was soft or easy with easy being erroneously perceived as either not rigorous or otherwise bad. She used to believe the easy is bad mantra due to CS's weed out culture but has since learned otherwise. This part of the talk was reminiscent of Felienne Herman's talk that's recently been making the rounds.

Aragon also talked about believing that ethics should be woven into all classes and not merely be a standalone even if required. This echoed with me as I've been saying this for years.

In another part of the speach, Aragon spoke on how she was working on a project to identify supernovas and how she moved it forward significantly in spite of two prior teams and team leaders failing to do so. She attributed the advancements to two factors. One was some technical, specifically algorithmic changes she implemented but the other was the human element. Aragon encouraged more collaboration and discussion and as this was an astronomy project, cross collaboration and discussion between the CS people and the astronomy people. It was that cross pollination that led to the technical improvements and was cited as an example of the "human side."

It was a good story and I agree but the sad part is that the example she gave shouldn't be an example of the "human side of AI" it's just how a good manager and good colleagues should work. Sadly it isn't always the case.

Another theme of the talk, and I mentioned this already is the fact that bias is already baked in. Our creations are biased because we are biased and that's something that we need constant reminders of and we have to do better.

A final point that was raised which I felt was particularly important given our current world situation, or more specifically, our current America situation was about a survey published by Stack Overflow. I forget the year and didn't copy down the specific numbers but you'll get the idea.

First question Aragon presented asked developers if they'd work on an ethically questionable. The majority said no which was a positive sign. Unfortunately, a later question asked something like what amount of responsibility a developer had when working on something unethical. In this case, the majority blamed management taking no responsibility for themselves. I was just following orders as it were. It was an old survey but I don't doubt the same mindset persists and it's greatly troubling.

Professor Aragon concluded the talk with recommendations on how we can do better.

Overall, a strong opening to the conference.

Has the CSTA jumped the shark?

Fri, 21 Feb 2025 11:50:55 -0400

That's the Computer Science Teacher's Association to you.

Has it?

Well, that's a personal question. I think it has for me but I'm sure for other people, that jump the shark moment will be far off or never.

I'm ruminating on this now as a result of a recent change in their membership policy. Going from two tiers - one free and one costing $50 a year to a single tier $50 membership. If you're currently a member, you can elect not to pay the $50 and still keep access to a limited set of resources.

Alfred Thompson wrote a great blog post with his take on the change here.

When CSTA started it was free. Back then, there were far fewer computer science teachers and while I joined early on, I saw very few opportunities to get involved. That's not to say there weren't any or plenty, just none crossed my path. At one point I did try to reach out to their leadership but I found them dismissive. I got an email back seeming to indicate "well, you're at Stuyvesant so what you do doesn't count."

Anyway, in 2009 or so I had the opportunity to get more involved so I did. That led me to feeling that back then the CSTA existed for CSTA and they expected the membership to work for the benefit of the central organization.

Some time around 2018 or 2019, I felt that changed. To me, anyway, it seemed like the emphasis really turned around to support the teachers and local chapters. I applauded the change and many of the efforts.

That was around the same time that CSTA announced CSTA+ - the $50 membership. I immediately joined. Not for the extras that it provided but just to support the organization. I think a lot of other people joined for the same reasons.

Personally, I've never found the CSTA materials and resources useful. That's not an indictment of them. I haven't found anything on code.org useful either. That's probably because I came in to teaching with a strong CS background and had developed a robust High School CS program before CSTA ever existed - I was ahead of the curve. I wasn't happy then and still am disappointed when I see a lack of actual CS coming out of CSTA but in any event, I knew that the materials weren't for me, they were for newer teachers and teachers just coming to the subject.

For a while there, even though I wasn't really the audience, I felt I could get behind and support the CSTA.

Fast forward to today. Why to me has CSTA jumped the shark?

Some time in the last year, I decided that I wasn't happy with the direction of CSTA. I was particularly unhappy with the board of directors election process. I had applied to be on the slate. I assumed there would be some manner of open process - the membership would either vote or they'd cull down the candidates in some open and fair way and then vote on the final slate.

Nope.

I was told that the committee decided to go with other candidates for the election.

Is my opinion on this sour grapes? Maybe but I don't think so. When I read the candidate bios I left feeling that some candidates were equally or better qualified to be on the slate than I was but then, some, in my opinion brought much less to the table.

Had there been an open process that left me and others off the slate I would have been fine with it but I was left feeling that the committee that made the decision was massaging the slate so as to get the board of directors they wanted.

I was also unhappy with the tone and direction of CSTA for a variety of other reasons.

So that's when I decided to drop down from CSTA+ to CSTA free tier.

I still wanted to go to this years conference, however. When I went to register, I saw that the free tier registration was $50 more expensive, which was fine, and that those additional $50 would pay to put me back into CSTA+. Every time I think I'm out, they pull me back in!!!!! :-)

Then I started getting the emails about the new dues structure urging me to upgrade.

Weird. I though when I registered for the conference, that took care of the upgrade. Turns out, I won't be upgraded until the actual conference. Okay, weird and not a great look based on emails but that's fine.

So that's why the CSTA has jumped the shark for me but I wouldn't be surprised if this membership change makes them jump the shark for a much wider set of CS teachers.

Alfred, in his post already talked about the cost and how many CS teachers are also from other disciplines and those costs multiply across subjects.

The big one to me though is that CSTA central will only support local chapter events solely for paid CSTA members. I hope I'm understanding this one incorrectly but a big part of local meetings should be working to get more teachers involved. That means meetings with non CSTA members in the hopes of roping them in.

If CS was like math and had been around as a school subject area forever then saying "hey, pay $50 and then and only then can you go to the club meetings" might play a little better. Might but that's no certainty and we're nowhere near being close to ready for it.

I get that money might be tight but in my opinion, if the CSTA wants to be long term viable as a teachers organization (as opposed to a top down organization paid for by say the Googles of the world) then they have to support local efforts that bring in more teachers even if those local efforts are still working to get their legs under them.

I'm also curious as to what the messaging will be with conference registration. If they're smart, since there will only be on $50 tier it will be marketed as "come to our conference and get a free years membership."

So, will this positively or negatively affect membership? While I don't have any numbers, at least CSTA meeting numbers in New York City isn't particularly high and is largely dominated by teaching adjacent people. This always struck me as odd since New York City's CSForAll initiative has resulted in tons of schools offering CS but those teachers don't come out. Add a $50 membership fee and I don't expect the numbers to rise.

Of course, outside of New York City, things could be very different.

In any event, we'll see how many more CSTA conferences I attend but I don't see myself engaging much with CSTA other than that under the new proposed model with the CSTA going in its current direction.

It'll be interesting to see how membership and attitudes change moving forward.

Getting Back in the Game

Wed, 19 Feb 2025 11:26:32 -0400

Even before I retired I knew that after a break I'd want to get back in the game. Not full time, but in a limited capacity when interesting opportunities presented themselves.

Well, it's time to start dipping my toes back in the water.

Next week I'll be at SIGCSE but that doesn't really count. Last year I attended both SIGCSE and the CSTA conference so that's nothing new.

After I get back from SIGCSE though, I'll be covering computer science classes for a friend at Stuy who's going to be out on Grand Jury Duty. It's three sections mid day, the second half of Stuy's intro class - now dubbed CS Foundations. The material should be straightforward. I think they're still learning the basics of Python.

I've got a couple of concerns though. First, in the email that told the classes that I'd be covering, I was introduced as the guy who created the course. Now, I'm a good classroom teacher but really, my value isn't in individual super lessons. I'm not Edward James Olmos as Jaime Escalante (though I wonder if Jaime Escalante was Edward James Olmos as Jaime Escalante). I've got my moments but my rapport with my kids builds over time - a semester, a year, or even multiple years. I'd rather under promise and over deliver so I hope I'm not being oversold.

My other concern, well, not really concern, is that I hope they're far enough along for me to do something fun. I'm only with them for two weeks. If they've gotten to or close to lists, I can do some fun and interesting applications. If not, well, teaching if statements can only be so interesting.

In any event, I hope to have fun with this.

One extra surprise is that I'll be getting a grand-student out of this. I'll be teaching a child of one of my former students.

I'm also getting back involved by supporting CareerWise, an organization that sets up apprenticeships for high school students. I was introduced to my new friends at CareerWise by an older mutual friend and it was exactly the type of intro I was looking for. A friend thinking "hey Mike and these people should do something together."

CareerWise sets up and supports apprenticeships in a bunch of fields but I'm of course working with them on the tech side. They've got a software engineering track and a cybersecurity track. The basic plan is that the students come from schools all across New York City, they receive training over the summer - think of it as some sort of boot camp, and then go to their apprenticeship along with continued support from CareerWise.

They're doing great work but tech apprenticeships are a particular challenge. As I wrote earlier, companies are expecting even interns and entry level employees to know more and more coming in. It's not like the old days when companies trained up their workforce. Of course it's even harder with high school apprentices. They'll come in knowing even less and will have even more limited exposure to how one operates in the workforce.

Now, as I said, the apprentices get some training the summer prior to their placements but that's a real challenge. First off, while a small number of the kids might have had some serious CS, most probably had APCS-Principles or maybe not even. Think about it, in a short summer program you have to take a high schooler who at most has written very short programs in a drag and drop environment and industry wants them to be ready to go with a text programming language, version control, using an IDE and more.

Not gonna happen.

Even harder, one company might be looking to place their apprentices on a front end team using Javascript and React, another a backend Java team, another something in Python and so on.

This is really tricky. How do you design a boot camp for that? Even if you focus on the basics, if you prepare them for one shop, you wont be preparing them for another one.

That's where I come in. The hope is I can help manage expectations with the partner companies - help them to understand that these are high school kids with limited backgrounds and experience. They're smart and they'll work hard but a twelfth grader is very different than a college rising junior starting an internship let alone an entry level grad. The other hope is that I can help direct the training so that the training and the company expectations meet in the middle.

This is not an easy task and I don't have any magic bullets but I do have some ideas.

Well, on the expectation management side, that's easy. Well, not easy to actually manage them but easy to have the conversation. We've already started so hopefully we're going in the right direction.

The boot camp / training is harder. What to do? After noodling about it a while, I thought to redirect the training a bit. Instead of focusing on a bootcamp where kids create their own frontend only app (well, really more of a demo with very limited functionality) or some other completed program, I thought, maybe we should make the application "testing." That is, as the kids are going through the training, have them write tests for existing code bases (provided by the company that runs the training). This could evolve into tweaking and extending code as well.

When the kids start their software engineering internships, they'd start by writing tests rather than creating new features and again, they could, as they get comfortable in their apprenticeship be pushed then towards creating new code.

Why do I think this makes sense?

From the companies point of view, writing tests is super low risk and is something they always need more of. It also keeps the kids out of the codebase until their ready. From the kids point of view at the company, it will probably be easier for a kid to read and understand a small chunk of code and write a test for it rather than creating something new to integrate into a complex codebase.

From a training point of view, this exposes the kids to code reading and understanding - something rarely done and writing tests is probably a more gradual ramp up than diving into full programs. The kids will also learn professional workflows.

True, they'll still only be doing it in one language so if the training is in Javascript and their apprenticeship is in Python there will still be a professional ramp up but it will probably be easier to write tests in the new language given the example code of the code base, the code their testing, and probably other tests to mimic.

We'll see how it goes

So, that's what I'll be spending some time on over the next few weeks and months. Still just part time but happy to be back in the game.

Know your (CS and tech) history

Sun, 16 Feb 2025 15:06:31 -0400

On Friday I saw this post shared by a couple of friends of mine.

Why did Musk and Trump make these false claims? Ignorance? Stupidity? Malice? Well, with the current Republican party, usually it's a combination of all three.

But wait, should I just be accepting Toshi's post explaining what's going on? I usually don't accept random posts as facts. Well, in this case, it was posted independently by a couple of friends of mine who are knowledgeable, smart, and also don't blindly share posts they can't confirm as fact.

And then, there's the fact that this makes sense to me given when I know about computing and computing history.

If one looks up the [[https://en.wikipedia.org/wiki/ISO_8601 ][ISO9601]] spec, indeed the Epoch for the standard is May 20, 1875, 150 years ago.

This isn't strange. I mean, the Unix Epoch is January 1, 1970. That is, Unix systems tracked time by the number of non-leap seconds since 1/1/1970.

Using a time representation like this can have some interesting effects and consequences. Uninitialized dates appearing to be 150 years ago being one of them. On the Unix side, the time has typically been stored in a 32 bit number. With 32 bits, we'll hit our largest time, or maximum date early in 2038 at which point, the calendar will roll over back to 1/1/1970. This is known as the 2038 problem and it's not good. Now, this problem might be averted if by then all systems use a 64 bit representation for the time and date but I wouldn't hold my breath. I mean, the systems which falsely show "150 year old people" are cobol systems probably written before the first second of the Unix epoch.

Another similar time/date problem was the Y2K problem. That's when computer systems written prior to the year 2000 used 2 digits to represent years, I mean, who'd still be running a program that long that it would cross over into the 2000s. Well, it turns out quite a few. This caused a big stir in the late 1990s. To the outside world though, 1/1/2000 came and went and it seemed like it was a big nothing but the truth is, a lot of people put in a lot of time behind the scenes to fix the problem.

Now, all three of these example could be problems or might not be problems but if you know a bit of your computing history then you'll understand what's going on and won't fall for the lies and misdirections hurled by Musk and if you're a tech person, you'll be able to deal with them should they occur in your day to day.

These artifacts of history appear all over. Here are a couple more of my favorites.

Using a single letter variable i as a counting variable in a loop. There's a school of thought that says you should never use single letter variable names - they should always bee clear full word labels. Using i for a loop index is unclear.

I saw a ~~argument~~ discussion of this online a few months ago.

Now, you could use something like index for your loops and in fact you could use text completions so you only have to hit i but personally, I think having things too long is just as bad as too short.

I see nothing wrong with using i for a loop and understanding the history helps give context.

In early programming languages, you could only use variable names of limited lengths. For example, in Fortran IV, a language I used early in my career (even though it was well out of date by the time I learned it), variables only had 6 significant characters. I also used a version of BASIC that only supported 2 characters.

That is, in that version of BASIC, you could have variables named index, indecisive, and in and they'd all refer to the same variable. Everything after the second character was ignored. Fortran IV gave us more characters but was still limited.

Then, add to that, in Fortran IV, variables that started with the letters i,j,k,l,m, or n were integers. Start with any other letter and the variable was floating point.

So, in this context, using short variable names made sense and using i as a counting variable or index for a loop or array made even more sense.

To me, at this point, it's an idiom in programming and makes for concise lines and when scanning code, it's easy to read confirm your loops are okay. It also makes the longer word length variables that live beyond your loops stand out so in my opinion it's a win win.

Sticking on the Fortran IV kick, let me talk about JCL - job control language used on big iron, that's IBM mainframes to you.

When I was at Goldman Sachs. before I started my teaching career, I was friends with some Cobol programmers. They would complain about the arbitrariness of JCL. I don't know exactly how they worked but it seems that before they could submit their cobol programs to run, they had to submit a JCL program right ahead of it to tell the mainframe about the forthcoming cobol program and how to deal with it.

They complained how certain things had to be in certain columns and how the entire process was incredible rigid.

Now, this was around 1989 so they were all programming on "modern" terminals. I started talking to them about this and they showed me some JCL and I immediately got it.

It's what I used on punchcards back at Stuy when I learned Fortran IV on our IBM 1130. I had no idea I was using JCL but I had to make a few special punchcards with very specific instructions and put them at the top of my program deck - all the cards that represented my fortran program.

Well, that was JCL.

The column specifications made instant sense to me. A punchcard had 80 characters and the reader was set up to read all 80 characters at once.

It made all the sense in the world to put column specifications and restrictions on the language given that it was for punchcards. In the 1980s and beyond the punchcards had gone away but JCL lived on.

I explained this to my friends and they got it. Didn't make JCL any less annoying but they were then able to better grok the system.

Another one is tar - the Unix archive system and zip - used by Windows.

A Zip file (and its later more modern counterparts) takes a bunch of files, compresses each one and the combines all of them in an archive. The Unix tar command combines all the files, uncompressed into an archive then you typically compress it with a program like gzip.

Each of course, has it's advangate. Since you compress tar files after combining them, your compression program can exploit similarities between the files in the archive. On the other hand, zip files allow you to easily extract any individual file without having to uncompress and unbundle the whole kit and caboodle.

Why the difference? I don't know why zip was defined the was it was but tar was a historic reason. Tar stands for "Tape archive." Hard drives were small at the time and backups were typically stored on tape which was cheap and big. The thing with tape though is that it's linear - no random access so you'd just send all your files one after another to the archive. You can do that if you're not compressing but not if you are. Later, when tar files were stored on disk drives, people started compressing.

Another and one of my favorites goes back to that religious war - Emacs vs Vi(m). I'm an Emacs used but see nothing wrong with Vi(m) but I get really annoyed when Vim users start talking about how Vim is inherently more efficient. It's modal design and using the h,j,k, and l keys were chosen to be faster.

I get annoyed becuase the people making those claims don't understand history. They're also wrong. Modal editing is not inherently faster and as a lefty I can say that using the h,j,k, and l keys are not optimal for me. Of course, if Vim works for you and is more efficient for you personally, that's great.

The thing is, the Vim design is from history. The modal thing comes from Vi being built out of a line editor. It was a Vi-sual interface to a line editor. The keys? Well, it happens that when Bill Joy wrote Vi, he was working on a keyboard that had the arrow keys inscribed on the sides of the h,j,k, and l. Back then, people were using much slower computers, frequently using slow modems or even teletype terminals. Blinding cursor movements were not high up on design decisions.

I actually made a 35 minute video on Emacs vs Vim that goes into the history. You can view it here if you're interested.

Now, the tar vs zip and Emacs vs Vim and even the i as a variable thing aren't really important, at least knowing the history isn't important but the date stuff is, particularly with Musk weaponizing lies about it.

On the other hand, the more of this history you know, the more you understand why things are the way they are.

As a whole, we do a pretty poor job teaching CS and Tech history. There have been some efforts recently to do better but they're mostly based on the people. Specifically on CS history from members of underrepresented populations. This is all good - it's important and I'm on board but adding technical history is also important.

Personally, I wish I knew more of it.

Savannah Trip Report

Thu, 06 Feb 2025 19:57:03 -0400

Devorah and I spent this week in Savannah Georgia. Our first time. For the past few years, we've taken a trip mid winter to someplace warmer. Not necessarily warm but warmer.

Last year was Charleston SC, this year Savannah.

The weather was perfect, at least for me. Mid 50s in the mornings so I could run in shorts and short sleeves in the morning and in the 70s by late afternoon.

We stayed on River street - right by, well, the Savannah river. That made running easy in the mornings. Each day I was up at around 6:00 out and off. The out and back by the river was a little shorter than my usual vacation run which was a bit annoying but still, it was mostly off street. Much better than other vacations when I had to just run in sidewalks and streets and watch for traffic at each intersection. Thursday morning was particularly foggy which also is kindof fun to run in.

Staying right by the river though did have one disadvantage. We were, staying in the cellar. Other than River Street, the rest of downtown is one flight up.

One interesting note was that the town is really empty. Busy season doesn't start until March or a little later but I was surprised. We read so many reports about how crazy lines were for places like Leopold's Ice Cream but there were no lines for anything. Well, one place - we couldn't get a reservation for the Old Pink House but that's fine, it's probably a bit touristy and we had lots of great food elsewhere.

Okay - here's what we did.

Houses of worship

Visited some really interesting houses of worship. First up was Temple Mikve Israel. America's third oldest Jewish congregation. The only older ones are Sheerith Israel in New York and another congregation in Rhode Island. All three predate the USA. Mikve Israel, like Sheerith Israel was founded by Portuguese Jews who landed in Savannah just months after Ogelthorpe founded the colony.

It's a reform congregation, having split with the orthodox faction years ago and the current building was built with that in mind. It's also buit in the gothic style which is somewhat unusual for a Jewish house of whorship.

Lots of interesting history including a Torah scroll from the 1400s.

We also visited the Cathedral Basilica of John the Baptist.

I didn't get a picture but it had an amazing pipe organ.

Having hit Jewish and Roman Catholic, our third house of worship we visited the First African Baptist Church. Great tour.

We learned about the church - its formation and development, largely by enslaved people, it being a stop on the underground railroad and more. Fascinating.

History

For the history part of our trip we hit the Massie Heritage Center - the first public school in the American South;

They had the history of the school and education, including information on the self educating efforts of the enslaved population along with the expectred artifacts. There were also other interesting exhibits in the school.

THe Owens-Thomas house and slave quarters was another must see. A house museum tour you get the usual info on the well to do homeowners and the history of the house but we also got to see the slave quarters and there was an emphasis on the slave experience. Highly recommended.

We also visited the Savannah History Museum which had some general Savannah history along with an exhibit on Jazz and one on Juliette Gordon Low - founder of the Girl Scouts and Savannah resident.

Right outside the history museum was the site of the Battle of Savannah from the American Revolution:

Art

We also got out art fix at the Telfair museums. Two art museums right next to each other. One, the Jepson center, had an exhibit on Venice and the Ottoman Empire as well as a terrific sculpture exhibit and an electronic media ehibit as well.

The other, housed in the Telfair mansion was a more traditional art museum but you also got to see the mansion itself.

These two art museums and the Owens-Thomas house were all on the same ticket. $30 and you had one week to visit all three.

Wel also visited the SCAD Museum of Art but only a couple of exhibits were open.

Bookstores

We also visited three independant bookstors - something we like to do while travelling. All three were terrific. The Book Lady was filled with both used and new books while the E Shaver bookstore all new. Both were great but I liked the feel of E Shave because it was a Frankenbuilding. If you've been to Powels in Portland you know what i Mean although E Shaver is MUCH smaller.

Third, Books on Bay was also a gem - a used book store but they had tons of old series and old kids series specifically. The Hardy Boys, Nancy Drew, Doc Savage, Tom Swift and more.

Devorah bought a "Sue Barton" book that she remembered from her childhood and I bought a couple of Agatha Christies.

Food

Food was also great. Lots of great meals but the standouts for me were B. Matthews Eatery, a farm to table place just a couple of blocks from our hotel and Carly's Cafe for diner type food. We also had solid chinese at E-Tang's.

Of course we went to Leopold's Ice Cream multiple times. We were told that Leopold's is rated as a top 10 ice cream place in the world by whoever rates ice cream places. I don't know if that's true - we've had lots of great ice cream in lots of places but it was really good. We noticed that the labels on the ice creams were changed tonight. Each was now named after a book since Savannah's having a book festival this weekend. As it turned out, the ice cream I got was named after the book Devorah just read and what she got was named after a book I just bought at E Shaver's.

Misc

We also spent a lot of time walking around - really beautiful city.

So, that was our Savannah trip. Flying home tomorrow morning.

Don't know if we'll get back here or not but if you get a chance, come down in early February, you'll have a great time like we did.

What we teach and when we teach it Java edition

Tue, 04 Feb 2025 20:06:40 -0400

It seems that every year or so we have a discussion as to what's the best programming language to teach first.

I thought today I'd look at a variation of that question. What language features should we teach, when and why.

I was thinking about this after someone in my StuyCS alumni community asked about which Java version was used in APCS. He was tutoring a current Stuy student and was wondering about language features. Specifically things like interfaces and abstract classes. Of course the question got murkier when he was reminded that Stuy doesn' just teach APCS-A but rather a superset of the old APCS-AB which included both programming and data structures.

Now, in terms of APCS-A, it no longer covers Interfaces and Abstract classes and I think inheritance is also on its way out the door.

I also seem to recall that the College board was thinking about removing Java Arrays from APCS-A and only requiring the teaching of ArrayLists but I might be misremembering.

From a pratcical "we want more people taking our exams and we want them to get reasonably high scores" point of view, removing all this material makes sense. I mean, when I took APCS the first year it was offered it covered a full year of college in a full year of High School. When they dropped the data structures, it became a half year of college in one year of high school. As they remove more and more material, the class is becoming less and less of a college course and more of a high school one. Now, this isn't necessarily a bad thing if those students are getting a solid foundation in the basics.

In my last few years teachig at hunter, I had about half my students come in having taken APCS-A. Hunter didn't award placement for APCS, partly because Hunter used C++ as their core language but truth be told, all my Hunter kids that came from Stuy and took Stuy's APCS could have easily started in CS2 and most could have started in CS3 with a short C++ for Java programmers workshop.

On the other hand, the vast majority of my non Stuy students who came in with APCS-A were not ready to be placed in to Hunter's CS 2 (data structures) class. A few were ready for CS2 or even CS3 but most weren't. Now, these were all very bright kids and had a very easy time with Hunter's CS1 it's just that the APCS-A classes they took in High School, by and large didn't cover enough to get them ready to be placed out of a college CS1, particularly one in a different language.

I do want to emphasize though that all my students who took APCS-A in high school breezed through my CS1 in college and then did well in my CS2. That APCS-A course they took was very valuable. Was it college level? Probably not but who cares, the courses they took did them a lot of good.

Now, as to what, why, and when, is it really important to teach abstract classes? In APCS, probably not. Abstract classes are useful and arguably important when writing libraries and when working on large codebases. You do neither of those things in APCS-A nor in pretty much any CS1 class. Sure, you might write a library but it'll be pretty small, probably contrived, and you won't then have to use it in multiple circumstances. Similarly, you won't be writing a large project, there just isn't the time. You might add code to an already existing large project - maybe implement something off of an abstract class but doing so will either probably be contrived or it'll be seen as just something you do, like implementing the comparable interface. Students probably won't "get it."

Same with interfaces. Java implemented them to get around the limitation that resulted by getting rid of multiple inheritence. They got rid of multiple inheritance due to its complexity and things like it leading to circular references. Still, this left some problems unsolved an interfaces took care of them.

Still, teaching them in APCS-A was always forced.

The truth is, for things like abstract classes and interfaces, even if they're considered basic building blocks in Java, one doesn't really grok them until they're writing a long term reusable library or working on and in a large codebase, again, usually over time. A single semester early in sequence class just doesn't have the time and the students don't have the experience.

Things like abstract classes and interfaces probably make more sense in a more advanced Programming Languages / Paradigms or OOP course or maybe better a Software Engineering course.

Unfortunately, if they're not tought in CS1 or CS2 they might never be taught. More unfortunately, if they are taught in CS1 or CS2, students likely won't "get" their importance so it won't be much better.

Now, does any of this matter? Maybe and maybe not. For interfaces or abstract classes, I don't know how important it is for someone on the academic computers science path. That is, the path that most college CS programs prepare you for under the false belief that most CS majors will pursue PhDs. Of course, the reality is that the vast majority of CS majors will seek to go into tech, many as software engineers so maybe the software engineering and software architecture concepts underlying abstract classes and interfaces are actually important.

There's the rub.

It's generally not a big deal if you remove some concept from APCS-A or any CS1 as long as, if it's important, you address it at some point in the future - hopefully at better point in the case of the two topics I've been discussing.

Of course, this has happened before. Back when APCS switched from C++ to Java, memory management disappeared. That was all fine and good, particularly since memory management's a tricky topic to cover early on but for some students, it was never covered. This is one danger with taking your CS1 (and back then potentiall CS2) in High School, or rather in one institution and the rest in another. If you took APCS-AB in high school and then went to a college that say, used C++ and tought memory management in their CS1 or CS2 then you'd totally miss out on an important topics.

Now, for memory managment you might say that most people will never need it - Java, Python and many popular languages take care of that for you. True, but many don't and having a good understanding of what's going on under the hood makes for a better software engineer or tech professional.

Same for the Java topics. While I don't think CS1 is a great place for Abstract classes or interfaces, the concepts beneath them are important for a well educated CS person or Software Engineer.

The trouble is, we spend too much time looking at courses and not at complete programs and pipelines.

Just something to think about.

Switching Domains

Thu, 30 Jan 2025 14:26:18 -0400

So, while I'm working on getting off of Google, I thought it was also time to unlink my blog from GitHub.

It's still, at least for the time being, hosted on GitHub but I've switched it over to my domain.

The blog's official address is now cestlaz.zamansky.net. The old address of cestlaz.github.io should still work since it's still hosted on GitHub. I'll probably move the actual hosting somewhere else at some point int he future.

IMPORTANT for RSS PEOPLE: If you read this blog using RSS, switch to the new RSS feed at https://cestlaz.zamansky.net/rss.xml. I think the old feed still works.

Finally, could somebody leave and/or reply to a comment here - I think I switched those over correctly but would love to test it out.

Thanks Mike

Choosing our platforms

Wed, 29 Jan 2025 15:23:26 -0400

In my Blog Reboot post I talked about blogging being a fading pastime. Over at Irreal, JCS responded that his experience says otherwise and that his RSS feeds in fact experiencing a rebirth.

Both can be true. I've also seen an increase in what are essentially blogs - more people on Substack and the like - I regularly read Kareem Abdul-Jabbar's substack as well as others. On the other hand, there's clearly been a dropoff in CS Education related blogging and also less dialog on the blogs that are around.

Part of what led me to my blogging is dying side comment is something that I've been thinking about a lot recently which is our choice of platforms.

The most active CS Ed communities I'm aware of live on Facebook. I've always felt that Facebook was a problematic platform. It's a closed silo. I'm not talking about the fact that groups are controlled by their respective admins. The CS Ed groups, by and large seem to have reasonable ones. I'm talking about the Facebook platform. The groups are not discoverable from the outside and while searchable, only from within each individal group. If Meta decides to change policies, the group or groups might go away, all data is lost and all membership is lost.

Contrast that with what I use for my StuyCS family (or Mafia as some call it). That's just a mailing list. The entire email archive is stored online but can be downloaded at any time and even if the provider closes up shop, I've got all the email addresses and can restart elsewhere. Of course, a mailing list has it's own limitations but that's another discussion.

While I've been concerned about Facebook for a while, Zuck's most current actions makes me even more disturbed about using it. While I can't speak for the entire CS Ed community, certainly the subset on the Facebook platform seem pretty concerned with DEI, equity, and the like. Well, we know what Zuck did there. Then there's also the fact checking.

Maybe as a community we should walk the walk and not just talk the talk.

Personally I've been spending less time on Facebook in general. Other than the CS Ed and a couple of other communities I'm finding it less useful. It used to be great to keep in touch with a subset of former students. The ones that I might not be close enough with to have an ongoing connection but still enough to care to see updates and send a hello now and the. Well, more and more, those people are opting off the platform, or at least the algorihtm isn't sharing there stuff as much and just giving me algorithm generated garbage.

Of course, one needs alternatives. It would be great to see a CS Ed blogging rennaisance but that's dependent on both creaters to create and readers to comment.

CSTA has forums but their platform is slow and clunky and I think it might only be for CSTA+ members. If that's true, I don't like the exclusivity. Reddit's a much better platform technically but is also its own silo.

Oh for the days of usenet news. a set of cs.education.* newsgroups would be just perfect.

SIGCSE has some mailing lists and those are easy to set up but I just don't think another mailing list would gain traction.

To be consistent I want to point out that I'm not just down on Facebook. I'm also looking to de-google. Their capitulating on The Gulf of America so quickly might have been my last straw with them.

I'm heading to Savannah next week for a little vacation but when I get back I'll probably switch over to Fastmail.

Then for me there's also GitHub. I've already moved most of my git based work to GitLab but this blog still lives at github.io. When I get back from down south, I'll also probably move that to another domain.

Of course everyone has to keep their own concience but I do really think that as a community of educators, particularly one that claims that equity and diversity is important, we should be thinking long and hard about the platforms we use and the products we support.

Hitch, Groucho, and Cavett

Mon, 27 Jan 2025 14:54:01 -0400

I spent yesterday evening at the Leanord Nimoy Thalia theater. Once an art house cinema, I spent many hours there watching all sorts of movies - timeless classics, B films, cult movies, and cartoon nights. Last night, my buddy Ben hosted a movie night screening two Hitchcock classics - The 39 Steps and The Lady Vanishes. The evening also included a screening of that Wagnerian masterpiece, What's Opera Doc!

Now, this doesn't sound like a CS post and it isn't, but I said I'd expand my range of topics. Hope regular readers don't mind or maybe even enjoy it.

While I've seen a fair amount of Hitchcock, I hadn't seen either of these two before.

Besides enjoying the evening it got me thinking - does the upcoming generation even know Hitchcock.

This wasn't a new thought for me. A couple of weeks ago I was at a party and I was in a circle of younger people. Something came up and I mentioned that just the day before I was watching a special on PBS about Groucho Marx, specifically, his relationship with Dick Cavett, who interviewed him on television a number of times. Now, I figured these youngsters would have no idea as to who Dick Cavett was but they also had no clue about Groucho.

I was a little surprised. I mean the Marx Brothers, and Groucho in particular as well as Alfred Hitchcock were just ingrained into American culture. Even if you never saw a Marx Brothers movie you knew Groucho's walk, mannerisms, and some one liners and even now, I just searched "Groucho glasses" on line and of course, the iconic nose/glasses/moustache combo came up. Hitchcock as well. The "Goodevveenning," the iconic silhouette and of course him being the master of suspense. We all knew this before ever seeing a Hitchcock movie as well.

I mean, I get why entertainment icons prior to film and then television haven't left their mark. Sure, we have the works of the great composers and authors, but since there was no recorded media, we have no real knowledge of great singers, comedians, play directors and the like.

I remember a bunch of years ago learning about George Washington Lafayette Fox - a super popular, well, clown in the mid to late 1800s. There are some photographs but of course no film. I only learned about him by seeing Bill Irwin's amazing tribute, Mr. Fox a rumination. Of course, Bill Irwin's tribute, while a wonderful show was in some ways something of a research project.

Years ago I wondered if cultural influences of the film and TV generations and later would have more staying power. True, media today has so much of a greater reach than say in the 1950s or even when I grew up in the 70s and 80s but at least there's a visual record from the relatively early 20th century, lost films not withstanding and a visual and audible record since talkies.

The thing is, media consumption has also changed. When I grew up, we only a handful of channels on TV. Even when cable came in, the choices were limited. We also didn't have onscreen guides so unless we kept the NYTimes program guide around or got something like TV Guide magazine we'd have to flip through the channels. We also, for the most part didn't have remotes so changing the channel meant getting up, walking to the TV and clicking the dial - if you missed your channel you had to go all the way around again.

This meant we'd leave something on becuase it was easier, or we'd be flipping through the channels and land on something new and we'd just check it out.

That's how I discovered a lot of old movies - both good and bad and developed my tastes. From classics aired at random times to Kung Fu movies on weekends and Chiller theater.

Now, everything's on demand for shows and movies - no discovery, all active decisions and even worse (from my point of view) the short form stuff like TicTock and Reels.

So what does this mean for the future? I don't know. It would be a shame if classics and classic comedy were lost to the future. Sure, some is just of historic interest but the good stuff really holds up.

Here's a bit of classic Marx Brothers:

and part of one of Cavett's interviews with Hitch:

More of the interview can be found on Youtube. Actually, there are tons of Dick Cavett interviews on Youtube and I highly recommend all of them.

Okay, that's it for today. Probably back to CS and Tech next time out.

Blog Reboot

Sat, 25 Jan 2025 14:32:53 -0400

Time to reboot the blog.

Why? Because of my new tablet.

So, I got a new tablet a week or so ago. A Oneplus Pad 2. It's a 12" android tablet. I really only use it in the mornings in the winter and inclement weather. I use it with my Wahoo bike trainer both to run the trainer controls and to watch videos. In nice weather I usually run outside.

This particular tablet has a pretty cheap add on keyboard so I thought, if I get the keyboard, it could be a lightweight travel laptop replacement. It's already good for media consumption and it can also run ssh so I can connect to a remote computer and do more serious work as well as things like write blog posts when traveling.

So, I went to a DigitalOcean droplet - a Linux box in the cloud, cloned my blog and installed Hugo - my blog generator, and the go language.

It didn't work. It turns out that the custom theme I downloaded had issues with the current version of Hugo and the older version of Hugo I was running had issues running under the current version of Go. Serves me right for not keeping on top of things since I moved my blog to Hugo in 2018 and updated the theme in 2020.

So, it was time for a reboot. Still using Hugo but a new but similar theme. I linked to the older versions of my blog in the menu up top. I'll probably also clean things up over the next few weeks but already the new theme has the Bluesky link up top to replace Twitter which I ditched months ago.

Even though blogging seems to be dying I'll probably continue writing on CS with some Emacs mixed in and probably expand the range of topics a bit since now, in retirement I'm not so immersed in CS Education.

We'll see where things go from here.

About

Mon, 01 Jan 0001 00:00:00 +0000

Who's this Z guy?

Hi, I'm Mike Zamansky.

I'm a retired computer science teacher living in New York City considering what to do for my next chapter.

I finished out my career at Hunter College but spent most of it building the computer science program at Stuyvesant High School.

Prior to that I taught at Seward Park High School and before teaching, did computer work on Wall Street at Goldman Sachs.

At Hunter College I created New York's first operating state approved computer science teacher certification programs. By the time I left, we had trained and the state certified New York's first 100 licensed computer science teachers.

I also created Hunter's undergraduate compute science honors program. A scholarship program that allowed students to receive a great undergraduate computer science education with many advantages that students at the "elite" programs of the city don't enjoy at a fraction of the cost.

Prior to my tenure at Hunter, I had built Stuyvesant High School's computer science program, arguably the finest high school computer science program in the country.

I also love tabletop gaming, cooking, baking, biking and a variety of other things.

Hope you enjoy my ramblings.

If you want to connect, you can always email me at zamansky@gmail.com.