C'est la Z

Making mistakes and what makes a great teacher

Thu, 09 Jan 2025 15:32:50 -0400

As he usually does (and of which I'm always appreciative) JCS shared my most recent Emacs post in which I go over a small Elisp routine that I wrote.

JCS commented that he liked that I don't post process or edit my videos. I work through whatever I'm doing and if I make a mistake, so be it, I'll try to figure it out live. I do things this way partly because of laziness but also because I think the result is more authentic. People doing something similar on their own will likely make mistakes and if every video and resource they find online is perfect, it creates an unrealistic expectation.

It's like how the usual self reporting of AP scores in the CS communitie unfolded a bunch of times with the only people sharing their results were teachers who's students all did well. Given national scores, we know that not every student gets a great AP score but it created an unrealistic and demotivating picture for teachers in the community who's results were more, let's say mixed.

From very early on, I've been pretty comfortable making mistakes in front of my classes. I mean, once I got to Stuy, I regularly taught students much smarter than I was. I almost always knew more material but they could stump me at times and once they dove into the mathy side, I was way out of my league. I felt being honest and open was my best approach and making mistakes allowed me to model that mistakes are part of the process as well as how to proceed. Of course a challenge would be recreating the mistake that was organic in period two later in the day.

Anyway, since I think my willingness to make mistakes in front of my students - my honesty in that regard is part of what made me a strong teacher, JCS's comment made me think of Alfred Thompsons recent excellent post on what turns a good teacher into a great teacher.

Alfred focused on three things:

Relationships with students - they care about their students as people.
They had to be enthusiastic about the subject they teach.
They had to innovate - that is develop their classes over time.

It's a good list. Many great teachers do indeed have these there qualities. I think it's worth considering though that while these are generally positive teacher attributes, they're not absolutes.

I've met some great teachers and teachers who have been considered to be great who don't have these qualities.

While none of the great teachers I know have disdain for their students, some don't have any real relationship with them and they don't really care about their success outside of them wanting all their students to pass their class.

Now, I've built my career on relationships so I never built a wall between me and my students and I don't understand teachers that do but I do have to recognize that building relationships isn't an absolute for one to be a great teacher.

Same with subject enthusiasm. Sure, it probably helps, but again, I've met some great teachers who are more or less indifferent. What's important is that they have sufficient subject knowledge and it probably helps if they also do care about their students so maybe you need at least one from column A and one from column B.

Finally, innovation. I always worked to improve and refine my work and would probably have gone crazy teaching the same thing in the same way for three plus decades but again, I've seen some great teachers who do the same thing pretty much the same way year in and year out.

On the flip side, I've seen some lousy teachers who have all three of these great teacher qualities. They care, they know their stuff and love their subject area, and try new things but they're just not particularly good.

JCS's post made me think about this because for me, open honesty was helpful in being my strongest teacher self, whatever that was and is. I also listened the other day to the most recent CSEd Podcast which had a professor sharing a particular practice he uses in class. I'm also planning to write a post on that but it also made think on how a teachers pedagogical toolbox can also have an affect on how "great" a teacher one is.

I guess the moral of the story is that while we can pick out some common attributes that many great teachers share, great teaching is hard to pin down which is why it's so hard to meaningfully evaluate teachers. I guess it's like Justice Potter Stewart's concurring opinion on identifying "hard-core" pornography. We might not be able to define it but we know it when we see it.

Extracting Emails (with Emacs)

Tue, 07 Jan 2025 09:25:00 -0400

Yesterday I found myself in a situation where I had a text document interspersed with a bunch of email addresses and I wanted to extract those email addresses.

Specifically, I had to copy email addresses from a couple of spreadsheets and other sources and I thought it would be faster to copy lines form the sheets instead of email cells as well as just grabbing text from the other sources and then isolating the email addresses in Emacs.

Now, writing a little "extract emails" script to operate on a text file is easy enough. You can search for an email address using a regular expression. Regular expressions are a text pattern matching language that most programming language support. I used something like the following regular expression to identify an email address:

[a-zA-Z0-9\.]+@[a-zA-Z0-9]+\.[a-zA-Z]{3}

The first [a-zA-Z0-9\.]+ matches a sequence of one or more characters that are letters, numbers or a period (the slash before the period is to escape it - prevent using the . for its special regular expression meaning). That's followed by the @ symbol, a second set of characters, this time without the period, then a dot and then three more characters.

This will match most email addresses.

In Clojure, I'd use the function re-seq which would take a string (presumably my text file) and return a list of all the matches. That is, all the email addresses. I could do the same in Python with re.findall.

In an editor, at least in Emacs, it's a little trickier. You can search for a regular expression but that just finds the next email address. I'd then have to manually copy it and then repeat the process.

I guess a keyboard macro could do the trick but easier would be to just write an elisp function and extend Emacs.

The video shows the whole walkthrough but the core of the routine is using the search-forward-regexp function which searches for the next occurrence of an email match. I then add that email address to a list. Putting it all in a while loop, you get:

(while (search-forward-regexp regexp nil t 1)
 (push (match-string-no-properties 0) matches))

The match-string-no-properties grabs the actual text of the match.

There's a little more detail but basically I wrote an equivalent of re-seq for an Emacs buffer - it grabs everything that matches a regular expression and returns all the matches in a list.

Then, it was a simple matter of looping over that list and inserting the results. I ultimately named the routine extract-emails. So now, I can be in any Emacs buffer, copy over all my source email material, run the command, and I'll have all the emails together one per line ready to be pasted into an email client.

I'm thinking that I should probably give the routine an additional optional parameter - have it extract emails by default but allow the user to pass in any regular expression.

Just another reason why Emacs is such a powerful editor.

Here's the video, enjoy:

Advent of Code 2024

Thu, 26 Dec 2024 14:30:56 -0400

Happy holiday season to all.

Blogging wise, I usually write up a bunch of posts on Advent of Code - a 25 day programming event where each day a new two part challenge is released. It's been running since 2015 with this being the tenth year.

Last year I wrote up five days, the year before only two but for 2021 and 2020 I wrote up around 25 of them. This year, I once again partook in the festivities but I haven't felt motivated to write up posts about the days. I still might do some writeups in the comming weeks and months since, as always there were many interesting and fun problems.

Still, I thought I'd share at least one post with some overall thoughts.

This year, I managed to earn all 50 stars - that is both stars on each of the twenty five days. I've only done that once before - in 2020. Looking back, I average 35 stars each year with my highs of 50 twice (2020, 2024) and lows of 18 in 2017 and 23 in 2018. I might go back and start completing more problems. I actually started to finish more of the 2015 problems this past month so we'll see.

Usually, I start on day 1 and keep going until either life gets in the way, the problems get too annoying - you know the solution but just don't have the time or patience to implement whatever algorithm, or the problems get too hard, like some of the mathy problems where I feel I'm just way underwater. I also don't do these "closed book." I give them some thought and a shot closed book but I'm fine looking at the Advent of Code subreddit which is both active and friendly. This year, I solved most closed book but there were a few where I got some helpful hints and two problems where I got some pretty significant help from the reddit.

It's also worth noting that I really hadn't done any coding since last December. Take that for what it's worth.

So, here are some thoughts.

Usually I do AoC using Clojure - basically a Lisp. It's a fun language to work in and Cider, the Emacs based Clojure editing environment is second to none. This year, I solved some in Clojure, some in Python, and many in both. I was giving a short talk on AoC in mid December and knew that nobody in the audience would have any Clojure or any version of Lisp but many would have probably played with Python. So, I coded most problems in both. After my talk, I usually went mostly to Clojure but a few problems were easier in Python so I went the lazy way.

That was the interested part - each language lends itself to it's own approach to problem solving. Some AoC problems lend themselves to state munging - you've got variables representing your things in your world. You have to keep changing them and they keep changing each other. Since Python started as a quick and dirty scripting language, it has an easier time on those types of situations. Clojure's big thing is immutability - once you set a variable you don't change it.

On the other hand, Clojure, being functional lends itself to other problems.

After solving a problem, I'd frequently look at other people's solutions to try to learn something and there was frequently a stark difference between the Python coders approaches and the Clojure programmers. In some cases, it was hard to believe that they solved the same problem.

I also found it hard to context shift as I went from one language to another lest I end up writing Python code in Clojure or vice versa.

This made me think about language choices in education. I've written about choices for first programming languages before (here, here) and of course there are also considerations for what to do after that first choice.

This is something I almost certainly plan to write about in the coming weeks.

Another thing that I always like about AoC are range of problems. This year there were a bunch of 2D grid pathfinding problems, a couple of mathy ones though none that were too hard, one cellular automaton, a couple of simulations, and a couple of what I'll call, problems with a non programming twists.

Those last ones I usually find really interesting although the last one from this year - day 24 required me to brush up on things I hadn't looked at in decades and many programmers may never have seen.

Those are problems where it's not just "run the program and get the result." It's more of write a program to get some info, do some stuff by hand or by "thinking really hard," write some more code, back and forth and get an answer. I really like these types of problems because they come up so frequently in the real world. Maybe not for the professional software engineer but in the more general programming population. You might not be writing a complete program soup to nuts that you can then send to the app store but rather snippets and small programs to help you get some job done. Love it.

Will probably look into that more next year.

Another thing I've been thinking about are people's motivations. I just partake because they're interesting problems. I'll learn a bit, and it allows me to prove to myself that I haven't totally lost my programming chops. Interesting though, I don't do these types of programming problems during the rest of the year. Others though are out for the global leader board - they get the problem when it's released at midnight (Eastern time) an they're off. In the last couple of years though there's been a problem with people using AI to game the leader board. The issue and the range of reactions have been pretty interesting.

Of course on top of these there are the actual problems so maybe I'll write up some of them as well.

We'll see but at least if I get motivated to write more, I've got plenty of blog fodder.

Learning About Startups at StuyCS

Fri, 13 Dec 2024 15:12:24 -0400

Most students who study CS in college plan to go into industry in one manner or another. Unfortunately most college CS programs are still designed with the assumption that all undergraduate CS majors want to pursue a PhD. I've addressed this periodically here and that's not the focus for today.

Of those who want to enter industry, it seems a fair number are interested in the startup ecosystem - they want to either be part of a startup - a small agile effort that they hope can change the world or even start their own company. Some eventually also want to turn towards venture capital.

Now, if college does a less then ideal job preparing CS majors specifically for industry, they do less ideal job preparing them to launch a startup. Before going further, I don't want to be unfair - a number of institutions are making efforts to modernize their undergraduate CS programs and a number of them are doing great jobs - offering courses in software engineering, teaching the "hidden curriculum" and integrating modern software engineering sensibilities like unit testing into core CS classes but change in academia is slow.

On the startup side, things are even further behind. There are a number of boutique masters programs in "tech entrepreneurship" that have popped up but financial aid can be sparse so many are, like code schools, workforce programs for the rich. For many of the programs I know of, I'm not sold on the quality. Some are good, many are meh.

In undergrad, if you're a CS major you're not really going to learn about startups and the startup ecosystem. I supposed if you're a business major you might, but then you won't develop the tech skills.

Since retiring, I've started to work more closely with the Stuyvesant Alumni Association and I've always had a tight bond with my former students and I wanted to do something to start to address this problem.

Last night was the first step. It was only a single small firs step, but it's something.

In partnership with Stuy's Alumni Association, we put on an event. First I reached out to my grads and asked for both founders and funders. People who've started a company and successfully raised money for it and people who've selected promising startups and provided said funds.

Since it's the holiday season, a number of my New York VCs couldn't make it as they had other holiday events to attend but we had a solid dozen volunteers available and I'm sure I could have gotten more if I had worked at it.

Then we had the attendees. I thought the event would mostly appeal to younger Stuy grads. Those in college and a bit beyond but there was a pretty wide spread of ages and even a number of current Stuy students.

We split them up at tables, gave the funders and founders some loose guidelines to follow - key points to address and let the founders and funders lead small group conversations. We also had open mixing and networking before and after the small table discussions.

While I'm still collecting feedback, overall I think the event went really well. People got to learn about the startup journey, some specifics from specific stories, got advice from knowledgable people who've both been there and done that and walked away with some new connections.

It wasn't perfect. It would be better if we had more balance between founders and funders - we had 2/3 or 3/4 founders and a 50/50 mix would have been better. Reshuffling the tables would also have been a good idea but, overall a success.

One thing that I really love about the StuyCS community is that we can do these events on almost any tech topic and really, many other topics as well. Last year I wanted to pull together an AI panel and instantly had far too many volunteers for panelists. This time, founder and funders and again no problem.

Now the next step is to refine this event and figure out follow ups as well as other future events.

The New Entry Level

Mon, 09 Dec 2024 15:20:32 -0400

Over in Facebook's Computer Science Educators group there was an article posted on CS grads going back for a "panic" master's degree since the job market isn't terrific. Now, going for more education, at least for those who can afford it has been a tried and true practice in weaker job markets. If you look at my former students with Masters degrees you'll certainly notice clusters around bad times in tech hiring.

What this post is about though, is one of the comments. On of my CS Ed brethren commented on how companies are looking for more mid-level experience and abilities out of college and not entry level.

Now, I've long been one who's felt that we can do a better job preparing out students for industry from undergraduate CS programs. I don't want undergrad programs to become code schools - they still need to teach real CS - enough algorithms, data structure,s theory and all but there are some easy changes.

Use industry standard tools when appropriate - version control for example.
Teach things like unit testing.
Make sure students complete some large group projects, preferably prior to their senior year.
Use techniques like code review.

These and other techniques and tools can easily be integrated into a CS sequence without removing any of the good stuff.

That said, I don't think we should be bowing to industry if they demand more mid-level ready college graduates.

Let's look at what industry had been demanding from their college grads over the past twenty years. Even if we just limit our view to the "good times" when "everyone was hiring" the typical entry level hire had a CS degree along with one or two internships. How much more can undergrad institutions actually do?

Even back then, the Googles of the world largely recruited from "elite" institutions and when they said "entry level" they really meant that you went to Carnegie Mellon or MIT, taught yourself all the stuff from the unwritten curriculum, and had a couple of killer internships.

In any event, once you got a job, you then might get some amount of ramp up - an in company bootcamp if the company was large enough or it might just be a helpful co-worker at a smaller company. Now, those company bootcamps also seem to be disappearing or at least becoming shorter and shorter.

Companies deciding they want more 'mid-level' ready new hires is just the next step in companies abandoning any responsibility to the tech ecosystem in terms of job preparation.

I'm not surprised.

About ten years ago I was speaking with a friend who was the CTO of what I guess I'll call a small large company. Multiple offices, multiple cities, but nothing like Meta or Microsoft in terms of number of employees. We were talking about internships. I asked him what his rate of return offers was and he gave me a number that I thought was low. He commented that he just couldn't hire that many but felt it was the companies responsibility to the tech ecosystem to have a robust internship program where college students could gain valuable experience and get paid a good summer salary.

Since then, I've spoken to a number of people in the industry but unfortunately, only a few understand that having an education, robust, internship program should be a companies responsibility not only for hiring their own employees but also to help bolster the tech workforce in general. A few got it, a few talked the talk as though they got it but their actions, or rather their company's actions spoke otherwise and to far too many, the idea of having an internship program as anything other than a direct recruiting tool had never crossed their minds.

Of course, in the most recent tech crunch, internships dried up. Never mind that companies like Google weren't losing money, they were only making less than before - one of the first things cut was this important piece of the tech education pipeline accompanied by a request for 'better prepared more mid-level ready' candidates.

All this made me think back to when I got out of college. We didn't have a tech scene per se, but companies expected college grads to be raw. They trained them regardless of the job. I forget how long my training program at Goldman Sachs was but it was long. It covered the business side and the tech side. They taught the languages we were expected to use - generally C or Cobol and the platforms we were going to work on. Once we were in our groups, if we needed more training, we got it. I ended up doing Windows programming so my manager made sure I had whatever I needed to get up to speed.

Other companies had even longer and more thorough programs. Many also had rotational programs - you got your training then you did rotations - multiple weeks or months even with different groups to find the best fit.

Over the years, companies have abdicated more and more responsibility for employee training and society has decided that responsibility falls on the schools - both K12 and higher ed.

This is neither sustainable nor healthy which is why I feel push back against the "mid-level ready" employee right out of college is both appropriate and needed. Unfortunately corporate America holds a lot of influence so I'm not sure how effective any push back will be. Those grads will still need jobs and who's more likely to blink first - the educators or the corporate execs?

Evolving As A Teacher

Fri, 22 Nov 2024 15:18:28 -0400

Over the past couple of months I've had a number of interactions with former students. A fourth of July celebration with some of my first students, a bunch of dinners dinners and a couple of alumni events.

It got me thinking that while I was the same person and same teacher for all of them, I was also a different person and teacher.

We all evolve over time.

One particular conversation made me think. I was talking to one of my earlier but not earliest grads. Someone who graduated within my first decade my career - a number of years that based on my scale would place me as moving out form advanced beginner and into intermediate level as a teacher. Devorah and I were getting dinner with this former student, who I now consider a friend, and his partner. He was talking about the stress levels and impersonally and large schools like Stuyvesant and in fact in many large public schools. Something I agreed with. He also spoke about his college choice. He was and is a really smart guy and had strong grades but he didn't want to go to a school like CMU or MIT since he thought they would be the same type of big machine hyper competitive education complex that some people perceived Stuy to be and so opted for another top CS school that he felt would be more chill.

All of this got me thinking. At that stage of my career, I was still enamored with my students all getting in to top name schools. I was patting myself on the back when my students hyper achieved and I was designing experiences for the top of the class.

Now, I didn't know it, but I would evolve. A few year later, I would refocus on all students designing my intro course and pushing the city for CS Education for everyone. I also became more aware on the college front. Whereas I don't think I advised this student particularly well, if at all back when he was a senior but a few years later, I would probably have tried to get some small liberal arts schools with great CS onto his radar. I learned that you can do just as well at a Tufts as a CMU or similarly a Stony Brook or a Hunter college, where I'd later end up teaching. I learned it was about fit not about ranking.

Later, as I got older, I'd also evolve in terms of my perception of students needs. Back then, I don't think I did a particularly good job curating my class or my office suite for culture. Sure, we had a fun quirky tech culture going on but I'm talking about a comfort zone for students. Later in my career, I heard back from a number of former students who thanked me for making my classroom or my office suite a "safe space" for them. I didn't think I explicitly did anything special other than try to be a decent human being but I think it was also part of my evolution as a teacher. Something I didn't know to pay attention to in the early years but fortunately something changed at some point so I did better as I gained more experience.

All this is to say that on some levels I was always the same teacher but we do change for better or worse. It makes sense. When I started I was only a few years older than my students. In fact, I had a couple at Seward who were piratically my age. I could be viewed as practically a peer or an older brother. By the end, I was older than my students parents.

The funny thing is that while looking back I can clearly see how I changed as an educator, for years I had my students evaluate me anonymously and my ratings decade in and decade out were pretty consistent. At the same time, my results, however you measure them, also remained pretty much the same as did, at least to my perception, my rapport with students and former students.

All of this is neither here nor there but it's something to think about and I encourage long term educators to look at their own careers and see how they've evolved.

Blue Sky

Mon, 18 Nov 2024 11:23:53 -0400

Blue skies smiling at me Nothing but blue skies do I see

Last week, Bluesky seemed to really take off. What's Bluesky? The simplest answer is that it's another possible Twitter, or do I have to say X, alternative.

Like many others, I left Twitter shortly after Musk took over. Actually, I just stopped using it. I didn't actually deactivate my account until relatively recently but other than logging on to do that, I hadn't been on Twitter since late 2022 or so.

While Twitter has certainly gone to pot, prior to Musk's takeover, it filled a particular niche in social media. Facebook was for people you knew. Instagram for photos. Linkedin for jobs so not much general social discourse going on there and then there were other things like Tumblr, Whatsapp, Pinterest, Snapchat and more. I'm not listing Reddit since I view that more as a modern day Usenet replacement and things like Discord and Slack are also separate beasts.

Twitter was more discoverable and open than Faceook - you didn't have to be "friends" to read or partake in a conversation and as a result, there was a different type of discourse. It also was more suitable for live conversations, similar to Slack or Discord based on hashtags.

Twitter was also good for aggregating news, or at least types of news (before things got bad).

Finally, Twitter became in many instances, at least for me, the last line of customer support. When you couldn't resolve a customer issue with a company in spite of hours on hold and speaking with people in call centers from around the world, a well placed tweet would often get results.

Post the Musk acquisition, Twitter went south fast and up until now, nothing has really filled the void.

Facebook continued to have its Facebook problems - closed silo, not very discoverable, and of course, you're subject to the whims of a different billionaire owner.

Linedkin? As I said, that's for jobs although I did notice an uptick in traffic on my feeds post the first Twitter exodus right after the takeover.

A few contenders have popped up. An early one was Mastodon. I really wanted to like Mastodon and I do. From a user point of view, it looks and acts similar to Twitter but it's open source and can be self hosted. Both big pluses for me. It also uses an open protocol - federated which, in theory lets it play nicely with other federated platforms.

The trouble was and is that because it's distributed you have an account on a particular server. Sure, you can follow people on other servers but finding them and dealing with the whole distributed thing is just a little to much for the non techie user. So, in the end, it seems to be ending up as small closed twitter like groups of technical people. I'm on Mastodon as @zamansky@csed.social and will probably stay on the platform but I haven't seen much life there.

Another one that had promise or at least backing has been Threads which is Meta's answer to Twitter. Again, the look is similar to Twitter but it has the huge advantage of preseeding your account with your contacts and everyone on Facebook could get on with a click. It was trivial to hit the ground running with Threads whereas on Mastodon you had to build your network one account at a time.

On the downside, it really just feels like a more general Facebook feed. Lots of posting, little conversation (at least from what I can see) and again, another platform controlled by a different Billionaire.

Bluesky has been another contender and the one showing the most promise. It grew out of twitter but has spun off and is it's own public benefit corporation - no affiliation with Twitter or Musk. Like Mastodon, it's distributed, uses an open protocol and has open sourced it's apps. Unlike Mastodon, it uses a single entry point so you don't have to figure out who's on which server.

I got an invite back in June 2023 - shortly after Blusky's launch but things were pretty quiet. I checked back periodically but never noticed much growth.

That all changed about a week ago. Suddenly Bluesky became very hot and tons of old Twitter contacts started appearing. Bluesky also seems to be kickstarting the onboarding and finding people problem through the use of what they call [[https://blueskydirectory.com/starter-packs/all ][Starter Packs]]. I'm also starting to see news outlets and companies on the platform.

Will Bluesky ultimately be the Twitter replacement people have been looking for? I don't know but it's the most promising one so far.

So, if you haven't yet, check it out. I'm there as @zamansky.bsky.social and if you know some great people to follow or any great lists in the CS Ed space or NY Tech scene, let me know.

Back to Amsterdam

Tue, 05 Nov 2024 10:10:17 -0400

We finished up the trip with a few days in Amsterdam. Sitting in our hotel now writing this post and trying not to stress out over the elections until it's time for dinner.

Getting to town was a little hectic. Turns out there was a train with some mechanical issues on the line so two thirds of the trains into Amsterdam from Leiden were crowded. It only slowed us down by about a half hour but it was a little hectic and much more crowded than I would have liked.

Since we had most of the day, after dropping our bags at the hotel we explored one of Amsterdam's street markets. The Ten Katemarkt in particular.

Of course, Devorah wasted no time in finding yarn and fabric stores.

A Dutch specialty appears to be Speculoos cookies - think Biscoff cookies you get on planes. One particular version is the Gevulde Speculaas which is an almond paste filled version. We got a particularly tasty one at the market.

We spent the rest of the day wandering through the city, through Voldelpark and eventually getting back to our hotel.

Had dinner at the Blin Queen which is a pancake restaurant by day and a really nice dinner restaurant in the evenings though we did get a great pancake there for dessert.

Next day was another walking tour. It was a bit of a disappointment. It randomly turned out to be the same tour guide as the one Batya and I toured with last year and then he was great. This time, not so much. I'd like to attribute it to just a bad day. In the afternoon we hit the Albert Cyupmarkt - Amsterdam's largest street market. Also great stuff there. I think visiting the two markets were the highlights of the trip.

To close the day we found Amsterdam's Beguinage:

Though here it's called a Begijnhof. We noticed it on Google maps and decided to check it out.

Next up, museum day. First back to the Rijksmuseum which is truly amazing and then an afternoon canal boat tour with Those Dam Boat Guys which was fun.

Between the two activities we went to The Carrousel Pancake House - recommended last year by a friend.

That brought us to yesterday morning when we visited the Ann Frank House. We also did the introductory program. Both were very impactful. Walking though the house and going up to the secret annex you can't help but get emotional.

We also ended up randomly finding our way to the Netherlands National Holocaust Memorial which was also moving.

And finished out the day in the Amsterdam Museum.

Today's our last day in Amsterdam - we're flying home tomorrow morning. Our first visit today was the Straat museum.

A really cool place. We had to take a ferry from Amsterdam Centraal to get there but it was a nice ride and free. The museum is dedicated to street art and you just get consumed being surrounded by all these giant pieces. Really great museum.

Last up was the H'Art museum for an exhibition of Kadinsky's work on loan from the Pompidou Center. Lots of great works but what got me were his earlier non abstract works. I had no idea he started with stuff like that.

Now all that's left is a dinner, a breakfast and a long flight. It's been a great trip but I'm looking forward to getting home.

Antwerp Rotterdam Leiden

Thu, 31 Oct 2024 14:35:10 -0400

Before heading off to the Netherlands we had one more day in Belgium. In Antwerp Specifically.

We stayed one night right across from the railway station which is particularly magnificent.

First stop, the Plantin Moretus Museum. This museum focuses on two 16th century printers - Christophe Plantin and Jan Moretus. The museum, which is based in their homes and printing business covers the history of the family, the evolution of their printing business and the industry in general as well as the development of technology. It also focused on some of their primary industries - religious texts, science, and maps. Really amazing place.

After lunch we visited the Red Star Line museum.

This was basically an immigrant experience museum. We've gone to tons of these in the states but it was pretty interesting to see learn about the story from the other side of the Atlantic where so many of the journeys originated.

The next morning we were off to Rotterdam.

We spent the morning at hte Boijmans art depot. They store thousands of works of art and many are on display but there are no descriptive plaques or anything, just the art.

In one room we saw master after master - Money, Van Gogh, Bosch, Rembrant and so many more one after another. Also, tons of more modern works.

A really amazing place.

Between the two we got lunch in their Markthal - a neat food hall set up with an archaeological exhibit. A couple of blocks away outside they also had an urban surfing pool.

Next was Leiden which was mostly a Devorah day.

We did hit the Windmill museum which was pretty cool. We got to climb up a working windmill, learn about it and also the local history.

We also hit a number of fabric and textile places starting with the Textile Research Center where they had an exhibit Shisha embroidery and also showed us a bit of their archives in the back

We also hit the Weaver House - a house from the 1500s still preserved. It housed poor families - mostly weavers throughout its history. The walls had a list of all the inhabitants.

We learned of the house's history and saw weaving demo. One of the docent's recommended we then visit the headquarters of the Leidse Deken - a group working on raising awareness of the history of Leiden's artisinal blanket making industry as well as bringing it back. We wandered in, went up the stairs, and introduced ourselves. We got the grand tour.

They told us about their mission, how they created the blankets from shorn sheep to end product and showed us around. Lots of cool looms, spinning wheels, and even a weaving machine that was pedal powered.

Our one additional stop was at the American Pilgrims Museum which was interesting. It was actually a split house museum a house that has been preserved through the ages. Half was as it was in the thirteen hundreds and the other half, in the sixteen and seventeen hundreds. There wasn't any evidence that any Pilgrims that would later sail to America actually lived there but the house was typical of the period when the Pilgrims were living in Leiden en route to America.

Overall a very interesting day.

We finished things off with an Indonesian meal. Tomorrow we're off to Amsterdam for the last leg of our trip.

Belgium 2024

Tue, 29 Oct 2024 02:40:17 -0400

After Paris, we spent a few days in Belgium. We have a friend there - Helen, who teaches at the International School. Helen and Devorah taught together at the Manhattan Center for Science and Math back in the day and while I got to see Helen on my last visit here with Batya last year, Devorah hasn't seen her for almost a decade.

This was more of a visiting part of our trip than a run around like chickens without heads touring part but we still got in some good stuff.

Helen lives in the Laeken suberb of Brussels so we spent time just wandering there.

Had dinner at the Atomium - built for the 1958 world's fair. The top sphere houses a restaurant while the others have various exhibits.

The restaurant's way up to so we got some cool views. Not at dinner mind you - we were on the upper level of the top sphere which doesn't have views but we were able to go down to the level with windows and walk around.

The meal was tasty but the restaurant was uncomfortably warm and apparently half the waitstaff called in sick so service wasn't great. There was only one guy handling our entire level and he was running around frantically all evening trying to keep up. Felt bad for him.x

We did also, though go to the inner city and do a walking tour.

Next day we explored Anderlecht, another essentially, borough of Brussels. We went to see the Beguinage there. Basically a housing complex for lay religious women.

The Beguinage was closed but we did get to a library that was exhibiting a local artist's watercolor work and Devorah found a fabric store.

The highlight was the Erasmus house. Completed in 1515, Desiderius Erasmus Roterodamus, or Erasmus of Rotterdam lived there briefly and the building is now a museum dedicated to his work. Really interesting stuff.

Final day in Belgium was a visit to Brugges.

The reason for the visit was to see the lace center.

Devorah used to make lace - mostly bobbin lace before we had kids but lace making and small children do not mix so she stopped. Now with the kids grown she plans to start again and add to her set of hobbies - sewing, knitting, spinning, and weaving. She's wanted to check out the lace center for years.

Very cool place and far enough outside the touristy areas so that we got a feel for the real town.

Of course, we also did the tour and played tourist as well.

As a bonus, on the tour we went in to the Brugges Beguinage so we got to learn a bit more about that and them.

Between the tour and the lace center we also sat down for waffles.

Savory for lunch and sweet for dessert but the latter vanished before we got pictures.

This was the more relaxed part of our trip - more visiting than touring but still got to see the sights.

We're off to Antwerp today to finish out Belgium and then to the Netherlands.

Paris 2024

Sat, 26 Oct 2024 02:02:42 -0400

Devorah and I are finally on our first overseas retirement trip. Our hope was to make up for lost travel time in retirement. We've already gotten away a bit - a few days in Charlseton SC and a couple of local driving trips but this is our first retirement biggie.

Started out in Paris. Left Monday evening and got in Tuesday at around 6:00am local time. Went right to the hotel. Stayed at the Hotel Marignan on the Rue de Sommerard - a very reasonably priced hotel in the Latin Quarter. Address given to differentiate from the much more expensive Hotel Marignan on the Chams-Elysees. We took the train into the city and rolled up to the hotel before 9:00am. Our host at the hotel, Bridgette, was terrific. Took our luggage, signed us in, asked if we wanted breakfast and once we were settled and caffeinated, gave us the lay of the land and assured us our room would be ready by noon (which it was). Can't recommend the hotel highly enough. Nothing fancy but clean with a great staff. We opted for a room with a toilet and shower but if you're comfortable with shared facilities the rates are even better. I'll only add that the hotel has six floors - that's seven flights of stairs since they zero index and there's no elevator. Not a big deal for us but be aware.

Even though we had just landed, once we dropped our bags and had breakfast we went forth and had a full day - a couple of museums, over 10 miles walking, and some good food.

Our first stop was the Rodin Museum and garden - that is the museum in Paris and not his former home on the outskirts. We wanted to hit it since we just recently visited the Rodin museum in Philadelphia. The Paris museum is much larger. Many of the same pieces - each have their "Gates of Hell," Assorted Burghers of Callais, Balzac, and others but the Paris museum had more, also included paintings and other media and was much more spacious and expansive and since Devorah enjoys sculpture more than paintings, which I prefer, this was time well spent.

After lunch we walked across the Siene and visited le Petit Palais. Great collection plus it's free. After that we made our way back to the hotel, rested up, got dinner then crashed for the night. We ended up walking well over 20,000 steps, about 11 miles give or take.

We were going to start the day with a tour but that wasn't going to start until 11:00 so we had time to kill. We tarted with a critical pilgrimage. We had to head over to the Sorbonne to view le Tour Zamansky - yep, there's a tower named after a guy named Zamansky - perhaps a relative of mine.

Marc Zamansky was the Dean of the Faculty of Sciences at the Sorbonne up until 1970. He was also apparently a resistance fighter during World War II. When we entered the campus, the security guard indicated that we were not allowed to take pictures. I asked if he spoke English and he said "non." I tried to explain in my really weak french that I was a Zamansky and I showed him my ID. He laughed out loud and ushered us in and even offered to take the picture. It was a nice moment all around.

We then wandered the Luxembourg Gardens until tour time.

We usually try to do a free (tip only) walking tour whenever we visit a new city and since we haven't been to Paris for 36 years, it was basically new. I think we ultimately pay the same as if we do a normal pay tour but these have worked for us. The tour was great and informative and gave us the lay of the land.

For lunch we decided just to get sandwiches - cheaper and lighter than a sit down lunch and the bread in Paris is amazing. Truth be told I would have been happy just eating bread the whole trip. They also do sandwiches right - not tons of meat and gobs of mayo. It's a slice of ham, maybe a slice of cheese and a little butter on amazing bread.

Afterwards was the Muse D'Orsay. Here's the elephant out front.

Inside we found more elephants

But no Krafayis:

Really amazing museum.

For dinner we ended up at a restaurant with Basque options and I had a great cassoulet:

We went a bit further afield on Thursday, our last full day in Paris. We visited the Muse Marmottan de Monet. A converted house museum that boasts a tremendous collection of Monet's work as well as other impressionists.

They also had an exhibit on Trompe L'Oeil - art with highly realistic depictions of 3D space. Really cool.

Great museum - I highly recommend and it was much less crowded than say the Muse D'Orsay. Afterwards we slowly made our way back to the hotel.

If course we found the yarn store for Devorah:

And took a selfie as we passed the Eiffel Tower (but didn't go up):

The plan was to go to the Muse D'Orangerie which also boasts a strong collection of Impressionists and Post Impressionists but the line was too long and it looked too crowded so instead we went back to the Sorbonne and to a small Marie Curie museum. It's a two room museum plus you can look in on her lab and office. Really cool. Also 100% recommend to fellow science nerds. As a bonus, it was free.

Finally on Friday we decided to visit the Cluny Museum - just a couple of blocks from our hotel. It's a mansion museum boasting a tremendous collection of art from the middle ages.

About a third of the museum was shut at the time of our visit but what we saw was amazing. We'll certainly be back to see the rest.

After another sandwich lunch, this time in the Luxembourg gardens, we stopped by at the hotel and then off to Gare de Nord - the train station to take the Eurostar to Brussels.

That's where I am now writing this post. It's Saturday afternoon. We're spending a few days in Brussels staying with a friend so it'll be more visiting and less frantic touring. That said, we did a walking tour this morning and have a couple of things lined up for tomorrow and then Monday we'll be heading to Bruges for a day trip, mostly because Devorah wants to check out the lace center there.

Great trip so far. Looking forward to the rest.

Authentic Languages

Fri, 11 Oct 2024 12:05:37 -0400

Language wars in tech are common. Java vs C++, Functional vs OOP. If you're doing functional is your language functional enough?

So to in CS Education. What's the best language for CS0? For CS1? Drag and Drop or text based? Functional? Object Oriented? Compiled? Interpreted?

The battles rage on.

One particular "war" that I've been a part of deals with the idea of a language's being authentic or real. I've been a part of this on two fronts.

I guess most visibly this has been with Scratch and other drag and drop languages. People sometimes perceive me as being anti drag and drop. This is not the case. I love drag and drop languages for the earlier grades. My issue is that I haven't seen an advantage to using them in the later grades. I haven't seen evidence that knowledge transfers when an older student - say high school 10th grader or beyond starts with a Drag and Drop language and then switches to text based. I've also called out what I see as disadvantages of drag and drop languages in the later grades - what I call jigsaw puzzle debugging or programming where students write and debug programs by seeing which block fits geometrically as opposed to thinking through the problems and also the small screens students usually use are a limiting factor when writing complex, large programs.

I never thought that drag and drop languages like Scratch were inauthentic or less real than, say Java.

That said, there are those who do. I think it comes mostly from parents and students who feel "they don't use it in the real world." Probably also from hardcore programmers who, even though scratch is Turing Complete, feel that it's less of a language.

I've never taught scratch in a formal class so I never had to deal with any push back. I've only used it in one offs like teacher professional developments or working with students for some side event or project.

On the other hand, I've heard the same complaints when teaching languages like Racket (nee Scheme) and NetLogo. They're not real languages. Nobody uses them in the real world. I'd particularly hear it about NetLogo which is a wonderful language, particularly for visual modelling and simulation.

Never mind that languages don't have to be used in the real world in order to be real and to be of tremendous value in the classroom. Racket is wonderful for teaching functional programming. It's also a super small language with a simple consistent syntax which makes it, when taught right, beginner friendly. NetLogo is interactive, visual and allows for students with little experience to explore complex concepts. In my classes at Stuy, we wrote a maze solver that used implemented a cellular automaton and also explored predator prey models all during one unit in our CS0 class.

Even if a lesser known language, something seen (incorrectly) as inauthentic can truly have an impact.

I saw that just the other day.

I found out that one of my former students was on the team that just won the Nobel Prize in Chemistry. Just seeing the news made my week.

I of course sent a congratulations. As part of his response, he wrote:

You know, I can trace a lot of my interest in computational science back to playing with NetLogo turtles simulating things, exploring cellular automata rules, and graphics.

All the stuff we did in Stuyvesant's required intro CS class.

I don't know if the intro course as a whole made a difference in his life and career - he might have gone down the CS path regardless. It is clear though that using NetLogo in the intro class did ignite a spark. Had we used say Java or C++ from the get go. maybe this never happens.

The whole line about inauthentic languages is really hogwash anyway. Sure, some teachers use the wrong language the wrong way at the wrong time but most languages can be the "right one" under the right circumstances.

In this case, it was NetLogo at Stuyvesant in tenth grade in our required CS class.

Community Across The Years

Tue, 08 Oct 2024 14:02:09 -0400

I spent last week in San Francisco. I didn't leave my heart there but the entire trip was indeed heartwarming.

I was there with members of the Stuyvesant Alumni Association for a StuyCS/Tech meetup. As one might imagine, many tech inclined graduates from any school end up in the Bay Area and Stuy is no exception. Sure, the majority of StuyCS grads stay in New York and help drive New York's tech sector but plenty decid to "go west."

We had a terrific meetup on Thursday with around 125 attendees - mostly my former students but a few older tech alums from the early 90s, 80s, and even earlier. The meetup was amazing and it meant a lot to me to see so many friends, friends who I don't get to see nearly often enough. Of course, the problem is that it was something like being a parent hosting a Bar Mitzvah - since you know everyone there, you don't have a chance to spend quality time with anyone - it's mostly drive by greetings and at most a minute or two with any given person at a time. Still, I had a great time. Around the event, I was able to have a dinners with just a few former students but of course, I didn't have enough days to have the number of small group gatherings as I would have liked.

This got me thinking about communities and alumni communities in particular. After a career in teaching that lasted 33 years, the StuyCS family might be the single thing I'm proudest of. Interestingly enough, there's nothing else like it at Stuy. I'm wondering if there's anything like it from other schools.

Like any school, each graduating class is a coherent group - some will be more active, some less but that's the typical unit - the class of XX. They all have the usual five year reunions and maybe other periodic events. Not much crossover though, if any, between the years.

StuyCS is different. It's a network of people who took the StuyCS classes regardless of whether they went into tech or not spanning the years. It also includes older Stuy graduates who are in tech but graduated prior to my starting Stuy's CS program and Stuy grads who found their way to tech after graduating. As a result, we also have lawyers, doctors, authors and more as part of a community who's graduation years start in the 70s and continue through this past year.

At Stuy, leaving CS aside, there are two other groups that I would have figured would have a similar network - Stuy's Chorus - a program that accommodates a couple of hundred kids a year and spans all four years at Stuy. There are chorus classes and they also do special events in addition to the usual at school concerts. The other is Stuy's Speech and Debate program. Smaller, but still spans the grades. I believe there's at least one S&D class and there are after school meetings and practices and plenty of events and trips. For most of the past three decades, each was led by a top notch teacher with a strong personality - someone who could really hold things together and drive the programs - far more together than I ever was. Stuy has a few other sizable communities like the theater community or maybe one of the sports teams but these two would be the one's that I would expect would have a strong, active, coherent alumni network.

Surprisingly, at least to me, Stuy Chorus alums might be connected to their fellow chorus members but not, for the most part, spanning the years. They feel an allegiance to Stuy's chorus but mostly if not entirely withing their class. Speech and debate seems similar, at least to me based on my discussions with Speech and Debate grads.

At the meetup, a few StuyCS members noted that we're the most active and coherent Stuy alumni group. Active meaning connected with each other and willing and able to help each other across the years. We wondered why, particularly since we're probably the most socially awkward group of grads.

It was posited that yeah, we're socially awkward but we're all socially awkward so somehow we fit together and it works. This sounds plausible.

Maybe it's because we have critical mass in a couple of locations - NY and San Fran. Maybe, but then NY has critical mass for any Stuy alumni group.

Another possibility is that so many StuyCS people went into the tech field so there's a direct professional commonality and a pretty direct way people across the years can help each other. Maybe. Not many chorus members are professional singers.

I'm not really sure.

We're also an interesting group because we're not affiliated with Stuy's alumni association. We grew organically. In the past, under earlier Alumni Administrations, I actually avoided working with them but now, the current crew is top notch and I'm hoping to bring my community and the official Alumni group closer together.

As I mentioned, I'm really proud of this community. I'm wondering if other schools have had similar groups develop. If so, that's awesome. If not, well, it's never too late.

Enemy Of The Good

Wed, 25 Sep 2024 11:14:27 -0400

I usually listen to podcasts on my morning run. For the past couple of weeks I've been listening to 99% Invisible's series on The Power Broker, Robert Caro's incredible biography of Robert Moses. Arguably the best book on New York City not to mention urban planning and politics. Amazing book. I read it a few years ago, well, actually listened to it at the same time as Natan - our own personal father son book club. We've done it with a few tomes - The Power Broker, Don Quixote, and most recently Wagnerism.

I highly recommend both the book and the podcast series but I'm not going to talk about either really today. As part of the series, each episode includes an interview with a guest. One guest was Michael Schur, co-creator of Parks and Rec. During the podcast, he told a story about how, when working on Parks and Rec, they interviewed a city bureaucrat or civil servant. He wanted to rebuild a small local park. He had all sorts of plans to return the park to its original glory including magnificent columns. Schur noted that since we live in a democracy, he couldn't just start building but rather all sorts of constituents had to weigh in. One group wanted less grass, another objected to the ornateness of the columns and so it went on and on. By the time the park was actually finished, it was a pale imitation of the original plans. Sure, it was a new, rebuilt park, but the plans had been gutted, scaled back and otherwise twisted.

In spite of this, when asked, the civil servant said he was happy with the results. Sure, he wanted his plans but he also knew how government worked and in the end, he as happy that there was some progress.

I see the same thing happen time and time again. Traffic calming efforts that would improve safety, air quality, and quality of life in general for many New Yorkers shot down because a small group of people (usually rich white ones) insist that the city can't remove the free public parking in front of their buildings. I won't even get started on the housing crisis in the city. If following our cited civil servant's point of view, the people pushing the positive changes in the city should be happy with their baby step successes.

I've always had difficulty with this point of view.

I'm not saying it has to be "go big or go home" but I've always tried to do it right rather than do it enough to check the checkbox.

I remember when New York City was creating AFSE (The Academy for Software Engineering). I was involved for a while until the Department of Ed froze me out. The DOE insisted that the goal of the school was to create industry ready software engineers right out of high school based on what I had built at Stuy. I said "great, but to do that you need to have an academic screen to get in" knowing that you can't get to industry ready if the kids are coming in multiple years behind, particularly in math. I also said that they could go without a screen but they'd have to adjust their expectations (as well as messaging) since without the screen and with students coming in at all levels, the best they would be able to do is give some exposure to computer science and if they did it right, prepare them for college CS or a boot camp after high school.

I said they could go either way but I was pushing for the entrance screen. Not anything like the Stuy test, but something to make sure the kids entered with enough of a math background so that the school could accomplish its goals. I really believed with a modest math screen and the right resources, the school could be something special.

At the end of the day, I lost that one. The DOE froze me out. The created an unscreened school (again, I'm not saying this in and of itself was bad) but they maintained the messaging that the kids would get that top level software engineering preparation.

Of course the students didn't. The school was fine as a generic school but in terms of Software Engineering and CS Prep, looking back, I'd actually say that there were public NYC schools that already existed that were doing as well or arguably better and I'm not talking about test schools like Stuy here.

In discussing this was friends, I was commonly told "don't let the perfect be the enemy of the good." I get it but it's something that I've long struggled with.

At the end of the day, I think I was right about AFSE for the students that went there. If the school was honest about what it was, I think it could have served its kids better. Alternatively, if it added a nominal screen it could have fulfilled its stated objective.

On the other hand, I was also wrong. The schools existence along with the Public Relations push and industry hype, all under the Bloomberg administration probably helped pave the way for Bill de Blasio's CSforAll initiative in New York City.

So, was I right or was I wrong? I don't know. Without AFSE we might not have had CSforAll so I was most likely wrong but part of me still believes that they could have done the school right and at the same time still built the hype.

I've also been similarly on both CSforAll and CS teacher prep programs. My position has always been that we need knowledgeable CS teachers in CS classes and that knowledgeable means actually knowing a fair amount of CS. Of course this would mean a long term ramp up and funding to actually pay to prepare the teachers. The majority trend and view is quite different - start teaching CS now even if teachers don't really know what they're doing and we might work things out at some later point.

I'm not going to address the people who feel that short form PD is enough to make a strong high school CS teacher other than to say they're flat out wrong.

My view is wrong in that the ramp up would take years - probably decades. That means that potentially another generation of kids would be denied any CS education. On top of that, the amount of time needed would be far longer than typical government office terms and who knows if initiatives would survive from one administration to another.

On the other hand, doing it quick and dirty can also be wrong. I've seen both teachers and schools who believe that once they have someone in an APCS-P class - regardless of that teacher's ability to actually teach CS - as long as they have the seal of approval of having gone through some rudimentary code.org style training, it's good enough and the school is set for life.

This is also a problem.

An even worse problem is when a school is teaching CS badly and either turning off students or teaching them things that are just wrong resulting in kids getting turned off later on. I've seen this as well.

So, I'm still muddled on this. I find it very frustrating when I see CS education at the school, district, city, state, or national level moving in a direction that I think is, not wrong, but not optimal even though I try to console myself with the fact that the movement is positive in a lot of ways.

I also remind myself of something Pete Seeger frequently said, attributing it to Rene Dubos - "think globally, act locally." I take great pride in the results I've gotten when I have been able to do projects my way with my team - Hunter, Stuyvesant, and an assortment of others. That reminds me of something that a very smart friend once told me years ago when the K12 CS movement was just starting. He predicted that "there will be small pockets of excellence" - referring to places such as Stuy but the mainstream will be mediocre at best. I think the mainstream overall is doing a fair deal better than mediocre but I think his point of view was pretty prescient.

In any case, I do wear my naysayer hat as a badge of honor. I think the movement needs people like me to try to keep the mainstream honest.

StuyCS - Playing the long game

Tue, 17 Sep 2024 11:49:58 -0400

Last Thursday, I was back at Stuy. I was there along with members of the Alumni Association to host Stuy's first tech meetup of the season. I guess that's the first "official" project I'm working on since retirement. In the past, there had been various issues with Stuy's assorted alumni associations (yes, there were at one time three competing entities) but for the past few years they've been under what we can call "new management" - people I both like and trust so I'm happy to be working with them.

This event was just a normal meetup - mostly networking time and a couple of speakers. One of the speakers was my friend, former student, former colleague, and successor as Coordinator of CS at Stuy - JonAlf. JonAlf gave an update on the "state of StuyCS" going back to when I first started to build things. I enjoyed the talk and since, as JonAlf said, even though the audience collectively knew the development, individually very few did. There were also a couple of old timers who probably didn't know any of it.

It made me think of how much education is playing the long game.

Things started in Fall 1994 when I started teaching APCS-AB - the one that was both a college CS1 and CS2 - four sections. Prior to that there had been a single APCS class and a random programming elective here or there but all taught by math teachers and all taught as one offs.

Spring '95 brought graphics and Fall '95, Systems.

At this point, we, or I was just starting but I wouldn't call it a program. Three courses but I'm not sure I'd be able to claim any strong coherence.

APCS was, well, APCS. Nothing special there except that we probably covered more and more deeply than most high school programs.

Graphics was the terminal course - second term senior year. It had a nice combination of the programming they had learned in APCS, Math from precalc and a touch of calc, and a little of their physics. Much like the rug, it tied things together nicely. That, plus the kids made pretty pictures which helped stave off but didn't eliminate senioritis.

Systems was just thrown in. It was based around Intel Assembly programming in our lab of MS-DOS/Windows-95 machines I thought it might be interesting and fun for the kids and I knew I could teach it but there wasn't any deep thought that went into why I created it as opposed to some other course to fill that space.

Around 1998 / 1999 things started to change. First, Systems had changed. We had switched the labs over from the extremely overpriced and obsolete IBM PS/2s (obsolete because IBM had stopped selling them prior to selling them to Stuy) for regular PCs which we could install Linux on. Since we could do much more with the platform I decided to change up the course. I also had a few years of graduates now in college and was able to get their thoughts on what was missing from their CS sequences.

That's how our new Systems class was born. It covered C programming and an assortment of Systems topics. Some was similar to what they'd cover in college but a lot was set up to support them in college. I continued to get feedback from my grads on this and other courses from that point on and it helped direct my approach to course building.

I also created the sophomore one semester intro course that would later become a requirement. This class was designed as a new CS0 - something that would be programming heavy but would both expose students to a wider range of CS topics than a typical intro programming class and also relate CS to other disciplines. In some ways APCS-Principles which came around later had a similar intent but I'd say my course was much more rigorous and much better for my population.

The goal would be to provide something valuable to the majority of students who might never take any more CS while at the same time, inspiring and preparing those who would.

Now I'd say we had a program. It was still basically just me along with a math teacher and then two picking up some of the APCS classes but we had a cohesive sequence of courses that made sense.

Intro would lead into APCS would lead into Systems/Graphics.

Being a StuyCS grad meant something.

A few other selective schools also claimed CS programs at the time but as far as I could tell they had intro programming followed by APCS followed by just teaching the next college course or courses in high school.

I didn't see the point of that. The kids would get algorithms, theory, compilers etc. in college anyway.

For me, Systems and Graphics made much more sense. My Systems was set up to support the college classes and Graphics was typically an elective so it would add to my kid's backgrounds and not just be something they'd take again. Besides, I ended up working some things like parsing and automata into the graphics course.

We also started to grow.

By 2007 we had 3 full time CS teachers plus five sections of CS classes being taught by math teachers. Intro was a single semester requirement. We had 12 intro classes each semester, 5 ACPS classes (still teaching AB), 2 Systems/Graphics (Fall/Spring), plus one research which was really a fake class to let me work with my kids to run the network although I did do my best to support the kids who did want to do legit research and within a few years, the class did morph into a legit research class.

That's a little over 10 years. The long game.

We gad a coherent three year sequence. It wasn't random electives. Being a StuyCS grad meant something. Now, I'm not saying that my way was the only way or the right way for everyone but I think for CS it made a world of sense. Contrast that with Stuy's English department. Through most of your four years, students could select which English "selective" to take and they ranged wildly. You could take Science Fiction, Women's Voices, Children's book authoring, science writing, detective fiction and tons more. This means that Stuy grads can have radically different English and Literature experiences. For English. maybe that's good. For a HS CS program, I think a coherent sequence is better. I guess it's also like having area requirements for a Bachelors of Arts degree - take one from this category, two from this etc. versus Columbia's core curriculum where everyone literally takes the same classes for the core.

I was confident though that it was working. At one point, I saw a tweet by Morgan Missen. I don't know her but was told that she was the top tech team talent builder in the Bay Area. She tweeted "Stuyvesant, the Stanford of the East." I also started hearing from VCs, investors, and founders about how they didn't really like seeing a high school on a graduate's resume unless it was Stuyvesant. If they saw StuyCS, the knew it meant something.

By the time I left Stuy for Hunter College we were mostly stable. Maybe one or two more APCS sections on any given year and I did create a Software Engineering course called Software Development to replace research. That course, like my others, was designed based on what my graduates were saying they needed.

So it took twenty years of work but in this case slow and steady might not have won the race, but it built the program.

Now, I've gone over why I left Stuy in some other posts but two important reasons were my stagnating - I had done everything I felt I could do and also, I felt that some of my team was also being held back as they were too deferential to me.

Regardless, I left, JonAlf took over team leadership and here's where they are today.

They now have an opt in full year CS requirement. Kids can either take a full year of CS or one of CS and one of drafting.

That translates to 13 sections of the full year course and 6 of the single semester. They also rebranded intro and now call it "foundations of CS."

They also have 7 APCS classes and that's probably the limit, at least based on teaching the Stuy version that includes data structures. They tried to have additional sections based on requests but that led to a portion of the students really struggling.

To address that, they created NeXTCS which covered APCS-A but at a more relaxed pace and had a lot of fun projects. I think they were using Processing. They're running 4 sections of that.

On top of that add 2 Systems to go along with 2 of a new Cybersecurity class as well as the usual 2 of softdev and a new Ethics class. The only reason they aren't running 2 graphics classes this year as well is because they had a crunch in math and need to send a CS teacher over to teach a couple of math classes - how times have changed.

To run this, they have 8 full time CS teachers.

There's no doubt in my mind that my leaving was the right move. StuyCS has gone on to new heights under new leadership and I couldn't be prouder of them.

It did take a span of just about 30 years to go from no CS to a three year sequence, a full year required option, a coherent sequence of electives and eight full time CS teachers. Just another indication that education is a marathon not a sprint. Now, we're in the infancy of CS teacher certifications and much of the country isn't even where Stuy was thirty years ago in terms of courses. I expect it'll be another thirty years or probably more before it's all worked out. I guess for old timers like me (and some of my readers and friends) we get to be Moses - we get to see the promised land but we won't get to enter it. It won't be here for another few decades but that's okay, that's just how education progresses. It's gratifying to look back and see how far we've come so far.

What they remember

Fri, 06 Sep 2024 15:10:30 -0400

I haven't blogged for a few weeks. Part of the reason is that I haven't felt that I've had too much to say. That's in part because it's summer time and in part because I'm no longer teaching so that means no new experiences with students, no new stories, and, at least right now, no new CS education adventures. Part of it could also just be that I'm following the trend - fewer CS Ed bloggers blogging less frequently.

Another part of why I haven't blogged for a few weeks is that I've decided to try to write a book. Since retiring I've had a few people mention it to me so I thought I'd give it a go.

I've been writing a few sample chapters over the last there weeks (around 35,000 words , 75 pages so far). The plan is to alternate chapters - one with a story from my career followed by one on teaching something in CS or some CS pedagogical technique. Rinse and repeat. The "story" chapters will also have some educational message but it won't be topic or in class specific.

We'll see where it goes. Even if it comes together I have no idea how I would go about trying to get it published

Since my memory isn't what it was and even if it was, our memories are malleable I decided to reach out to my network of former students. I know that overall I've done a solid job as a CS teacher but what specifically was memorable. I might have loved a topic or something I did but maybe it was a flop. On the flip side, maybe there were things that had a major impact - good or bad and I thought they were just throwaways. I sent an email to my alumni list and posted to Facebook asking for the stories.

What I got as a result is what made me think about the title for this post.

Some of what I got wasn't a surprise. The competitive kids enjoyed things related to the USACO. Some talked about some of the bigger themes in the advanced classes. Everyone loved Halloween!!!

I was surprised though with other things that came up.

A surprisingly large number of responses singled out my "Homer the robot" lesson - my day one or two activity for new CS students. I used to do "Homer the Robot" in APCS when students took AP without any prior class and then moved it to Intro once that class was established and a requirement.

Basically, I would play the part of Homer the Robot, named after Homer Simpson, largely due to Homer's well known mental capacity and also because I've always been a huge Simpsons fan - remember I started teaching this lesson or a variant of it back in the mid to late 90s. I'd set up a simple scenario - instruct Homer so that he writes today's aim on the board.

They would shout out instructions and hilarity would ensue.

The idea of course was to show that computers follow simple instructions and follow them literally. As I always tell my classes "computers are horrible at doing what you want them to do but they're great at doing what you tell them to do."

I'd interpret or rather, misinterpret the class's instructions but eventually we'd get an aim on the board.

Now, I know this isn't a unique idea. I'm sure while I was coming up with it in the 90s in New York City others were doing similar things elsewhere. It's like pair programming. Industry "invented" the idea as part of agile programming but just about every CS teacher ever invented a variation of pair programming prior to industry ever dreaming of two people on one computer every time that teacher was in a computer lab with more students than working machines.

Nowadays you can find similar unplugged exercises a simple web search away - things like "the paper plane exercise" or the "peanut butter and jelly sandwich exercise" come to mind. Still, I prefer my version but if you want to know why, you'll have to read my book if it ever comes out :-).

When reading my student messages, I was surprised how often Homer came up. Sure, it was fun and the class always got laughs but there were plenty of other classes with more laughs. Maybe because it set the stage. Many of the students who were subject to that lesson ended up taking two and a half years of CS and many classes with me. Maybe because it started to demystify computing. I can't say for sure but it's something I've been thinking about. Why that lesson and not any of the hundreds to follow.

As I said, it's interesting what they remember.

Another theme that came up in more responses than I would have expected didn't refer to a lesson, any particular activity, or story. It had to do with the CS class, or in many cases, the CS office area - room 301, being a "safe space." Interesting because I don't think people even used that term back in the day. I found that interesting because first, it again came from more students than I would have expected. Second, it also came from some people who were never my student but are now in my network. They were never in my class but still found there way to the CS corner of Stuy. Third, I never thought I did anything special or even really thought about making a "safe space," particularly since that can mean very different things to different people.

Those were the two biggest surprises. Not that they were mentioned but that they were mentioned more than I would have expected.

There were plenty of other responses. Some I expected and some not and lots of good memory triggers as I stroll down memory lane.

With the new semester starting for teachers around the country it's worth spending a minute thinking about it - you'll all have an impact teaching your subject area but you'll also have an impact in ways you probably wouldn't think about.

Have a great semester everybody!!!!

The Danger of a Bad Teacher

Wed, 14 Aug 2024 10:51:51 -0400

For years I've been one of the few people banging the "CS teachers must know CS" drum. There have been a few others out there but in terms of being vocal about it, we're in the minority. I get it. There are many teachers who volunteered to teach CS and are working hard to acquire the knowledge needed. There are also those who were voluntold who are trying just as hard.

The problem is, at least from what I've seen, there are far too many CS teachers who after years of training still can't navigate CS1 or sometimes even CS0 level code and have very limited context when it comes to CS. They know the bullet points in the scripted curriculum they're using but that's about it.

I also get that it's not easy to be vocal about it. When I talk about this, it can seem like I'm calling out my CS teaching colleagues and while in a way I am, I'm talking about those who've stayed at barely CS0 for multiple years, not those working to improve their CS knowledge.

I was reminded today why this is a problem, why people should be paying more attention to it and why more should be calling it out.

I was chatting with a young man today. He has a Math and CS background - was a double major, and is considering a career change from a non teaching job in to either teaching or something education adjacent.

One thing that came up in our conversation was that he took a CS class in high school and that almost caused him to drop the field. My new friend mentioned that the teacher in question was a math teacher and a pretty good one at that but noted that in addition to not knowing the CS, the teacher didn't know how to teach a CS class. Of course, my friend acknowledged that he's not a teacher himself but the end result to him as a student was that he was turned off.

In college, in spite of that bad experience, he decided to try CS again and after a couple of nondescript courses he took one that he described as life changing. A great instructor that reignited his love for the subject.

This happened to me as well but on the math side. I was always pretty good at math and enjoyed it, until I wasn't - this was Calc II. I thought I just hit my wall and I started to dread math even though I knew I'd need some to be a CS major. I figured it must be me. I slogged through my required courses in college barely getting by until I took Discrete Math - the instructor was "just" a grad student but he was terrific. Turned me back on to math. I realized I could do this stuff. A couple of years later, I realized that my problem wasn't that I was stupid but I had a bad teacher who didn't teach me what I needed to know to succeed in Calc II.

We always here about that small handful of teachers that many of us had who inspired us or set us up for our futures but there are also many tales of that bad experience. That reason why someone is math phobic or why they abandoned some prior passion. Maybe it was a bad teacher, maybe a good teacher severely out of their element but throwing kids into a CS class with an unprepared teacher just to say they got to take CS might not be the best idea. We might be exposing them to the field but we also might not and worse, we might be turning them off. If the teacher is well prepared, all good. If the teacher isn't but recognizes their deficiencies they can still pretty much ensure that they won't be harming students and some good will likely come out of the class. An ill prepared teacher who doesn't know any better can be a dangerous thing.

This of course isn't to say that there's one way to prepare a teacher or for a teacher to be prepared and certainly there are many ways for said teacher to run their class. In my conversation today, my new young friend shared thoughts on a few of his CS teachers noting how many of them did things very differently from each other but they each played to their strengths and to my friend's view, the were effective.

I'd like to think that this situation will ultimately work itself out, at least in states that have strong credentialing but even there, we could have issues as I've seen CS credentialing programs that barely cover CS0. If we want to make sure that ALL students receive strong CS education, it's up to all of us in the community to call out the situations preventing that and that includes calling out bad prep programs, bad PD, and schools and districts that tolerate or even encourage CS teachers who don't know CS.

Teachers learning, using, and teaching AI

Wed, 31 Jul 2024 16:33:08 -0400

How do we prepare teachers to teach AI to their students, teach their students to use AI effectively, and how to effectively use AI themselves.

I was having this discussion with a friend this morning and I'm sure many educators have been having this and similar conversations over the past year or so.

While the discussion around AI is different from the discussion of CS, after all an AI writing assistant can be used with zero understanding of AI and in any subject area but AI is a subset of CS and there are a number of parallels when thinking about preparing teachers in both cases. Unfortunately, we're still having the same discussions about CS so while we can look at what's been done so far, everything's still too new to know what works or not.

Should students learn about AI in all of their classes or in specific AI or CS focused classes? What about in a specific "Using AI" class? Can "all teachers" learn sufficient AI to teach their students or do we need specialists? Are the answers different for the early grades than from high school? Will teachers learn on their own? Should we provide professional development? What would that look like? Or, do we need to integrate all of this into our teacher preparation programs.

These are all issues we've been dealing with in CS education and we don't have any definitive answer.

I don't have any final answers but I though I'd share some of my current thoughts both on delivery and also on training content.

In terms of delivery, the two options are Professional Development (PD) or preservice coursework.

While PD can be very valuable, quality PD is extremely rare, at least if you listen to teachers. On the other hand, if you're in a state with minimal teach requirements or the powers that be can't convince the colleges and universities that prepare teachers to bring in AI (or CS for that matter), then you're left with PD.

A local department of education, even one as large as NYC can also roll out PD much more quickly and at a wide scale like what NYC has done in their CSforAll program without requiring any changes at the state level, in legislation or by including colleges or universities.

A good example of some good work being done here is what Nora Burkhauser and Nick Yates are doing in Maryland. I wrote a bit about it here. They run a 2 part PD series. It doesn't prepare teachers to use AI effectively as it's meant to educate them on what AI is but it looks to do a great job at that. Part 1 is a great introduction for non CS teachers of all grades and part 2 goes even further. Ideally there would be even more for CS teachers, as I wrote about in my earlier post, but that's a limitation of the PD model and they're doing all they can do under their current system. They're also doing more than so many others.

On the other hand, I've had an opportunity to really see the pros and cons of the PD offered in NYC for their CSforAll program - here PD is generally offered by a PD provide - that is, someone like code.org or the College Board.

I got to do a deep dive when I evaluated applications for the CS Certification programs at Hunter College. I reviewed hundreds of applications over a three year period. If I remember correctly over 100 each year and close to 200 multiple times. There was a small amount of overlap each year in applicants but not much.

Applicants needed rudimentary programming knowledge in any text based language. Write a program that used a loop, had variables, a conditional, a function, and used an array or list. That's it.

The vast majority of the applicants had NYC CSforAll training and many, if not most had multiple years of this PD.

In years one and two I could barely accept 20 per year. In the third year I was able to accept close to 50 but that was by additionally offering multiple bootcamp sessions.

Clearly, the PD model wasn't working. We had a large number of teachers having gone through multiple years of training by the big name players in the field yet far too many couldn't tell the difference between a for loop and a forklift.

Sometimes PD is the best you can do and often it's all you can do right now but it's not the long term answer.

If this doesn't convince you I'll add that we all know that the attrition rate for teachers is among the highest of all professions and every time a teacher leaves the profession after two or three years, all that effort is lost. A PD model at best is a never ending treadmill.

On the other hand, if you work AI into your preservice courses, you have more time to cover it, the students (teachers) have some accountability (which rarely exists in PD sessions as one basically becomes certified by just showing up), and teachers enter the profession with AI as part of their skill set. Of course, ongoing PD after that then becomes a bonus.

Of course the challenge then becomes convincing states to require it and educational institutions to offer it.

In my Hunter programs, there's certainly room for both AI as a CS subject and also how to leverage using AI but my program is really made for high school CS teachers. I've had elementary school teachers but it goes so far beyond what they need to know in terms of content and there's less focus on their needs. It's also made for CS specialists so we'll never get the history teacher for example.

That said, the topics and ethics CS classes both give platforms to learn about AI. Even in 2020 we had a unit in the topics class on neural nets and many in the ethics class on bias. Meanwhile the methods classes and curriculum development class give a platform for exploring and using the tools, although, again the focus in on a CS class.

I can see a number of places we could work AI, from a generalists point of view into a typical preservice program and, between you and me, there's enough cruft in most programs that nothing would be lost. Even if you do have to cut something, AI isn't going anywhere so methods classes, for example, really will have to deal with AI tools if they really want to prepare teachers.

In any case, the next question is what to teach the teachers. I can think of a number of activities, both on the computer and unplugged which would give teachers a basic understanding of some key AI concepts. While not AI, a Markov Chain text generator show how you can use statistics and probability to generate text. Now and LLM is much more complex and uses a neural net on the inside but an unplugged activity around this simple generator can go a long way in demystifying an LLM and once demystified, they can be used more safely and more effectively.

I've done a number of other proto-AI lessons like "Who Played Spiderman?" or my Caesar Cipher unit which can be converted into unplugged activities to demystify other aspects of AI.

There's also a good amount of low hanging fruit for other uses of AI like generating discussion topics or having groups of students using AI to generate or research content and then have them discuss or debate.

Another advantage of doing this in preservice is that you can spread the AI over many courses and you'll have the time to also discuss issues like bias and dangers like using an AI tutor at the expense of developing teamwork and interpersonal skills with your classmates. In A PD time is much more limited and while there's good work like the Maryland project I spoke of above, far too much is given by vendors who's bottom line is selling a product. On top of that, things like bias is really beyond AI - yes, AI exhibits all sort of bias but the root is societal bias and where an AI PD is unlikely to be able to sufficiently handle that, an entire masters program can at least make a dent.

So, those are my thoughts for now. They'll probably change and evolve and in any case, given my current role as "retired guy" I'm not sure when I'll be able to act on any of this, though I do have some thoughts.

AI without guardrails

Fri, 26 Jul 2024 14:37:10 -0400

CSTA was only a week ago but I've already been trotting out things learned there in conversations. Specifically, during the panel keynote, one panelist, while discussing the affects of AI cited a study.

Two groups of programmers were tasked with writing some safe and secure code. One group used AI tools like copilot and chatGPT. The other didn't

When all was said and done, the group that didn't use the AI tools produced more secure code. The group that did use the AI tools not only produced worse code but they were actually more confident in their code.

To be honest, this really didn't come as a surprise.

Today I was thinking about this again. My friend Bethany Crystal wrote up a post on leveraging AI. Lots of good stuff in the post, one point was when Bethany talks of when she was with StackOverflow and how it became such an important part of developers workflows and how AI is now becoming the same but beyond coding.

The post is a great post to get people thinking about using AI in their workflows but we also have to be reminded that AI without verification can be dangerous.

Bethany gives an example of using chatGPT in a similar vein to StackOverflow - as a timesaver to generate content from which she would then build her actual work.

To do this, she asks chatGPT:

Hi, I'm a principal at a public high school in the South Bronx. Can you help me make a case about the importance of project-based learning to teachers who are skeptical about why we need to teach kids how to code?

If one is an educator and already knows their stuff on Project Based Learning (PBL), the results are a nice start if you're building a presentation, going to address your faculty, or a number of other tasks.

On the other hand, if you're someone who just got out of Teach for America and are now making the next greatest education startup but don't really know the real deal in education, it's not nearly as helpful. I ran this query twice, the second time, the answer didn't talk at all about some of the problems with PBL - class size and time issues, project storage over days for physical projects, prep time needed to set up each experience and many more. The first time I ran it, it mentioned some of the issues with PBL but just in the most generic sense "make sure teachers have the support they need."

Both answers also conflated coding and PBL so if the person using chatGPT doesn't have as strong education and in this case programming background the chatGPT solution can do more harm than good.

I then, for kicks, tried this prompt:

Hi, I'm a principal at a public high school in the South Bronx. Can you help me make a case about the importance of the use of lectures in teaching who are skeptical about why we need to teach kids how to code?

I had to change the wording a bit but it's basically the same ask but with the use of lectures instead of projects.

While parts of the response are very specific to lectures, the entire answer actually reads very similarly to the PBL answers and its justifications are just as strong (or weak depending on your point of view).

All of my queries resulted in some incorrect assertions.

Now, I'm not saying at all that one shouldn't use these AI tools - we should - if we already know about the domain we're using them in. They can be great at generating code for well understood algorithms - we just have to make sure to verify what's been generated. They can help phrase a difficult paragraph or reframe an argument and assist us in many other ways. The important thing, and the thing that that first group of programmers from the experiment up top forgot, is that we have to verify what these tools give us. We end up using them like a curated library and it isn't.

Tools like chatGPT look at a corpus of documents and probabilistically generate its content. This means it's basing its results on the documents out there.

In the case of code, we know that there's a lot more bad code out there than good. This means that while chatGPT can probably easily generate code for a really well known algorithm like a sort, we've got to question and test anything it puts out.

Likewise for education issues, there's a lot of bad material on education out there. A lot of garbage has been published and treated like "the true way" to teach and to run schools. Things that "reformers" love because they cut costs but teachers know are crap. Things that are likewise pet theories of education that have become popular but are no better and are frequently worse than what teachers know are tried and true methods.

What's more a chatGPT response can lack nuance. I can picture someone who's starting a school - a non educator let's say, heard from a person that PBL is good. Now, I know some people who swear by PBL and say to use it all the time - those are people in super selective private schools with tremendous resources and ultra small classes. The reality is that PBL is good and should be used as appropriate but there are other good techniques that will be better under some circumstances and worse under others. This person starting the school then dives into the AI with queries like I made on PBL and get a very one sided view. Now we have a problem.

I loved Bethany's post and it got me thinking more of how I can leverage AI, of course the problem for me is that I'm currently between gigs so don't have any work to apply it to. It also reminded me of the dangers of using it blindly and without guardrails.

CSTA 2024 final thoughts

Tue, 23 Jul 2024 15:10:14 -0400

Got home last night. Only delayed a couple of hours so we were much luckier than many friends who had to scramble and deal with canceled flights. It helped that we flew JetBlue, an airline not affected by the CrowdStrike issues and also because we were staying until Monday anyway so as to play tourist.

What's left to cover. Let's start with the exhibition hall with all the vendors. There were the big companies - Google, Amazon, and Microsoft, curriculum provers like CodeHS and CMU (CS Academy), and lots of robots and robotics companies. There were a few others mixed in as well.

I didn't speak to too many vendors though as I'm not particularly interested in using someone else curriculum and while some of the robotics stuff looked very cool, it's not something I'll be using since I'm no longer in a classroom and any toys I do want to play with have to fit in my Manhattan apartment.

I did spend some time with the people in the Kira booth. I thought I'd check them out since they seem to be getting a bit of buzz recently. Their home page touts Kira as "The AI platform for schools." There are courses which seem to basically be MOOCS, although they can be used (and appear to be meant to be used) as a tool by a teacher with a class. There are pre-made courses and units and teachers can also build their own content. This all looks fine but seems similar to other course platforms.

They seem to want to differentiate themselves with their AI piece. This comes as an auto-grader which I didn't get to see along with an AI tutor for the students. It looks cool but to be honest, I don't know if it's even a good idea as a classroom tool. I think I'd rather have a student talk with a neighbor first. I plan to do a deeper dive into Kira sometime soon and then will probably report back. Right now, if they keep the buzz, they'll sell well but I'm not yet sold. If I were still in the classroom, I think I'd stick to my old school ways.

To me, CSTA has always been a people conference rather than a session conference and this year was no exception. I got to see and spend time with many old friends and met a couple of new ones. For the sessions though, while I found a couple of great ones, many didn't do much for me. That's not to say they were bad sessions. I think I even noted that in my previous posts where I felt that a session would be great for a teacher at another point in their career but it didn't scratch my itch.

Of course, given that any attendee can only go to one session per slot and since descriptions don't always tell the whole story, perceptions vary. I just read Amy Ko's CSTA report and Amy lamented the lack of equity and justice focus. I felt there was good coverage there. I on the other hand felt that there should have been more coverage on actual content and pedagogy. Both of us could be right, both of us could be wrong. At the end of the day though, CSTA provided a platform with CS teachers from across the country could discuss whatever issues they desired with a diverse range of colleagues and that's a pretty special thing.

I'd be remiss if I didn't close with a few words on my tourist time. I wasn't excited to come to Las Vegas - I'm not into casinos and coming from New York I've got better and cheaper food options as well as pretty much any entertainment I want. In NY I also don't feel like then entire town is trying to scam me for my money. I don't want to come down though on residents of Las Vegas though since I do know a couple of great people who live there so I'm sure people there are just like everywhere else it's just that the tourist industry there isn't for me.

That said, we did do a number of things we enjoyed.

We did a couple of day trips with Pink Jeep tours - one to Hoover Dam and the other to Red Rock Canyon. Our Red Rock tour guide was the better of the two but both were solid and both trips 100% worthwhile.

We also visited the Atomic Testing Museum and the Mob Museum. Both were really interesting. I've always been interested in the mob thing, particularly the Jewish Mob in New York so the Mob Museum was a particular hit for me.

We walked Freemont Street, which was basically another touristy strip but got a great apple fritter at a donut shop across from one of the ends.

On our last afternoon, after our Red Rock tour we grabbed lunch at Sayulitas - a hole in the wall burrito place. I've got to say - probably our best meal in Vegas. All the strip stuff was overpriced although some was tasty but Sayulitas was dynamite. When we got in, we were concerned about the lack of seating but there was a huge crowd of locals and pretty much no English being spoken - all good signs. Hadn't had burritos that good in decades.

From there we went to the Pinball Hall of Fame. While I'm not into casinos, I love pinball. Basically a pinball parlor with machines ranging from the early days up until today. Also a bunch of novelty machines like a vertical circus game with three mini pinball games in it and a couple of two player games where each player is on opposite sides and the playfield tilts back and forth. Super cool. Price per game range from a quarter for some of the old machines to a dollar for the new ones. Devorah and I spent a couple of hours and $20 and had a super fun time.

So that's CSTA 2024. Before flying to Las Vegas, I was already not hot on the town and also not so hot on some aspects of CSTA, the national organization. I think the most positive thing I can say is that I left the conference not only energized but thinking about proposing a session or two for next year.

CSTA 2024 - the final day

Sun, 21 Jul 2024 19:51:07 -0400

The conference ended this past Friday. After a couple of morning sessions I started to play tourist which included spending more time in the 100+ degree out doors. Fun stuff that I'll talk about later but left me too drained Friday and Saturday evening to continue my write ups.

First up was another AI session "Unlocking AI: Teaching Artificial Intelligence in K12." Slides and other materials can be found here. The speakers were Nora Burkhauser who gave the talk on "Machine Learning for Kids" and Nick Yates. The talk focused on two professional development opportunities they've developed to bring K12 teachers of all grade levels and all subjects up to speed on Artificial Intelligence. Well really one PD opportunity split into two part.

Part one, offered via remote sessions with interspersed assignments followed by in person PD covered the basics. Basic definitions, classic ML and AI, a typical AI pipeline. From where I sat, it looked like this course provided a good overview for any teacher new to AI. It didn't, in my opinion cover enough for a CS teacher in a HS class to do a full AI unit but given time constraints of a typical PD, it's a great start and looks like it provides plenty for the early grade teachers and the non CS teachers while at least getting the CS teachers on the road.

Part 2 goes over available AI curricula, touches on modern AI - generative AI, neural nets etc., and also goes over applying AI. In fact, they presented the same lesson, more or less, as was presented in the "Machine Learning for Kids" talk I attended on Thursday. This part 2 course certainly adds to the foundation for the early grade teacher on non CS teacher, but still wouldn't be enough, in my opinion for the HS teacher. I understand that the speakers can only do so much - they can't offer one or more college or graduate level courses in AI - they have to work within their state's professional development system but we have to become vocal about the fact that so long as our CS teachers are not coming in with strong content knowledge, add on professional development isn't a solution, it's just window dressing. Sure, right now it's frequently the best we can do but more people have to start banging the drum that we need our CS teachers to actually have a good amount of knowledge in, you know, CS.

Before moving on, I do want to say again that this was another strong session and the speakers are doing great work, they're just trying to do their best within the limitations of the system under which they work.

After this AI talk I attended some lightening talks. The first was given by a new teacher on her just barely staying ahead of her class. It was a nice talk but got me thinking that it might be useful for CSTA to specifically set up an affinity meeting that combined brand new CS teachers (particularly those with limited content knowledge) and match them with groups of slightly more experienced teachers - say those with 3 or 4 years experience. Teachers with enough mileage so that they can provide encouragement and guidance to the neophytes but not with so much experience that they've forgotten the struggle.

Another lightening talk was about creating a social media presence. I was interested in this because the speaker was going to talk about making short form videos and I've spent the past decade+ in the world of blogs. Unfortunately, the talk was really more about how to tweak your already existing video content to maximize viewers. Valuable if you've already got a video presence but not the info I was hoping for.

I didn't attend the last keynote so that was it for me but that's not it for my SIGCSE post. Next up, the exhibitors, Las Vegas and final thoughts on the conference.

CSTA 2024 day 3 part 2 - AI and a keynote panel

Fri, 19 Jul 2024 09:56:09 -0400

So, I stopped my previous post after my thoughts on the DEI talk. The actual post wasn't all that long, the reason I said for stopping the post but rather because writing that post was emotionally draining. Can I talk about that stuff? Will people interpret it the way I intended. Will people who I don't want to offend and don't think I'm offending, will they be offended? Will the people I want to offend not be offended (okay, just joking here). It's really hard for people to have these conversations and too many people these days don't allow for nuance but I felt it important to get my thoughts out there. Not so much because my thoughts are anything special or that they're right or wrong but one, because the session was terrific and I am documenting my conference experience and two because having these conversations won't get easier unless people are willing to put themselves out there.

Anyway, back to our regular programming.

Thursday afternoon I sat in on "AI Lessons using Machine Learning for Kids." Machine Learning for Kids is a site/platform that allows people to set up and train machine learning models. The speaker's slide deck with links to resources can be found here.

The speaker walked us through how she used ML for kids with her class. She basically teaches a typical Machine Learning pipeline, that is:

collect data –> prepare/clean –> select model –> train –> validate –> predict

Her specific project was sentiment analysis of text. The platform can also work with images, sounds, and numbers. After logging in (or clicking to try without registering) you start a project. You select the model you're using, "recognizing text" in this case. You then use their interface to add data. The instructor, of course, would have to provide guidance so that a student would know what and how to enter.

When finished, you click the train button and the platform builds the model. I'm pretty sure you can then play with the model but I didn't actually test that. The cool thing, to me was that you can then have the platform generate code - scratch, python, and a couple of other platforms. I clicked the generate button and all of a sudden I got a chunk of Python code that I could embed in a program. Specifically, the code let you use your now trained model as a library (or more specifically, web based API) and make calls with data and get your results.

While this is very cool, it raised some questions. The big one for me is that you can go through all of this and have almost zero knowledge of the underlying machine learning. Contrast that with the session I wrote up a couple of days ago where the students learned all the little details of gradient descent for linear regression (more on this later).

Is this really the right way to go?

The instructor noted that when we use a tool like ChatGPT we're using an already trained model so this is no different. That's true but in a class learning about, say sentiment analysis as an ML technique, I think students should learn how the sausage is made.

Now, this approach of leaving the actual ML as a black box does still allow for some good learning to go on. Students will learn the pipeline. They can also discuss issues like false positives and true negatives - the effects of different training data, over training and much more. They could also see, from a user point of view if different models produce different results. A teacher might also have constraints to keep them from diving deeper - insufficient time, toolset limits, student backgrounds and I'm sure others.

That said, I'm a firm believer in seeing what's under the hood.

Now, going back to the Linear Regression talk from Wednesday, that might be too much for an average high school class and almost certainly too much for a pre APCS-A class so where do you draw the line?

It turns out that in a CS0 class using a text based language like Python, students can pretty easily write a simple sentiment analysis system. You can also scaffold pretty thoroughly.

The basic gist is:

Make two lists, one of positive words, one of negative
Read in the text you want to test
Split the text into words
Count how many words are in the positive list and how many in the negative.
If there are more positive than negative, it's a positive piece, otherwise negative.

This is extremely bare bones and has issues. For instance if a document has "not bad" in it it'll just see the word bad and classify as negative but that's something the class can talk about. The project can also be easily extended in any number of ways. You can also include all of the good things from the speakers lesson design - the pipeline, different training sets, etc.. The important thing is that the students will have a stronger understanding of how sentiment analysis actually works.

While this was a strong session, I'm a big believer in students understanding the inner workings of things and I think this is going to be even more important with the current AI craze - this is something one of the end of day keynote panelists actually pointed out (more on that down below). Of course, you have to know your population, where they're at, and what your restrictions are.

In middle school with scratch, the black box approach is most likely the way to go, particularly since ML for kids hooks in to scratch. Teaching a High School class using Python? You should probably do a unit where your class codes the whole thing.

Just my opinion and approach.

One final note before moving on to the rest of the day. I've noticed that while there are plenty of sessions that, by description, talk about ChatGPT tools - what's considered the cutting edge of the AI fad, there have been a few sessions on Machine Learning and more traditional AI. This has been great to see as so much of what runs and what will run the world operates using these techniques and technologies and it's important not to drop them in education for the flavor of the month.

That was it for the Thursday sessions for me, I spent the rest of the session time looking at the posters and the hallway track.

The Thursday keynote was a panel moderated by education journalist Natasha Singer. The panelists were (as seated left to right):

Christy Crawford of the New York City Board of Ed's CSforAll initiative
Maggie Johnson, Global head and VP, Google.org
Mehram Sahami, Chair of the CS department, Stanford University
Kip Glazer, Principal, Mountain View High School

Before diving in to the panel, I've got to express more pride in my Hunter CS Teacher Cert program graduates. While delivering some opening comments, Natasha Singer gave a shout out to another Hunter CS Certified teacher, Marisa Shuman citing some of the work Marisa's done bringing AI to her students in New York. Here's Natasha's article about it: link

Now to the panel. Overall it was a great session but I did notice one thing missing. There were no actual teachers on the panel. Maggie Johnson's from Google and he background's in higher ed as is Mehram Sahami's. Christy Crawford doesn't show any classroom teaching experience on her linkedin. Only Kip Glazer listed any K12 teaching experience on linkedin and that ended back in 2016 - about 11 years total which is a nice chunk but in my mind generally not anywhere near the time needed for mastery.

It would have been nice to have at least one actual K12 teacher up there. Even better if it were an experienced CS teacher who also had strong content knowledge.

The first question was on the overall impact of AI in education. Mehram was first to respond and hit the nail on the head - AI can't transform education, only people (teachers can transform education. Bravo. He added that AI can enhance teachers ability to transform education but it's just a tool. As a side note, I really felt that Mehram was pretty much spot on throughout. I was very impressed with his responses.

Christy chimed in with a common teacher's fear - will AI replace them.

I think this was an interesting juxtaposition. You have the academic speaking the truth of how things should be and the person immersed in K12 noting the unfortunate reality that the politicians and private interests are always looking to save money and turn a buck by cutting teachers and directing money to private companies to "compensate."

The next question asked were coding should live in K12 with tools like copilot and other code generating LLMs.

Maggie noted that even if AI can generate code, it still needs to be understood, checked, tested, extended, and embedded. She said she didn't know how much coding would be necessary but certainly it's something we much teach in K12.

Mehran noted that teaching coding syntax is still important and that we can use tools like chatGPT like a code library or API to enhance our student's work. This echoed something I wrote about when I wrote up this year's SIGCSE experience.

In response to Natasha Singer asking if there should be limits on AI on AI Maggie spoke on how it's something that Google talks about top to bottom every day. When she followed up with "what about in the classroom" Christy chimed in yes but added an oh so important "but not on the backs of the teachers." I do lament that there wasn't an actual teacher on the panel but applaud Christy for consistently having teacher's backs throughout the panel. It's not a surprise, she's always had teacher's backs in her role at CSforAll. I wish others in the NY DOE would follow her lead more.

Natasha also asked about corporate influences on curriculum as this is something somewhat unique to CS. As a new field and a new field in K12, few teachers and schools knew what to teach so for better or worse, outsiders filled the void. That's the college board, code.org, organizations like PLTW, companies liked CODEHS. This doesn't happen in other older fields. A friend of mine later pointed out that it sort of does but the outside influencers are the book publishers. Anyway, Maggie posited that technology changes so fast that industry is needed to keep curriculum up to date. I think she misinterpreted the question though. I believe Natasha was specifically talking about content providers like those I listed above while Maggie was thinking about pure tech companies informing curriculum which, if they did and if they "stepped back" to go back to yesterday's post would be great.

I think Mehram also misinterpreted the question as he talked of the symbiotic relationship between industry and education. I think he was thinking more about the university industry relationship.

Once again, though, Christy had the teacher's backs when she on the one hand noted how important industry can be - they can inform, provide opportunities like mentoring, and support but there has to be a hard line. They should not be driving curriculum or education. Again, bravo.

At this point, we ran out of time. Great panel. As I said, I particularly liked Mehram's comments which included a number I didn't mention here as well as Christy's constant defense of teachers but everyone was terrific.

That's it for day 3.

In the next day or so I'll get to day 4, the exhibitors, and some closing thoughts.

CSTA 2024 day 3

Thu, 18 Jul 2024 17:37:43 -0400

Started out today in a session titled "Keep Calm and AI on." The panel was made up of CSTA Equity Fellows. It turned out that the session wasn't what I was looking for. According to the description the panel was supposed to cover "the promise and the challenges of various artificial intelligence tools, especially generative AI tools such as ChatGPT, Copilot, and Bard." Instead it was a general panel and discussion of the use of AI. To be frank, the session wasn't really focused on CS educators at all and could just as easily been plopped down int a PD session for any subject or grade level. Now, that's not necessarily bad but it just wasn't what I was hoping for.

The panel consisted of a number of questions where each panelist had their say and then the moderator, also a panelist solicited thoughts from the audience. The questions they asked were:

What ethical concerns do you have?
In your setting, how is AI used?
What's one piece of advice you'd share?
What work have you seen being done for AI in education?
What is your call to action?

The moderator ran the panel well and we got to share many points of view. Personally, I was hoping for a discussion of some specifics with using CS specific tools like copilot or issues or possibilities with applying ore general tools like ChatGPT in a CS classroom or lesson but I suspect that the panel as it went own was probably pretty valuable to a lot of people in the audience

My second session was "The Power of Words in CS Education" - another panel. As per the description, "this panel discussion aims to explore the intricacies of intercommunication and confront uncomfortable conversations" with the key word being DEI.

The questions were:

What comes to mind when you hear the term DEI?
Who can talk about DEI?
What issues have you had speaking your truth in this space?
Do we need DEI offices and spaces?
Who owns this space?

Due to time constraints, the panel never got to the last two quesitons (other than a quick mention of question 4 where the consensus answer in the room was "yes").

The panel was run much like the AI session I described with the panelists sharing their thoughts and then the moderator (again, also a panelist) solicited audience input and again the session was run smoothly and effectively.

I want to highlight some thoughts that came out on question 2 - who can talk about DEI as well some thoughts I had based on something one of the panelists said.

One panelist commented that anyone can talk about DEI but talk is cheap, we have to look at actions. My take is that you've got to walk the walk and thought about all the companies who crow about their DEI programs but really just pay lip service. Companies who claim they're pushing say black SWE hires but really only fight with with the other top companies to monopolize the candidates from say Stanford. They talk the talk but aren't really doing anything to improve the situation but rather they're just trying to gobble up the low hanging fruit.

Another panelist mentioned that we need white people involved but they need to know when to step back. This is a tricky one. I get it - I can't understand the life experiences, situation, or needs of say a black person in the USA and I certainly shouldn't be telling one what they should feel or what they need but on the other hand I might actually have some background where I should be stepping in in a complementary way or might be raising a relevant issue.

I'll share two personal situations where I was frozen out of a community for not "stepping back." One was a conversation with some people about the LGBTQ+ pride parade in NYC. My position was that while I'm all for the parade, I had a problem with what I saw at the parade a number of years ago - near naked people on floats essentially performing simulated sex. Apparently I wasn't stepping back. Now to me, I wasn't telling anyone in this young subset of the LGBTQ+ community how to celebrate or behave and I wasn't telling the members of this set of LBGTQ+ community members how to live their lives. I was conveying my opinion that as a parent I really didn't want to have to start explaining people having simulated sex mid day on a float to my young kids. Please note, my objection wasn't specific to this parade, I've had the same complaint to other non LGBTQ+ parades that cut through family areas during the day. Was I out of line?

The other time was in a discussion of the SHSAT - the exam you take for admission to Stuy/Science/Tech etc.. There's long been a problem with the low number of black students being admitted to these schools and Stuy in particular. A number of ideas have been floated including one particularly bad one - admit the valedictorian of each middle school. I asserted why this was a horribly bad idea. I as accused of being a racist and frozen out. To them, I wasn't standing back. To me, I was pointing out the reality that this solution didn't get to the root cause of the problem and also wouldn't end well. I obviously can't walk in the footsteps of a black middle school student or their parents but I had much greater knowledge of the test and test process, what Stuy was like and how it operated and think I had and have a much better understanding of what the outcome would be from this plan.

Was I out of line and should have stepped back or was I appropriately contributing to the discussion. It can get complicated, particularly if recognize that many of these situations are pretty nuanced.

A final comment I'll make is that one of the panelists mentioned that they don't want allies, they want co-conspirators. The difference being that a co-conspirator will step in front of a car for them. I get this and agree but it also irritated me. Not about the speaker or the panel but something I've noticed going on with some subsets of DEI supporting communities. I'm hoping that these subsets are small but vocal minorities but it's hard to believe that given what I've seen in the media.

I'm a white male Jew. Back in the day Jews were recognized as co-conspirators with the black community and the LGBTQ+ community. We would not only figuratively step in front of a car for those communities but at times literally do so. Recently I've noticed the rise of antisemitic voices coming from some of those communities. This forces me to examine my relationship with these communities and makes me sad.

I really enjoyed this session but think it's important to note that this was another session that really had no focus on CS.

I still want to write about one more session I attended as well as the Thursday keynote and the exhibit halls (including a bit on Kira Learning) not to mention tomorrow but I think this is long enough for today's post. I'll cover the rest in the next few days.

CSTA 2024 day 2

Wed, 17 Jul 2024 23:01:02 -0400

What, no day 1? Well, day one was all workshops followed by a keynote and a reception. I generally don't do workshops and got in too late for the keynote but I did hit the reception but not too much to write about there.

This year the CSTA conference is in Las Vegas. I've never been to Vegas and there's not really much of a draw here for me. I'm not into gambling, the food, for a New Yorker is nothing special and while there are a number of shows that are unique to Vegas, again, as a New Yorker, it's not like I'm lacking for local entertainment options. That said, there are some things I'll enjoy taking in while I'm here - the spectacle of everything, the Mob museum, Hoover Dam, and Red Rock Canyon in particular.

To be honest, though, I really don't get putting a teacher conference here. Everything is so expensive. The rooms were reasonable (even with the resort fee) but everything is nickle and dime. The room only comes with 2 WIFI connections so if you're 2 teachers sharing a room and each need a phone and computer on WIFI, that's $30 extra a day. Then you've got food. A medium coffee in the morning is going to set you back $7 or $8 and today lunch cost me around $25 for a mediocre sandwich and a cannoli - around $10 more than I'd pay in NYC which is not a cheap town.

So, I don't get the location but I still really enjoyed the opening reception, had a blast today as well and am a big fan of the CSTA conference.

To start out the day I sat in on John Dutton's session "Creating a toolkit for all CS learners." I was hoping to avoid an AI talk since I knew I'd get plenty of those later on but I really wasn't sure where this session would lead. It turns out I didn't avoid an AI talk but that was okay.

The first half of John's talk was on how he set up his class culture - something that's both very important and also frequently neglected. He spoke about empowering the class by asking them how they would like this class to be different than others they've had - it reminded me of my friend and former student Davin, who, whenever he starts a new company spends time with the founding team discussing what they did and didn't like about the culture of their past companies and then uses their combined experiences to make a strong positive culture for their new venture. John' talks about student "jobs" in the class and more. I ran into John at lunch later and we chatted a bit. I felt that our thoughts are pretty aligned on class culture being really important but also confirmed my suspicion that he only had around 14 kids per class which affords him more opportunities to set a positive culture than a teacher of say 34 per class in a crowded room. This is not to take away anything from his work as he's aware of this advantage and appreciates it.

John also talked about some CS teaching techniques. Well, they're not really CS specific techniques but CS teachers frequently use them - rubber duck debugging, asking classmates before asking the teacher and a few other strategies. Then, the main event, a template to set up an assignment that then used a script that used AI to take the data in the form and create multiple, differentiated versions of the lesson.

To be honest, there wasn't a lot in this session for me given that I'm not teaching right now but for a new CS teacher, I could see a lot of value. I think I thought there would be more actual CS in the presentation but that's neither here nor there. John shared this link with materials which includes along with his lesson template, lists of frameworks and other resources, and a whole lot more. If you're relatively new to teaching CS I'd strongly recommend you click on the link and check out everything provided.

Next up was a teacher certification talk given by three educators fro Gwinnett County Georgia. The talk covered the state of things in Georgia and what they've been doing. I learned that in Georgia, one has to have certification (or rather, an endorsement) to teach CS. This can be earned either by taking an exam or by earning an endorsement through an approved program. This group from Gwinnett designed a one year and also acquired funding to cover costs. I like a lot of what they're trying to do. The Gwinnett team developed a one year boot camp for current teachers and area trying to emphasize a community model along with mentorship. Philosophically, at least from the pedagogy and implementation side, their program sounds similar to what I designed at Hunter in New York. My concern has to do with the depth of CS knowledge they're covering. This isn't necessarily the fault of the Gwinnett team though, it appears this is what Georgia's requiring and if they require content far deeper than Georgia requires, the state might not approve of their program and teachers might not attend. Their program maxes out at APCS-P which to me, is way to elementary for a certified High School CS teacher. Those teachers need, in my opinion, at least a bit beyond APCS-A along with a good deal of breadth.

It was interesting to hear about what was going on in another state and while it sounds like Gwinnett County Schools are doing some pretty good work within the framework and constraints their state provides, I fear that by setting the CS bar so low, Georgia is shortchanging their students.

After lunch, I attended John Chapin's session on "Machine Learning Made (more) simple." I loved this session. John's material is all here. John described and walked us through how he teachers linear regression. His goal was to teach us (his students) how it works rather than just use a library or black box toolkit. I'm 100% behind this. Now, it is true that John's at a public magnet school with high achieving students just like I was at Stuyvesant and what he presented, out of the box, probably can't be implemented in a typical school but the approach and attitude certainly can and should.

John motivated the problem by asking what data would be helpful in predicting home prices, he then had his class (us) work for a time in groups to try to figure out relationships and even models to go from the data to something to predict prices. He then stepped us through unplugged hands on exercises that led us to and through gradient descent. This in turn led to code and in the end, the class had both written and understood a key machine learning technique.

Great session and I hope the attendees take this approach in some of their CS classes. As I said, I don't think this specific assignment will go over well as is with all populations but I think the approach can. I've successfully done similar things both my AP classes and in pre AP programming classes like when I cover writing a Caesar Cipher decryption program. The difference is that in the Linear Regression lesson the class is learning a specific machine learning technique that requires a certain level of mathematical sophistication to really "get" while in my case, we do a simpler problem that uses many similar techniques but with simpler mathematics and simpler application. My problem sets the stage for that ML lesson a couple of years later. I also do something similar in my "who played Spiderman" unit.

Great session. I was concerned that all of the AI sessions - and there are many - will all be, taking out policy sessions about using existing tools and libraries and not result in students really understand what's going on under the hood. This talk allayed that concern.

I just hit the posters and exhibits for the last sessions and then attended the closing keynote delivered by Gholdy Muhammid - "Cultivating Genius and Joy in Education Through Culturally and Historically Responsive Pedagogies." I thoroughly enjoyed the talk. Muhammid is an engaging, entertaining and funny speaker. My takeaway was that our teaching should be evaluated by the Joy in the classroom and the pursuit of knowledge as opposed to just the standards based evaluation currently used which just measures facts and only facts. Note that I say "just the standards…" and not instead of them. To her credit Ms. Muhammid noted that standards - the specifics, the skills, the facts are important as well. This is critical. One of my annoyances at a lot of the K12 CS movement is that there's frequently only emphasis on the joy and they ignore the facts that at the end of the day the teacher must know some, I don't know, actual CS and that students must learn it as well.

That said, Ms. Muhammid's talk was really all about the type of education reform we really need in our schools. I had a couple of quibbles but it was a great talk. If CSTA releases a video, take the time to watch it.

The day ended with Devorah, who played tourist by day, and I went for drinks and some light dinner with the New York contingent at the conference. A highlight was getting together with four students from my CS certification program at Hunter including Marisa - a CSTA Equity Fellow and incoming CSTA-NYC board member, Jessica and Josh - both CSTA-NYC board members, and Victoria - the first person in New York State to earn a Masters in CS Education from an approved certification program and earn the associated license. So proud to have been privileged to work with all four of them.

The Dino Podcast

Fri, 12 Jul 2024 16:57:44 -0400

I love talking shop but I don't have the opportunity to do so as much now that I'm retired. So, when my former student Dino Beslic emailed me and asked if I wanted to be on his podcast, The Dino Podcast, I was both honored and excited.

Dino was a student of mine in the early 2000s but we haven't been in touch for years. I reached out to his brother Denis, also a student of mine, in order to connect yet another student to him for a job referral. Denis talked to Dino and that ended up in me being invited on the show. Of course, I'm also excited about being reconnected with both Denis and Dino.

Dino's a Tech CFO and investor and his podcast is about "business, technology, and geopolitics."

The tech job market has been in a downturn for the past couple of years and at the same time, we've seen a rise in AI an AI based programming assistants and related technologies. Dino and I spent a little over an hour talking about reasons for the downturn, if and how AI and the downturn were affecting the market and education. We also projected a bit into the future and made some recommendations for youngsters starting out.

I particularly enjoyed the fact that we were talking about education at the college level up into the workforce - most of my recent shop talks have focused on K12 and while I'm passionate about my work there, I've also done a lot with preparing students for the tech workforce and navigating early careers and am wondering if that's an area I'll go back to focusing on for my next adventure.

So, if you've got the time, check out the Dino Podcast. Here's a link to the episode I'm on: link

Hidden Communities of New York

Mon, 08 Jul 2024 15:09:27 -0400

Now that I'm not teaching CS every day, I find the desire to maybe write about some other things. I've still got a bunch of CS and education posts lined up, but for today, something different.

I was out for my morning run a couple of days ago. I usually leave my apartment between 5:00 and 5:45, go to the Hudson River and run an out and back. It's still pretty quiet at that time. When I leave later, the park is much more filled with runners, running groups, and bike commuters.

I'm heading south and walking north is an older couple. I say older couple but let's be fair, They're probably around my age and I'm just older than I care to admit although based on my current pace and recovery time, I should know better.

We wave good morning to each other and keep on our way. A little later, I see a younger man walking and running south (it was really hot and humid). We cross paths regularly but since we were going in the same direction and frequently wave, I slowed to join him. I said that after waving high for a couple of years, we might as well learn each other's names. We fist bumped, or rather he went for the bump, I went for the handshake, we switched but then finally synced. So, that's how I met Chris. We chatted for a few minutes - we both live in Chelsea, how we both get an energy boost whenever we see our "running buddies" in the morning, and also about some of the other early morning runners we've met.

He named Vicky, who I've been waving to for three or four years. Then I gave him Suki - the first person I started saying hello to. Suki and I would cross paths near Stuy, her walking north, me running south. She would greet everybody with a warm smile, a wave and a strong Good Morning! After a while, I slowed my run one day to get to know here. We also talked about Priscilla, the Mayor of Hudson River Park - at least to us early risers. Priscilla would always be walking or running on the path along with her magnificent dog Buddy. Unfortunately, Buddy passed about a year ago. We both lamented that we haven't seen Priscilla lately and we both hope it's because our schedules have changed.

We talked of a few others and I noted that I really have to introduce myself to that couple I passed earlier and then went we on to complete our individual morning workouts.

I got to thinking about this community. It's not a normal community. We don't all know each others names and really very little about each other but every morning, or at least a few times a week we've come to expect to see each other and greet each other with a smile, a happy hello, or a thumbs up or some other encouragement. We get a boost when we do but we grow concerned when we don't see our morning amigos.

In New York, we don't have that small town everyone knows their neighbors thing. Heck, I live in a 20 story apartment building and barely know my neighbors. We also don't have neighborhood schools and the mom and pop store everywhere is a dying breed. Yet, we still have these weird little communities that crop up. I look forward to seeing my running buddies again early tomorrow morning.

Good Schools for New Teachers

Mon, 24 Jun 2024 14:13:06 -0400

Usually, when I write about a teacher get together it's either my StuyCS people or CS educators. Last week, it was another group. Each year, Devorah, along with a bunch of her former colleagues from the Manhattan Center for Science and Math (MCSM) get together for a year end celebration. They've been doing this for something like 20 years even though most of the crew is retired now and all but I think one or two have left MCSM. For the most part, they all started close to the same time and most were in the Science department. I've joined in from time to time and they usually get a good crowd of between one to two dozen teachers. Last year, the gathering focused on Devorah's retirement and this year on another from the crew.

To be honest, I'm a little jealous. sure, we used to have periodic end of term lunches at Stuy but to keep a group together even after so many have left the fold is pretty cool.

David, a terrific Physics teacher in the group (and presumably the retiring honoree next year) made a comment, noting that so many of the group started at around the same time. He said that they were really lucky because MCSM was a really good school for new teachers. He commented on how much support he got across the board and how his colleagues, supervisors, and administrators really helped him and others grow.

This got me thinking. My first few years of teaching were at Seward Park High School - actually, a school with some similarities to MCSM. Both were large schools. Not giant monsters like Stuy or Brooklyn Tech but still pretty large. Both also had a range of students in terms of ethnicity as well as academic performance. In those early days I had classes with kids on parole, drug dealers, and thugs and I also had classes with the nicest kids you'd ever met. I had kids who in 10th grade or beyond could barely add and I had super high achieving honors students. All that and everything in between.

The biggest difference between MCSM and Seward Park is that at Seward most of the faculty were at war with the administration.

I've been told often that "bad" schools - rough students and/or horrid administrations frequently result in very close knit faculties and that seemed to be true at Seward. They all banded together and as a newbie, they took me under their wings. Now, I personally had no problems with the administration - I got along very well with my direct supervisor and he loved me but I still was adopted by so many others. New teachers frequently get the worst program but my chairman had a different approach - he rotated everybody - he worked to give me - the new kid - some tough classes and some easier ones and worked to get me through the entire math sequence quickly. As to the faculty - they helped me with lessons, sat in on my class, and let me sit in with them. They also saved me with discipline issues. I had one particularly rough class and a colleague with nonchalantly wander in to use a computer in the back of the room. He came in because he knew his presence would help with the discipline. The kids settled down and were non the wiser. Mike (his name as well), would then on a day to day basis work with me between classes until I could handle things on my own.

Without the support of teachers like Mike, as well as a handful of others, I would never have made it to year two let alone a three plus decade career. Very little of this support I got was official but it was critical in my staying in and succeeding as a teacher.

Now, MCSM didn't have a dysfunctional administration. In fact, from what I saw, they were pretty good. It wasn't until three principals later that they got their disaster, which ultimately prompted a handful of retirements and many transfers so might be fairly unique that Dave, Devorah, and all of them had a school with both a good administration and a united faculty, or at least science department that really fostered the development of new teachers.

I think something that also helped both at Seward for me and MCSM was the size of the school. Being large schools, there were always multiple teachers for each subject. That meant a new teacher was never on an island. It also meant that there were many other teachers to observe, learn from, and get support from. It also meant that there was never just one or two of the more desirable classes so while new teachers might have gotten less desirable programs, they didn't get the worst of the worst.

Contrast those two schools with Stuyvesant when I started. Stuy was also a large school - in fact much larger, with few discipline problems. I'd say that Stuy, back then, was much worse in terms of new teacher development. Sure, teachers were nice enough - I became friends with a number of them, but truth be told the most you'd get would be some notes from a previous semester or you could quietly sit in on class. At Seward, I got that, plus debriefings but check ins and more. At Stuy, I got the least desirable classes until I took over CS (although even the worst classes at Stuy were pretty desirable) as well as the worst rooming situation - I was running from the third floor to the 7th and back between three periods early on. At Seward, my program was pretty much the same as everyone else. I'd like to hope I helped foster a better environment as the CS program grew and we brought in new teachers but I can't say.

Another problem with Stuy, at least for new teachers, and this is probably true at many high performing schools, both public and private, is that new teachers don't really have to learn how to teach. They can be bad at their craft and the students will still pass the standardized exams. I've seen new teachers at stuy who never taught elsewhere who thought they were great but really weren't - at best they got by on the fact that they were young and had some charisma. Their kids did alright but deserved better. Some were self aware enough to figure it out but by and large it was an issue. Of course at a tough school the new teacher would just leave the profession while at the high performing school, the teacher in question might go on to a career of just not teaching that well.

In a profession with an extremely high attrition rate and also one where much preservice preparation is, let's say, limited, we have to do better. Obviously the biggest issues are things like class size, pay, and benefits but starting out at a school that's friendly to new teachers can be the difference maker.

What I have (and haven't) been up to

Wed, 19 Jun 2024 11:42:10 -0400

It's been about a year since I taught my last class. I miss working with the kids and truth be told, I do miss working with other educators and tech professionals on programs, policy items and related but I don't miss any of the bureaucratic nonsense, policies made by people who don't know what they're doing, nor the record keeping, grading or paperwork.

I spent the first few months in what I've been calling my retirement hibernation but more recently it's been time to emerge. Time to see where, if anywhere, I now fit in the CS/Ed/Tech space moving forward. That's both in terms of where I'd like to be involved and of course the flip side, if those places or people want me involved.

So, what have I been doing in that regards?

Connecting.

Not me connecting with new people (although I love doing that, just haven't had new opportunities) but connecting other people. A big part of this is connecting younger former students with older, more experienced ones - frequently for referrals. I like doing this but to be honest, I can no longer keep track of who's where – too many former students at too many companies. Still, this is an aspect of the StuyCS and how HunterCS families that I'm really proud of building and I really get nachas when more experienced family members are there for the younger ones.

I also had a former student who decided to spend this summer mentoring college kids who couldn't get an internship - he's between jobs and wanted to give something back. Trouble is, he wanted to work with too many kids than would fit in his apartment. Fortunately, I was introduced a couple of months ago to someone who just opened a new co-working/learning space and was looking to support worthwhile projects. I put them together and it's been a win win.

I've been enjoying doing things like that.

Advising.

This is interesting, because I generally don't actually give advice. I try to repaint the picture of the advisee so as to be sure I understand it and then I generally talk about the questions I would ask if I were in that situation.

In one case, I was put in touch with a team building a workforce exposure/entry experience for NYC high school students. They had a lot of workforce program experience but little tech experience. I think I was able to add value to their efforts given my tech background and educational experience.

In another instance, a friend has been working on a new school proposal. Now, I never actually started a new school but have been involved in school design a few times and have certainly built out a number of programs. It was a lot of fun talking to him about his vision and school's aspirations and to share with him my perspectives.

One other recent conversation was between me and a friend who's been building a company to help in both hiring and in leveling up employees. This is someone who I really enjoy working with. She cold emailed me around nine years ago when she was starting a direct CS Education program and we've been friends ever since. I really enjoyed this conversation because it wasn't the usual "I want to start a coding program" but was still something that, with my education background and range from small kids through workforce, I think left me with something of value to add.

I enjoy this type of "advising" or rather, as I look at it, talking shop and I do think that my background - having been in multiple levels of education from early grades through college as well as in industry and having also build numerous programs lets me add value. Maybe I should try to set it up as a consulting thing rather than just being there for friends.

On the other hand, then there's what I'm not doing.

I think I'll be less and less involved with the national aspect of the CSTA. Note I'm talking here about the national CSTA. I'm looking forward to continue to work with my local NYC and regional CSTA people whom I both like and respect.

A couple of interactions this past year really rubbed me the wrong way. Last year, they were looking for people to join their new Imagining CS Pathways effort. I applied for one of the roles. I wasn't selected. Now, that's fine. I mean, I personally felt, after I heard who was selected, that I had more to offer than many but that could just be my ego. What bothered me was that I was then sent an email asking if I was interesting in participating in a capacity that I had explicitly said I wasn't. I also wasn't particularly impressed with the report once it came out. I felt it was lacking in actual CS. Although I didn't think about this at the time, I suspect the people who built the committee already had an opinion as to what the end report should look like and built the committee to match. If true, not cool.

Later, I decided to throw my hat into the ring to run as an at large member for the CSTA board. Now, again, I didn't figure I'd be elected - I think I've pissed off a number of people over the years but I was put off when I got an email saying that they elected to place other people on the ballot. Why did this bother me? Because it says that there are people within the organization who can basically shape the board by only allowing certain people on the ballot. Again, not cool.

Now, CSTA can of course do what they want. They're the big players who get the Google money. I'm just a retired educator but since the organization appears to be going in directions that I think are wrong, there's no reason for me to continue my involvement or show support.

So, that's what I won't be doing. I guess it also appears that I'm blogging less frequently but that seems to be the norm these days. I'll probably continue apace until something clicks and I'll do a deep dive somewhere.

Philadelphia Freedom

Thu, 06 Jun 2024 19:02:04 -0400

Time for another travel post. Figured that by posting here there's more permanence and search-ability than a Facebook or other post.

So, Devorah and I are spending a few days in Philly. We started coming down here a year or so ago and this is our third trip. It's a nearby getaway from NYC and we've got friends here that we rarely get to see.

Sports fans notwithstanding I'm surprised that Philly doesn't get more love from NYC people looking for relatively nearby travel. It's under 2 hours away, is a very walkable city, and has the things we love, in no particular order - food, culture, history, art, architecture, and neighborhoods.

We did a lot of the mainstream history, rah rah America stuff in our last couple of visits along with such gems as the Barnes Foundation and the Eastern State Penitentiary so this time we've been looking at some less mainstream attractions so I thought I'd share.

On the way in, we stopped at the American Swedish Historical Museum in South Philly. Great museum in a gorgeous 100 year old building. The museum covered the early Swedish/New World history (including, for New Yorkers like me, interesting tie ins with Peter Minuet and Peter Stuyvesant) but had so much more. They had a room on inventor John Ericsson and another on Swedish Nightingale, Jenny Lind, brought over to the USA by P.T. Barnum and played all over including at Castle Garden (nee Fort/Castle Clinton) in lower Manhattan. Also lots more including some sketches by painter Anders Zorn and a whole bunch of artifacts. There was also a special exhibit on Swedish candy.

The highlight for Devorah was an exhibit on Swedish weaving:

Today we started off at the Ben Franklin Museum. Pretty cool but nothing unexpected. I recently watched Ken Burn's documentary on Franklin so I was up on his life but it's a nice little museum with some neat artifacts and it was only $5.

Next up was a real gem - the Science History Institute. This free museum has artifacts and the development of science from alchemy through modern day.

A special exhibit on dyes:

And education in science:

Of course, Devorah found the knitting:

We were a little "museumed out" by the end so will have to come back but this is a must visit museum!!!

Next up, a pit stop. We ended up at Shane Confectionary. America's oldest continuously running candy store. They had some old-timey candies but they're really a chocolate place, making products from bean to bar so to speak.

Had a great conversation with the person working there, purchased way too much chocolate and headed off to our next stop - another gem - The Museum for Art in Wood - also another freebie ($5 suggested donation).

Finally, lunch time. We went to a hand pulled noodle shop that we had visited two years ago. Really tasty and super inexpensive:

Our final stop was the Fabric Workshop Museum. Lots of interesting stuff going on behind the scenes but for tourists like us, think of it as a general art gallery where the medium is fabric. There were a couple of really neat exhibits and again, free ($5 suggested). Another winner.

We then picked up doughnuts for later at the Reading terminal market and wandered back to our AirBnB. On the way, we saw a number of murals:

And also these people holding up a skybridge:

Spent the evening relaxing, having reheated leftovers for dinner from yesterday.

Hope you enjoy some of these less known Philly attractions when you get to town. Tomorrow we're going back to the Philadelphia Museum of Art for their special exhibit on Mary Cassatt and will also probabyl hit the Rodin museum and then we'll figure out Saturday.

CS Research by HS Students

Tue, 28 May 2024 14:02:01 -0400

Last week Devorah and I attended the Stuyvesant High School Alumni Association's annual fundraising gala. We were there as guests of the Alumni Association since, as retired teachers, gala tickets are generally out of our budget range.

We had a great time. I spent time with a bunch of former students spanning the years, some current teachers I hadn't seen since I left Stuy, as well as new Stuy alums and current students. I had a great conversation with one particular young man, a graduating Stuy senior, who's working on an interesting app. He asked if he could contact me to talk about it. I told him, I'd love to so I hope he reaches out.

At one point, I was talking to a current Stuy teaching involved in their research program - that is, the STEM research program, primarily focused around the Regeneron Science Talent Search - Westinghouse to us old timers. She lamented to me that she wished she could get one of the CS teachers into the fold - to basically run a CS version of the other research classes that already exist for Biology, Chemistry, Physics, and Mathematics. Each of those classes target students in their Junior years, hook them up with a researcher, and shepherd them through the process of ultimately submitting a project to the big competition.

I had to explain that I actually did run a CS "research" course back in the day but it was actually a bogus cover course since I didn't think that high school students, even with Stuyvesant's multi-year program had sufficient CS and Math to by and large do good research.

The class was created primarily because, at the time, I had set up and was running Stuy's network. I got neither pay nor time compensation for this. I also typically worked with a handful of students. They learned a lot about Linux and networking and helped keep things running. They also got nothing for their troubles - no grade, no pay.

The deal was that we'd create the class, they'd learn the networking stuff and we'd all have some time to keep things running. The kids would also get a nice high grade rewarding their work. They also had to cobble together a "research" project that they could enter into Westinghouse/Intel/Siemans/Regeneron.

Over the years, the class had some students who wanted to do the research thing and were not so much there for the networking side. Those kids had to do a more legit research project. Now, occasionally I had a kid who was able to do an excellent project. Spike's Sporknet project which was a design for a scalable, anonymous network. Really good work and ahead of his time. Brad's work on Captcha was also really great. The thing is, though, that most kids just didn't have the background to do solid CS research.

Even though Stuy's program was and is arguably the most rigorous HS CS program in the country, by second term Junior year, they're just finishing up CS1 (APCS-A). They'd be studying data structures during the second half of 11th grade. Since the research projects were always due in the early November, senior year courses don't have much affect.

On the math side, some might be taking calculus during their junior year but many will be in precalc. Few will have taken statistics and none will have taken discrete math or linear algebra.

These kids were and are tremendously talented but they just don't have the knowledge base yet. On top of that, on the programming side, they haven't yet gotten to the point where they can write large complex programs reasonably quickly. That doesn't typically happen until some time senior year, or the semester after data structures.

Now, in the sciences, the kids were always hooked up with a researcher in a lab. In the best case, they can do legit research. At Stuy, in all areas, including CS, there are usually a couple of outlier kids who had exposure early on and come in with a wealth of experience in their resaerch field. In the worse case though, these kids become glorified lab assistants. The thing is, though, this can be a really good experience. They are there for the entire process and learn a lot. In CS, the equivalent is basically having the kids implement an already designed algorithm or create part of a software system that isn't really showing them the research process. From what I saw, it just wasn't as rich of an experience.

After a few years, I dropped the network - too many headaches and soon after, got rid of the "research" course. I think it was the right choice. The research course could have continued and been a good experience for the couple of outlier kids each year who came in super strong but that wouldn't be a sensible use of resources with each class having over 1000 students.

So, unfortunately, I wasn't able to help this Stuy research teachers case for CS research in high school. I did point out that the kids might be able to leverage their CS and coding skills in their research in other areas. I had one student who ultimately did Physics research who made extensive use of his CS skills for data analysis and visualization. I believe he was a semi-finalist. I've also had former students do Astronomy research in college and grad school using CS skills they got at Stuy.

So, I guess that's it. Am I wrong on this? I certainly could be. Maybe I wasn't back then but times have changed. Sure, if a school had unlimited resources for unlimited classes, having a CS research class would be great even if it only had a handful of students. With budget and space constraints though, I think a full class of another advanced elective is more valuable.

All this said, I'd love to hear differing opinions.

It's teacher appreciation week again

Wed, 08 May 2024 11:34:06 -0400

It's teacher appreciation week again so I thought I'd show some appreciation.

In the past, I've specifically written about some of the teachers who were particularly influential on me growing up. I've also written about how I'm not a huge fan of teacher appreciation week as a whole. Today, I wanted to talk about colleagues who deserve more appreciation from their students and schools than they might be getting. Now, this isn't to say that they're not getting the love but I wanted to add my two cents to make sure.

First, my partners in crime from my days at Stuy. People like JonAlf Dyrland-Weaver, Sam Konstantinovich, and Topher Mykolyk. Not only are they great teachers but whenever I've had a project that benefited either kids or other teachers, they've gone above and beyond. Those three were instrumental in helping build and run my teacher certification program at Hunter college. They were way underpaid, way overworked, and in some (but not) cases, underappreciated by some of the students. Most of the students in the program were great but the third cohort, if I'm to be honest, had a few, let's say, challenging participants and I'll always appreciated my big three for working through and providing the best experience possible for the majority of students.

Now, I also had a number of other great teacher colleagues through the years and I do appreciate them but I wanted to single out DW, Mr K, and Topher (why am I blanking on what the kids call him now?).

The other teachers I wanted to single out today are the teachers who went through my certification program. We had 20 in the first cohort, 20 more int he second, then 60 in the third. The program continues on under the expert guidance of my friend and former student Edgar Troudt but I'm focusing on the three cohorts I worked with.

Yes, all my advanced cert teachers received scholerships to cover tuition but they all decided to give up a summer and spend all day every day in July with me and my motley crew to learn more CS and how to teach it. They then spent the following school year taking courses as well as recording their classes, remote observing their fellow CS teachers and in general supporting each other as they all worked to be the best teachers and the best CS teachers they could become. They worked hard and learned a lot. The teachers who participated who came in with less CS content experience spent countless out of class hours with classmates, TAs and instructors to develop their skills.

As it turned out, with the State's SOCE, it turns out that some, maybe even most of these teachers didn't and might not need the actual certification but they all took it upon themselves to work on improving their content knowledge and their refining their pedagogical techniques so they could be the best version of themselves for their students. It's also worth noting that they all elected to take on my program which I feel is pretty rigorous. At the time, we were the only game in town but if they decided to enroll in a certification program now, they'd have the option of rigorous programs, like mine and a couple of others or by taking one of the less rigorous programs that have since been approved by the state which barely cover any legit CS.

I don't know if their students really know or appreciate that these teachers have really gone above and beyond for them and I'm sure will continue to do so.

So, hat's off to them. A bunch of great teachers super worthy of appreciation. Not just this week but throughout the year.

Should we use MongoDB in High School

Wed, 01 May 2024 11:14:59 -0400

Should we use MongoDB in High School? That's the question. Usually when I give a post such a title, I already have an opinion - it'll be a yes or no and if it's yes, I'll have a plan.

Not this time. My gut tells me there's a place for it but I'm not sure where. Besides, that's not what this post is really about.

At one point during SIGCSE 2023, I stopped by the MongoDB booth. Had a couple of great conversations with the people there. I wasn't looking for how I might use it in my classes - I knew I'd be retiring soon but I did want to convey how much I liked the product and how supportive they've been towards some of my efforts both K12 and while I was at Hunter, I mean I still have more MongoDB shirts than pretty much any other type of swag shirt from when they donated so many for some high school hackathons I've run.

This year, at SIGCSE, I reconnected with my new MongoDB friends from 2023 and we started asking if there were a place for MongoDB in a typical High School. Clearly there was at Stuy but I covered MongoDB there in a senior elective after the kids had already finished a CS0 class as well as a college CS1 and CS2.

Now MongoDB is much more advanced than when I started using it but even back then there was a lot to like. It stored documents essentially in dictionaries which made things much simpler than SQL. You also formed dictionaries to form queries which might have been trickier to "get' than SQL with things like "SELECT NAME from PEOPLE where AGE=30" or something like that but much easier to debug and automate since one deals with a data structure to do it in MongoDB as opposed to string manipulation for old school SQL. It was also helped from the fact that most student projects didn't benefit from the relational SQL model. There was also an open source server you could install and self host. Nowadays MongoDB is much more capable for example it's data lake features.

I don't think MongoDB is the tool to universally use in High School but my gut tells me there's a place for it.

The cool thing to me is that we're now exploring this. Me, a few teachers I've gathered, and some new friends at MongoDB. The interesting thing is how it came about. Many of the K12 company based initiatives were the result of the company (or college) deciding that they wanted to do something in the K12 space. This frequently comes with a certain level of hubris that they can do it better. Not always, but frequently. The small companies and startups then work to generate interest and ultimately market share. The big boys are more frequently anointed since so few K12 CS educators are currently armed to really evaluate what they're seeing.

This was different. Nothing coming from the top. What started as a serindipidous conversation between someone on the education team at Mongo and me, an old education curmudgeon. We both asked "does it make sense?" We didn't know the answer so we're exploring.

Tomorrow morning I'll be going to MongoDB World Local in New York and hope to meet a couple more people on the MongoDB team. In the next few weeks, I'll be talking to some teachers who also hope to explore the MongoDB in high school question. After that, maybe not until the Fall if people are away for the summer, we'll get together and start to get to work.

Either we'll see that MongoDB probably doesn't make sense in high school or we'll figure out where and how. It'll be a fun process and if we do figure it out, you can bet it'll make more sense than any of the typical initiatives that come from above.

CS Certification - I should be exempt

Fri, 19 Apr 2024 13:05:03 -0400

Over the past few years, a number of states have acknowledged Computer Science as an academic area of study and have created pathways for certification. The actual requirements vary from state to state and indeed, even within a given state, there's a lot of variation. Here in New York, for instance there are some strong programs that will truly prepare a candidate to teach CS in K12 but too many others that are just "certificate mills" and the state doesn't seem to care.

The question that I noticed garnering discussion recently wasn't the creation of new certification areas but rather, what do we do with the people currently teaching CS?

Do we grandfather them in? If so, do we do it indefinitely? Do we force them to go through an entire certification program? How about making them sit for an exam? How about something else?

I'm writing about this now because on social media I noticed a post from a teacher who was just informed they'd have to take the Praxis exam in order to continue teaching CS. Here, and in other forums I've heard some teachers shrug this off since they did similar or had to do more, and some say that it was ridiculous to make someone already teaching CS sit for an exam or take additional courses.

I wanted to push back on that.

Just because someone has been teaching CS doesn't mean they should. In New York City, due to their CSforAll initiative hundreds of teachers have been thrust into CS classes over the past few years with minimal preparation. Many of them have worked super hard and have done a yeoman's job but the majority, from what I've seen, don't know CS beyond the super basics and haven't really had the opportunity to learn how to teach it.

On the other hand, there are some who indeed do have a solid CS background and any additional exams or coursework are just a waste of time.

On the third hand, a teacher who has already been certified in some other area and has been teaching CS in a public school shouldn't have to pay out of pocket to take additional coursework or exams.

The problem is that you can't really assess who knows their stuff and who doesn't just because "I've been doing it for the past ten years."

I'm comfortable saying that even though I'm certified in math, I'm a very strong CS teacher but can I prove it? Test scores don't necessarily prove it. I taught for many years at Stuyvesant. Those kids would do well on standardized tests if a potato taught the class. In fact, over my career, I've seen, in regents classes where a class taught by a horrible teacher ended up with some of the highest regents scores. There are reasons for this but it's not worth the time to explain here.

Should we look for student testimonial? Maybe but that could just mean that the teacher was popular and unless you're going to find students who are a bunch of years out, they might not have the perspective to give a fair evaluation.

How about observation reports from supervisors? Since supervisors rarely know CS that won't work either.

It's a tough nut to crack.

So, we're left with some states where they're requiring sitting CS teachers take an exam and/or classes which will be an inconvenience or burden on the few teachers that already really know their stuff and other states like NY where they're fine with unqualified teachers for the next ten years by having teachers merely fill out a form, which I wrote about here.

Neither solution is ideal.

Probably the best solution would be to have current CS teachers sit for an exam like the CS Praxis exam and have the state pick up the fee. I'm not a big fan of standardized exams but as far as they go, the CS Praxis isn't horrible and if you pass it, you should have the CS knowledge necessary to teach most High School CS classes. If you waive the fee, then it isn't really a big ask for a qualified CS teacher. Maybe you also only do this for teachers who've been teaching CS classes for two or more years or something like that. Combine a no cost exam with the teachers past CS teaching track record and you probably hit the sweet spot - fair for the teacher and you have a pretty good idea that your CS teachers know their stuff.

Seems like an easy common sense solution and it wouldn't cost that much. Instead we get states requiring way too much of working teachers or on the flip side, allowing teachers with no credentials to teach our students. I guess we've learned not to expect more from our governments.

SIGCSE 2024 - Papers, Experience Reports, and Lightening Talks

Mon, 01 Apr 2024 19:36:48 -0400

Last up we have paper sessions, which include experience reports and lightening talks.

Probably won't write as much on this - I'm a little SIGCSE posted out.

One interesting note was a semi-recurring theme of "the hidden curriculum." I say semi-recurring because while it appeared in three sessions I attended - two experience reports and one lightening talk, each time, the speakers were referring to a different "hidden curriculum."

The first time was a talk from UCSD describing a projects course. It was similar to many projects based courses - students are shepherded through a large project, but in this class, students worked exclusively on large existing open source projects. The hidden curriculum consisted of all those real world tools and techniques not covered in a typical CS sequence - testing, code maangement, tech integration, navigating a code base and more.

The course sounds great. If given a single choice, I'm not sure if it's better for students to work on a large original project or to learn how to work in an existing one as both skill sets are valuable and important but do like what they described as going on at UCSD.

The second time the hidden curriculum came up was in a lightening talk from instructors at the Milwaukee School of Engineering. What was interesting was that they were talking about something altogether different from what UCSD considered their hidden curriculum. In tnis case they were talking about things like study skills, stress management, homework strategies, registration and advisement, time management, and life after graduation. While I certainly appreciate everything about the software engineering skills form the UCSD talk, the skills talked about here are arguably more fundamental and they're skills that many students, particularly those from lower income households have never had the opportunity to develop. This took me back to our first keynote when the speaker mentioned that he while he got great support from his family as the first one going to college, he got very little help. His family supported him and wanted him to succeed but as nobody else from the family had ever gone to college, they didn't have much to offer in terms of concrete help in navigating this new world. At the school. the instructors taught their hidden curriculum through extra curricular sessions during common free times (providing free food to draw in the students).

The third hidden curriculum was a talk titled "From Degree to Developer…" In this case the hidden curriculum as interview prep both in terms of the soft skills for interviewing and also the technical interview skills - how to pass those LeetCode problems. The solution in this case was like the first talk - create a course that was essentially an algorithms course with interview style implementations along with the soft skills.

What struck me here is that we've got three different institutions all resorting to something outside of their school's major requirements to address a real student need (the two courses that were created were free electives). Something's wrong here and unless it's addressed we'll see the gap between the haves and the have nots continue to expand since well to do students at elite institutions are best able to learn these hidden curricula on their own.

Personally, I've found that if you're strategic you can add a fair amount of the software engineering practices to core CS classes without too much hassle - certainly if you tweak the content a bit - do we really need every single one of those data structures and do we have to dive so deep into proof and theory? Maybe we should realize that our programs are still set up for 100% percent of our grads to go on to get their PhDs where the reality is that the majority want to take tech skills to industry. I'm betting programs could easily accommodate most of the hidden curricula with some sensible re-evaluation of requirements and everyone would be better for it.

One other session that I want to mention before I wrap up was one on "Creative labs in CS1" at Georgia Tech. This was interesting. In this course, they added a handful of what they considered "fun" projects. The concepts would not be tested on exams. These projects would not be in Python, which I guess is the core language for the CS1 class. The first project might be an HTML/CSS web site. The second, maybe adding some JS. To be honest, I didn't get this at first but then the speakers explained that at Georgia Tech, CS1 is almost entirely made up of non-majors. I'm still not sold on their plan here but I think I get what they were trying for. What got me was the fact that this is a real problem and not just for Georgia Tech. We're seeing more and more schools requiring students to essentially pre-declare in CS before they're even admitted to the school. Kids without sufficient opportunities, as a result, are being locked out of the field. I know we've got a booming enrollment problem but we've got to do better.

So, that's it for the sessions. I think I'll do one more SIGCSE post as a general wrap up and talk about our Portland adventures. Then it'll be back to my usual curmudgeonly rantings.

SIGCSE 2024 - Panels

Mon, 25 Mar 2024 14:23:03 -0400

I generally enjoy both panels and special sessions at SIGCSE over paper sessions. Don't get me wrong, paper sessions can be great but with those you can always read the paper and all to frequently there's not a lot of value added beyond the paper itself. Panels and special sessions can't just be read in the proceedings.

For one panel session I had to choose between

Enduring Lessons from ‘Computer Science for All’ for AI Education in Schools

and

AI in Computing Education from Research to Practice

I opted for the latter. The panel was interesting but not what I expected it to be based on the title and description. The moderator provided a list of questions and the audience voted to order them using an app. The moderator then asked the top ranked questions along with additional audience questions.

Many of the questions went over what is now well trodden ground - will kids go into CS if AI will take their jobs? How can kids be motivated to learn to program since they have AI assistents? How can we detect the use of AI tools in assignments? What are useful current AI tools? Not a lot new here.

Still there were some interesting points that came up.

One panelist discussed how through the use of AI early CS classes, CS0 and CS1 can get much further - student can do more advanced work more quickly. I started to wonder - has anyone actually checked this? I mean, yes it's easy to see that they can get through more material but do they learn it? Have the later courses been adjusted to accomodate? Has anyone studied the AI enhanced CS0/1 students vs those without it in CS2?

I've seen countless times where students seemingly achieve advanced results through the use of libraries and other tools along with extreme coaching. Many code schools as well as some highly lauded youth programs do this but the students invariable have little retention and hit a wall or regress soon after the class or camp ends.

Now I'm not saying that you can't make a better early CS experience with AI. I'm just saying that one, it's not as easy as saying "they can do more" and two, the jury's still out.

Another "positive" the panel brought up was that LLMS can help or guide students at scale - LLM based TAs and tutors. Now, I'm all for adding resources for students but people can be so hyped for some new technology they miss the problems. In this case, adding LLM based tutors to hide a courses shortcomings - something I wrote about dealing with CS50 here. Another potential danger I see is using LLM based teachers and tutors to replace human interaction.

Another concern that came out of the panel when a panelist claimed that AI tools will allow for the creation of a "curriculum in a box" - the panelist framed this as a good thing - a school could offer a class they otherwise couldn't but I read it as another sign of the deterioration of the craft of teaching.

I'll end my thoughts on this panel though with an LLM strength - that they can be multilingual. I haven't seen this exploited yet but this could certainly be a boon.

My second panel was

Data Science and Computer Science in K-12

The panel started with an interesting fact. In 2020 no states had anything in their requirements or standards dealing with data science. Now in 2024, nineteen do. That's some pretty fast growth.

The panelists brought up a number of issues:

Data Science and Computer Science frequently overlap
Data Science provides opportunities to develop CS skills
CS has DS in their standards but frequently they need to be expanded on
The question as to if DS (and CS) should be standalone classes or embedded in other subject areas

It's doubtful that data science will spin off into its own discipline in K12. It's much more likely to be integrated into other subjects but the placement of data science and computer science is an interesting one. Currently, the trend in the early grades is integration. In the later K12 grades you've got mainstream players like the College Board with standalone classes like APCS-A and APCS Principles but also integration efforts like Bootstrap which integrates into classes like high school algebra.

Now, in general I'd love to see more subject integration but looking at CS in particular, at least at the high school level, I've always pushed for separate classes.

Why? Well, first off, separate classes means separate teachers and ultimately departments. Yes, this further silos the system but without it, schools won't have long term advocates for compute science and the subject will be permanently relegated to second class citizen.

Another problem with integration is that subject teachers are overloaded as it is. Now, a science teacher or history teacher might already be doing some data science and can just relable things but CS is another matter.

At the panel, I thought of another reason. What happens when the curriculum changes? Bootstrap is integrated into a typical algebra curriculum. What happens when the curriculum changes. On the one hand, if the curriculum is just tweaked, things will probably be fine but what if there are bigger changes. Over the course of my career there were changes to both the Geometry and Algebra 2 and Trig curricula. Big enough that if the embedded curriculum is scripted and the teacher doesn't really know the CS well, that add on is going out the window. This will likely be the case with minor changes - moving or removing a single unit or changing an emphasis but what about bigger changes - adding calculators in math or the types of changes we've seen in APCS-A over the years.

This isn't even counting major changes like when we went from Algebra to integrated math to a new version of integrated math back to Algebra all in my career.

This is all to say that integration probably isn't sustainable if you're just providing a drop in and some PD. It can work if the teachers are truly trained to know both their core subject and the CS or DS but that's a big ask.

So those are my thoughts on the two panels I attended. Good stuff and good food for thought. Next up paper sessions.

SIGCSE 2024 Birds of a Feather Sessions

Sun, 24 Mar 2024 10:30:26 -0400

I saw three keynotes, I also went to three Birds of a Feather sessions (BOFS). Three was possible as SIGCSE added a third "flock" this year. I didn't take notes since BOFS should be pretty interactive so I won't be able to give too much in terms of content summaries bu hopefully I'll be able to convey the gist of what went on.

Two of the BOFS I attended were good, the third was, if I'm to be honest, pretty awful.

Here's the rundown.

The first session I attended was "Creating University Teacher Training Programs." I figured I pretty much had to attend this one.

Overall, this was a good session but it did suffer from a couple of insurmountable problems that I think all the BOF sessions had. First, there were a lot of participants, second there just wasn't enough time. This meant that we just didn't have that much time to get into meaningful discussions. We were able to share some basic facts about our state's certification processes but we couldn't get much further.

Most of the BOFS I've attended at SIGCSE follow the same basic format. The session leaders say a few things, the room splits into groups of varying sizes, the leaders give questions, the groups discuss. Between questions, there's usually a share out. This takes a lot of time. It might be worth it if SIGCSE made one flock of double length rather than two flocks or if organizers split their sessions into smaller groups and just let them have at it without the intermittent reconvening.

In any event, this session covered the basics and was worthwhile but I would have liked to have been able to talk about actual program designs with the other session attendees.

The second BOF I attended was titled:

K-12 CS Teacher Education: What are we teaching and to whom? How do we know we are moving the needle?

with a sub heading that included:

…it is imperative that the CS education community understand the myriad ways in which CS teacher education is enacted and begin developing best practices for the field.

I was hoping that this would be a dive into what and how different programs were teaching. I couldn't have been more wrong. The session was run by two people but only one really facilitated. I won't go into details but I will say that the individual who actually ran the session displayed poor teaching practices in working to direct the groups in specific ways rather than facilitating an open discussion.

I left midway through and had a much more valuable hallway track discussion.

My third BOF was the following day. It was titled "Computer Education and Capitalism." Like the first BOF I described, this was a good session but fell short on time. The organizers framed the discussion but as "computer education and capitalism" can mean many things, they gave groups freedom to explore as each group wanted to. This meant we could talk about education for jobs vs pure education, capitalist influences on curricula, teacher prep, and how teachers have to teach or pretty much anything else. Overall strong session. I think most people, including myself left with a new wrinkle in our thinking.

So that's it for today. Two strong BOFS and one bad one. I think it might be cool if there were a way for each BOF to summarize the discussion to share back to attendees. Sometimes it feels like we come together in a room, have a discussion with new friends and then it's all lost in the ether. Still worthwhile but maybe they could be something more.

SIGCSE 2024 - keynotes

Sun, 24 Mar 2024 10:29:30 -0400

I wrote about the first keynote in my day 1 post. Today I'll talk about two more. There were two additional keynotes during luncheons but I didn't attend those.

First up was Jane Plane titled "Find Your Drop and Add to the River." Strong keynote but it spoke to me in a very different way than the first keynote. The first keynote was maybe the first talk I've heard by someone thought of as an education expert who actually spoke the measured truth rather than eduspeak propaganda. This keynote resonated as I've recently been thinking a lot about the drops I've either contributed or failed to contribute to the CS Education cause.

"Drop by drop a river is made" was the lead quote. It's a translation of an Afghan proverb. Plane's plea was for us all to find our drops to contribute and when dealing with CS Educations major issues - diversity, equity, and inclusion - it's going to take all of us adding our drops and while any one action might seem insignificant, they add up over time.

The talk started with Plane's journey. How she had the opportunity to and then actually built the first Masters in CS program in Africa and then later did built similar programs in Afghanistan. She also spoke of the drops she added at Maryland and now in Wisconsin. One might say that not everyone will have those opportunities but the truth is, most people, if they did have them, probably wouldn't have taken them nor accomplished as much. Regardless, even more mundane sounding drops can big time contribute to the river.

In the first keynote, Todd Zakrajsek told of how he almost dropped out of college but the right words at the right time from one single professor changed the course of his life. Without those words, none of his drops would have been added.

In the micro-teaching session, one instructor mimed an interaction he had with a student. Before "class" he asked what the student planned to do after college. She said, she wasn't sure. The instructor encouraged her to come to office hours to discuss. The "student" was at my table and said afterwards that the interaction was real and ultimately it what led to her becoming a CS educator. That's a lot more drops.

The other day I noticed a comment on LinkedIn by my friend Leigh Ann DeLyser. Leigh Ann commented about the number of "multipliers" at SIGCSE. Made me think of the fact that teachers in general are tremendous multipliers. One word can send a child in one direction or another and the drops we add can turn into streams or more. As a high school teacher, I was charged with 150 or so students each semester. If even a small number, say, become teachers themselves, or go into tech and as a result of my classes pay more attention to DEI, well, that's one drop that's turned into a steady flow. We might not all be creating new programs in other countries but every teacher has an opportunity to be a tremendous influencer. We just don't get credit on the internet.

The other keynote I attended was given by Rachel Rose of ILM. The talk started as a journey talk - from early interests through college and ultimately to ILM. One point was when Rose emphasized that she collaborated more in her CS classes than in her other classes. This was interesting. It makes sense - particularly in project based courses but flies in the face of the stereotypes. It's also interesting in light of the ongoing tensions in schools between collaboration, individual assessment, and academic honesty.

The second part of the talk was a showcase and inside look into the technologies Rose's team developed at ILM for use in productions like "The Mandalorean", "Solo," and others.

Next up was a talk about the Academy Software Foundation. How the movie industry banded together to form a foundation and how each major studio contributed open source platforms. Now the industry can use standards and tools like materialx, opencolorio, and openfx. You can check out ASFW here. Looks really cool and having all the major players rally around the virtues of open source is really cool. My only issue with them is that their logo has an open comment with no close comment and that just bothers me.

The final theme, which appeared throughout the talk was the industry's work to increase diversity. A major point Rose made was that youngsters have no idea as to what they can do within the movie industry. Neither do they understand the educational path from school to Hollywood be it in tech, the arts, or some hybrid of both.

All three keynotes were strong. All conveyed worthwhile messages and all were delivered in an entertaining way.

Good job SIGCSE2024!!!!

Still to come - BOFS, Panels, and Paper sessions.

SIGCSE 2024 - Microteaching

Sat, 23 Mar 2024 23:42:56 -0400

We stayed an extra day in Portland to do some site seeing. I'll write about the tourist side in my final SIGCSE2024 post. Right now, we're back in the hotel resting before dinner so I thought I'd start the next part of my SIGCSE2024 report.

For Day one's first session, I attended Colleen Lewis's Microteaching session. I've wrote about this session last year and also about Colleen's recent appearance on the CSEd podcast where she discussed manipulatives. Much of Colleen's public facing work involves CS pedagogy and if you haven't noticed, I'm a fan.

Microteaching has fast become my favorite SIGCSE session. Six teachers each give a quick, abbreviated mini-lesson and afterwards Colleen highlights pedagogical techniques.

It's not perfect. The mini-lessons are taught out of context, the audience aren't the presenters real classes, and it's only something like a 10 minute snippet but still, it made for a great session.

In one mini-lesson, Colleen pointed out that the presenter gave the "class" (us) time to process a question before asking for responses. Questioning technique and timing is an important part of K12 teacher training and I'm betting it's not covered nearly enough at the college level. There's a huge amount to explore on questioning but a simple technique like this - waiting say, 10 seconds after asking a question before entertaining a response can be huge.

In another lesson, Colleen commented how the teacher provided to concrete but different examples to set the stage for a class exercise. I liked that particular lesson, delivered by Adam Blank because it wasn't a core lesson in his class but rather a fun side topic to set the stage for the cybersecurity work yet to come. He elicited suggestions from the class how they could first crack a safe (using the typical movie trope of listening with a stethoscope) and how to crack a keypad lock by looking at the worn out keys on the keypad. This led to a class "lab" where we had to figure out which mystery toy was in an egg shell without opening it. All of this was to teach the class about side channel attacks. It was terrific.

This went on for six mini-lessons in all.

It was a great session but I had one thought that might make it better. Instead of six mini-lessons, how about only three so that we'd then have time to discuss the lessons. This could be tricky because it puts the people presenting on the spot and open to criticism but it could be so valuable.

For example, in one of the mini-lessons, the instructor presented a binary tree. Each node had a student name on it and also a priority value.Later the teacher removed the names. Colleen highlighted that putting the name in to the tree (along with priority values) was a strong technique as was later removing them. I partially agreed with this but also felt that removing the names could also have weakened the lesson. We didn't have full context so I can't be sure. In any event, this could have led to some really rich discussion.

A second example, and one I loved was when one teacher engaged the "class" prior to the lesson. He asked a student their future plans and after an exchange invited the student to come to office hours to discuss. This was a great example of a hidden technique for a number of reasons. One reason in particular I'll address when I talk about the other SIGCSE keynotes but it's also a technique that can be a negative. On the plus side it's a way to engage the class at a more informal manner, it can be used to build class community and culture, and also, if it's a morning class, start to get sleep minds stirring (not to mention the plus I'll talk about later). On the other hand, if you do this consistently with just a couple of students, you risk building resentment or concern from the students that aren't part of the "in crowd."

All in all, a super session that gave attendees a chance to see aspects of teaching that are never covered in research papers and rarely talked about in general. Looking forward to next year's edition.

I think I'll organize the rest of my SIGCSE report into session types - BOFs, paper sessions, and panels so stay tuned for more.

SIGCSE 2024 - Opening Day

Sat, 23 Mar 2024 22:34:50 -0400

Last time I was in Portland Oregon was March 2020. Came in for SIGCSE - the big CS Education conference. Got in a couple of days early, played tourist, and then morning of, everything was canceled. Flew back to NYC on a near empty jet and Covid became very, very real.

March 2024 and I'm back in Portland and SIGCSE's giving it another go. This time I came out with my daughter Batya who's developed an interest in CS education and my wife Devorah who's been playing tourist while we've been conferencing.

The conference kicked off with a keynote by Todd Zakrajsek of UNC. I was ready to be disappointed. A researcher, that is, not a teacher who's spent decades all day every day in the classroom, who's published lots of books about how to teach and he's going to tell me how to teach.

I couldn't have been more wrong. Content wise, he was spot on. Maybe the bet SIGCSE keynote I've attended. Delivery? He was great there as well.

He gave some backstory - first in family to go to college. Almost flunked out, wanted to be a cop - never thought of becoming an educator and over the talk also kept hitting a few key points.

One of the overarching themes was that there is no single true way. Right from the start, he emphasized, for instance that lecturing wasn't necessarily bad and that active learning wasn't always necessarily the best. Lecturing can be effective but good lecturing with some good active learning smatterred in can be even more effective. He later said "I love constructivism when I have the time." I loved that. I spent five or six years around a decade ago hearing from the "experts" that constructivism was the one true way ignoring the fact that it was neither practical for most public school CS teachers nor was it optimal due to a variety of constraints. It was refreshing to hear this expert say what my team and I have been saying for years. There are lots of ways to teach and what's best at any given time depends on a lot of factors and the most effective teachers figure out what tool to use when and where.

Related to this, Zakrajsek emphasized that failure was a combination of student and teacher. It's not solely the student's fault nor solely the teacher's. This is related to another pedagogical thought I've long pitched - the best technique to use at any given time is best determined by evaluating a combination of student needs and situation, teacher's strengths, and contextual circumstances.

Even though this first theme considered both teacher and student, it came across to me as a teacher issue - ultimately, it was how to teach. Another theme I felt was much more student centered.

The focus was introvert and extrovert.

Group work? Another "one true way" to teach that has been foisted upon teachers on and off for decades. Zakrajsek claimed, and I agreed, that while some students love and thrive under it, some hate it. Similarly some like and do much better with lectures.

He also spoke of the extrovert who would raise their hand before the question was even finished vs the introvert who took their time and pondered before even considering responding and that as teachers we must work to accommodate both.

I chuckled when he told a story of when he was a guest speaker in a class and when a student kept raising his hand before the question was even finished, he'd call on him and ask him to answer. I chuckled because what he did was very similar to how I handled a similar student in my class.

Back to the group work, another theme was that students don't know how to do it unless you teach them - the hidden curriculum and that's frequently ignored.

This led Zakrajsek to another theme - that when he started he thought teaching was about content but later he realized it was about learning to learn.

This resonated with me. Zakrajsek was 100% right but it's 100% not how teachers are prepared today. Too often teachers are not given the toolbox of methods and the deep content knowledge to develop educational experiences appropriate for their populations. Rather their given canned curricula, scripts, and instructions - what to teach and how to teach it. Far too many of today's teachers are trained to deliver specific content in a specific way and far too few are prepared to teach students how to learn how to learn.

Todd Zakrajsek's keynote was really amazing. Maybe I'm just saying it because it echoed so many of my thoughts about teaching and teacher preparation and he was really entertaining but I'll stand by that statement.

It was a great way to kick off the conference. There was a lot more to come on day 1. A great session on teaching, some papers on software engineering, lightening talks, and then a god BOF and a really bad one.

I'll save those for my next couple of posts.

For now, I'll just say, if and when Todd Zakrajsek's keynote is made available online, watch it.

The SAT - should it be required

Mon, 11 Mar 2024 15:15:57 -0400

I saw a few days ago that Yale was reinstating the SAT. They're joining Dartmouth and Brown of the Ivies as well as MIT. I have no idea what percent of applicants, let alone percent those receiving offers, took the exam anyway for other school applications.

Some of my friends thought this was great - a return to standards!!!! Others felt it was a regressive move and it will hurt diversity and equity.

For me, I think it's all about branding.

As my friends know, I'm no fan of the college board. They make standardized tests and do their best to both make money and to influence education through programs like the SAT/PSAT and the AP program. Many have noted that a high SAT score is not a reflection of intelligence but rather of income. This makes sense. Children from well to do backgrounds will have better educational opportunities, more support at home, and be better prepared for a standardized test. It also makes sense that these students will be better prepared to face college where life is less forgiving than high school. It's not fair and it's not right, but well to do kids are better set up for college success.

If I had my druthers, both the SAT and the College Board would just go away.

That said, let's dive in to why I believe that the whole SAT required vs SAT optional isn't really about building an incoming class, equitable or otherwise, but rather about image and branding.

For "elite" institutions, image and branding is huge. The truth is, you can get a class for class equivalent education from a second or third tier school as you can from Harvard at a fraction of the cost. You can't get the resources that Harvard provides due to its immense wealth nor can you get the old boys club connections nor the name that opens doors even if you're less than competent. I've seen this time and time again in industry where a candidate from an Ivy or similar will get offers, opportunities and promotions while just not being that good whereas someone from a no-name school has to actually be good (yes, I know plenty of Ivy students are actually quite good but I've met plenty of dopes).

Go SAT optional and you appease your more liberal fans. SAT required? The elitists and "meritocracy" crowd will cheer.

Image and PR win the day.

Take CS50 for example. It's the golden CS0 course with all the hype but if you look more closely it's a flawed implementation of a flawed course. Not horrible but nothing special. It wins because it's from Harvard's rep and ballyhoo - a charismatic barker at the helm and a class filled with students preculled to succeed by Harvard's admission process. Nevermind the huge waits for TAs when projects are due or all that cheating we heard about a couple of years ago. It's Harvard and it has hype so it's good.

Same for the SAT. You've got a high SAT cut off? You must be elite!!!

Elite institutions can shape their incoming classes however they want. A few years ago Harvey Mudd proudly achieved gender parity in CS. Now, Harvey Mudd could have and probably did make some changes in what they did once students were in their walls, but let's look at shaping the incoming class.

I found data for accepted students for Harvey Mudd online. I didn't check the year - it doesn't matter. Here are the raw numbers:

	Applied	Offered	Enrolled
Men	2931	271	116
Women	1509	322	121

That's 13% acceptance rate and a 40% yield.

They also boasted a 3.96 GPA and an average SAT score of 1520.

Note that more women were both offered and accepted in spite of only roughly half the number applying. I don't have older stats but I guess 10 years ago the offers skewed more male.

With just a 13% acceptance rate and Harvey Mudd's reputation it's easy to imagine that a HUGE number of qualified applicants were rejected. I've never looked at the numbers for Harvey Mudd applicants from Stuy but typically, for elite schools, far more qualified applicants are rejected from Stuy than are accepted. In any year, I might have five students accepted to say MIT, another 20 or more that were clearly qualified and capable of doing great and a handful that probably were right to be rejected. What's more, each year there were some rejected and the teachers all wondered, how did person A get accepted when person B didn't.

I'm betting it's the same in general for the application pools for all "elite" institutions. All Harvey Mudd had to do was skew their offers more female noting that the applicants they offer should have an inclination to go the CS route. They could do the same for minority applicants. None of this really changes the big picture. They're just rearranging the deck chairs on the titanic - grabbing the high performing CS inclined women who might otherwise go to, let's say Stanford or Berkeley.

It's like a friend who once lamented to me that even though the FANG companies (yes, old acronym - old conversation) made all sorts of claims about equity and diversity, their equity hiring program though consisted of out battling the other FANG companies for the handful of black CS students at "elite" universities.

Harvey Mudd or any of these institutions could probably keep the handful of "superachievers" - Westinghouse winners, IOI and IMO team members etc. from their class offers, throw everyone else out take the next n students in and the class would probably be remarkably similar and perform just as well. They could probably even do this a second and maybe a third time.

They can shape the classes however they want to achieve whatever look or result they want.

They can even let in students that require more academic support and still end up fine. All those division one athletes that didn't "come here to play school" are prime examples (of course, there are plenty of D1 athletes that did come to play school and are terrific students) - they're shunted to easier majors and receive huge amounts of support that regular students can't dream of.

Malcom Gladwell once wrote that the only fair way to form Harvard's incoming class would be for them to take all the qualified applicants, put them in a hat, and randomly select the offers. He was right and given the way colleges run their processes they can do what they want and come up with whatever they want, SAT or not.

Okay, so given this, it should be clear that requiring the SAT or not is really a nothingburger but I'll go further and say all requiring it does is that it puts stress on high school students and money into the coffers of the college board. It doesn't really give you a "stronger" incoming class, at least not if your admissions office is actually doing its job.

I spent my years at Hunter reading applications for my undergraduate CS honors program. I also spent a few years reading applications for the Macaulay Honors College - CUNY's elite branch.

For the first few years, we looked at SAT scores. For me, they were only useful for one thing - a very coarse math cull. My students were all coming in as intended CS majors. That meant they would have to take calc 1, calc 2, discrete, linear and a couple more math classes. If a student had a very low SAT math score, it was a red flag. SAT score didn't really tell me more than that.

Even then, I had to match this against transcript data. If a kid was struggling with geometry in 11th grade, they'll probably struggle with calc as a freshman so maybe not a good fit for my program. In fact, the transcript information - what class, what grade, and what year was far more valuable than SAT score in determining if a student was prepared for my program.

Now you might argue that high school grades are inconsistent and not every state has course level standardized testing like the New York State Regents Exams. I'll tell you what, it doesn't matter.

In very short order, I got to know schools and classes. I also got to know recommendation writers. By year three I knew what a grade in a given class meant for most schools in my applicant pool and how to read the recs. Sure, I'd have to merge in students from unknown schools but after a while you get a feel. The SAT was a nice little crutch but wholly unnecessary.

But, you say - you only had a few hundred applications and they were mostly local. Harvard and MIT have applicants from all over the world.

Yeah, they also have big admission offices with full staffs. I know that the "elite" colleges have specific people who read applications for sets of schools. Over the years, I've met the people who read Stuyvesant's applications from a few of these institutions. Like me, they knew what the courses meant at Stuy and their other schools and they also knew the recommenders - who were straight shooters, who dialed it in etc…. As they say "you gotta know the territory" and they did. I won't name names or schools but I bet if you honestly asked these readers that I've met if they really needed SAT scores to make a decision, they'd say no.

So, there you have it. SAT or no SAT? I'd be happy to get rid of it but everyone should recognize that it's really about image and not about forming an equitable and/or diverse incoming class. Colleges can do that already it's just that many don't really care to.