cseducation on C'est la Z

Blog For Self Professional Devleopment

Thu, 16 May 2019 07:12:44 -0400

Lat day of the 2019 NCWIT Summmit. Got up crazy early, got back from my run and saw this when I checked Twitter:

A1: CSTA, Twitter chats like this one, @guzdial and @alfredtwo's blogs among others. In all cases because those places start from assuming I'm a CS teacher, rather than starting from assuming I'm a teacher who is better at teaching something else. #csk8 https://t.co/6DO77fzq0j
— Sarah Judd (@SarahEJudd) May 16, 2019

I don't usually participate in Twitter chats. The format doesn't really work for me. I do frequently lurk though and while I didn't see this tweet until the morning after, another one by my Todd Lash caught my eye and I'll probably comment on that at some point in the near future.

On Sarah's tweet I'll say that CS teachers should certainly frequent Alfred and Mark's blogs. I subscribe to and enjoy both. I'll add something that I've said before and that is if you're a CS teacher you should also blog.

I get it - it takes time, it's hard to put yourself out there and if you're a relatively young teacher you might feel like you don't have much to say but I say so what.

I've been blogging on and off for nine years now. I have no reason to believe I have anything more than a small readership but I do periodically get surprised at a conference with a "hey, I love your blog." Comments are few and far between and from "the usual suspects" but as I sad above, so what.

Pick a lesson you just did and describe it. That's all it takes. Try something different a week later? Write it up. Rinse and repeat.

Why have I found writing a blog to be good self-PD?

Writing the experience or the lesson makes me think through it again. What worked, what didn't.
When I do a similar lesson in the future, I can look back on what I did. It helps in iterating towards becoming a better teacher.
I get to see how I've evolved as a teacher over time. You don't get this immediately but blog for a few years and then reread some early posts.
You could get some comments. I don't get a lot but the ones people have left have proven useful to me. Some of the ones from people who disagree have been the most useful.

On top of this, some other new teacher might discover something you wrote up and find inspiration from it.

So, jump in and try it. Blogging for self-PD.

Implicit Curricula

Sun, 07 Apr 2019 16:23:43 -0400

The state got back to us the other week on our CSEd programs. Still no approval.

One of the requirements is that we have 12 graduate CS credits in our program and that in those courses these five major areas:

Algorithms and programming
Computing systems
Data and analysis
Impacts of computing
Networks and the internet

as described in the K12 CS Framework. Even though many of the items described in the framework fit better in an education side course NY is requiring that it's all covered in CS courses offered by a CS department. The state has more information here.

To be honest, I'm not a fan of the K12 Framework and others have also voiced their concerns about it. Ultimately I think many states will use it as a certification guide and it will open the door to substandard preservice programs that will certify teachers lacking in CS content knowledge.

Still, the programs we designed at Hunter more than covered all the required bases.

So, why did the state come back asking for clarification? Because not everything was explicit in our sample syllabi. As we all know, Syllabi are usually rather terse documents with just a couple of lines per topic, maybe also an assignment and some readings. If they contained all the richness of all the lessons they'd be volumes rather than the usual page or two. So, to provide the requested clarifications I've been spending the weekend breaking them out to show more detail.

This got me thinking about the concept of implicit items in curricula. Much of the richness of a lesson comes not from the description of the topic in the syllabus but also from related and ancillary items that aren't specifically listed - implicit items. One example I wrote about before is the Towers of Hanoi. When I teach it, on the one hand, it's a lesson about a recursive solution to a problem but on the other hand there's all sorts of other richness to the lesson. We talk about

Proof
Run time
Hard problems
Moores law and the limits f computing power
Alternate approaches to problems

None of which is in a typical Towers of Hanoi lesson plan nor in most text books yet to me they're the most important, interesting parts of the lesson.

Another example is when I talk about changing the data structure rather than the code when I teach search.

In our Ed proposals one such place was when we move from using an online IDE to a local development environment and discuss all sorts of ramifications another is when we create a web application in our topics class and things like authentication and security are discussed.

I spent most of this weekend going through our proposals and making sure it was clear everywhere we covered something relating to those "5 areas" of computer science.

The unfortunate thing is that of curricular material and knowledge isn't easy to collect and share with others.

I and others have written about lesson ideas on our blogs but there's no searchable, shareable, discoverable place for either basic lesson material or these rich implicit items. I wrote about EngageCS and Nifty Assignments before but they're really core lesson ideas and aren't really easy to find grab and use.

I guess the closest project moving in this direction is probably CSSplice Curriculum Materials Working Group that I found out about from Cory Bart at this past SIGCSE.

Check out the project link and if you've got a great lesson with lots of implicit learnings going on, please share it in the comments.

Use A Real Language

Sat, 09 Feb 2019 15:08:18 -0400

Why can't we use a real language!

This topic has been coming up a lot recently. Now I'm not talking about the Drag and drop vs textual language thing. Let me be clear. To me a Drag and Drop language can certainly be a real language and many are. I also think they're terrific when used correctly. I just think they're frequently misapplied in later grades.

No, I'm talking about people asking things like "Why do we have to use Java in our class, why can't we use a real language like __." Or at Hunter, I might get "Why do we have to use C++, why can't we use Java." Back at Stuy I'd get the same question about Scheme and Netlogo. It actually doesn't really matter what language you use. Someone else, usually a non-teacher always knows better.

All of this reminded me of when I was young.

I started with BASIC and then in high school programmed Fortran IV using punch cards on an IBM 1130. At the time BASIC was considered a toy language and Fortran IV not to mention punch cards were yesterday's technology. I also took APCS the first year it was offered and that's where I learned Pascal.

I then went to NYU - The Courant Institute of Mathematical Sciences. There we also used Pascal as the core language. That was the language used in CS1, Data Structures, and Algorithms. It was also the default language for other classes. That is, unless the class lent itself to a specific language such as Lisp for AI you'd probably use Pascal. I learned C early on in my college career and also took classes that used Ada, and 8088 Assembly. I mostly tried to stick with

It was the hot language at the time and all the cool kids were

using it. Some kids continued to use Pascal but it was universally derided. It wasn't a "real language." Nobody used it for anything serious.

Senior year it was time to look for a job. I ended up going to Goldman Sachs and developed for PCs running Windows 2.1 in C. They also hired a bunch of college grads to program the Stratus systems in PL1 and Cobol programmers for the mainframes. This was around 1989. Goldman and most other big companies started you in a training program. I already knew C but they ran a crash course for my group. They did the same for the Cobol kids although most of them had learned it in their business schools. The PL1 kids were grouped in with the C and Cobol groups and had to figure out PL1 on their own once they were assigned teams.

My story tough, isn't about Goldman. It's about another big bank I interviewed with. I won't name them but it's a big firm that still exists today. I think they had at least a whole big building down on the street back when I interviewed but I don't remember. The interview process was tough and thorough. We talked about the firm, what they did, the platforms they developed and of course they drilled me on what I brought to the table. The funny thing was that their entire PC platform was built in, yep, you guessed it, Turbo Pascal.

Here was a really respected financial firm doing really interesting tech and they were using that fake toy language that no kid would ever use for real.

So there you have it. All this "it's not a real language" stuff is largely nonsense. If it can solve your problem it's real. If it can teach your concept it's real. The rest is nonsense.

As a side note, my first programming gig was in the mid 1980s for a ticket printing company - Arcus Simplex Brown. I worked on a computer ticket printing system for them. It was in BASIC and ran on a little PC running CP/M with a big line printer. Best I could figure they were still using the same system in 2012.

So, don't get caught up with someone else's opinion of what's a real language just make sure your students are getting what they need.

CS For All The Time

Fri, 07 Dec 2018 08:20:42 -0400

A question that frequently comes up with respect to CS for All is what does it mean?

To me it means giving all students some fundamental exposure to computer science so that they can:

Make use of those thinking skills, problem solving approaches, and practical tools regardless of their future paths.
Provide a path for the subset of "all" that desire to further pursue computer science.

But even with this, how do you do it?

Does it mean that you require computer science? If so, do we require it every year? some years? One time only?

Does it mean you offer it as an elective? If so, when?

I think about this a fair amount but decided to write something up when I saw this tweet:

STEM related career interests are largely developed before high school (Sadler, Sonnert, Hazari, & Tai, 2012) #csforall #csedweek https://t.co/TiAUvpP3bQ
— Lana Peterson (@tyLanasaurusrex) December 6, 2018

I responded that I've seen many kids get their first exposure in high school and from that go on to careers in tech so it's clear to me that high school isn't too late. I was reminded that I was talking about kids who went through my required HS CS class so the kids had no choice but to be exposed to CS but the people advocating starting at an elementary level are talking about a compulsory experience so to compare a required elementary experience against an elective high school one is really apples to oranges.

Even thought it's an unfair comparison there are a lot of things it can lead one to consider.

Should we teach CS in every grade? Probably not. How about every band? Maybe. For those of you not living in the K12 world "by band" usually refers to a set of grades - kindergarten to 2, 3 through 5, 6-8, and then high school. Where should it be required and where an elective?

Teaching a required CS class in every grade or band is going to be a huge undertaking. I'm also not sure it's required. To get the best bang for your buck you want to expose the kids to CS in a way and at a time when they might pursue it further. If you're having only one required experience, I don't think the elementary grades are the place.

There's that point in time, I think around first through third grade where dinosaurs are all the rage. Maybe more so in NYC where we have access to the American Museum of Natural History but still. Lots of young kids want to be paleontologists but it doesn't hold. Okay. maybe I'm being a little silly here but I think the basic point is accurate. Kids are exposed to many things in the early grades that just fade away.

I'm not saying that we shouldn't do CS or CS related things in grade school. Things like the activity Fred Wilson described in his post on this year's Hour of Code or the problem solving and logic games good teachers do if given the time and resources. Things that have always been done by some teachers in some schools but are now being relabled CS or Computational Thinking. These are great practices and should be made available to all students as part of their elementary education but I doubt that they'll lead to many kids following CS as a career path. I'm not an elementary school expert but to me, even though I see great value in the early logic / problem solving stuff, I don't see hyping CS early as making a difference unless there's some follow up before the interest fades.

That brings us to the middle school bands. I've dabbled a little here but it's also not my wheelhouse. In my experience, kids in lower middle school can dabble in DnD tools and start to develop some coding skills and explore some of the CS concepts that they'll see in late HS or college and those in upper middle school can play with the text based languages to do the same. What I also noticed is that what can be accomplished in a month with those kids in middle school can be done in a week or two when the kids are a year or two older and with greater understanding and retention.

As a side note, before people get up in arms with "I teach middle school kids APCS-A and they all master it" let me say that I'm describing my experience across a range of middle school kids. There will always be few outliers who can do more when younger. That said I'm betting that most people who say that any middle school kid can indeed master APCS-A hasn't done so in an inner city poor public school. Have you done it while teaching a class of 34? With a class of low income students with all the disadvantages and challenges inherent to that population? What's more did you then have to teach them a data structures course a year or two later to see if they really mastered the original material.

This brings us to high school. As with my original tweet, in my experience if you give kids a good first CS experience in high school you can indeed give them everything needed for CS for All and it's close enough to college so that they might very well continue on the CS career path.

So where does this leave us? We certainly should embed computational thinking practices (whatever that actaully means) in the elementary school, particularly if we can do it in a way that also supports other academic areas and while it's nice to have a CS middle school experience and if we can afford to offer one that's wonderful. On the other hand, I don't think it's as important as elementary exposure and high school exposure so if we have to cut a corner, that's the place. To me, and of course I'm biased, the critical time for a requirement is high school. Sure, if it's an elective kids won't take it but that's true of a lot of subjects. I don't know what the research says but people do indeed change directions. How many of us are doing what we thought we'd be doing in grade school or middle school. Heck, even high school. I have a friend that did computer work in his twenties and thirties, then went to law school and practiced law for a bunch of years and then in his forties started a production company in LA.

I'm a big believer in CS for All but to me it doesn't mean it has to be in every grade or every band. If I could only do it in one place it would be required in the high school. If in two I'd add the elementary school piece and then finally the middle school.

As a final point I'd like to add that school is a zero sum game. There are only so many minutes in the day and as we add CS we remove something else. I believe we can add CS or, more specifically CT to the elemntary schools without cutting anything and I know there's room in high schools in NY. I'm not sure about middle schools. In any event, I'd hate to see music and art programs - already decimated by budget cuts and misguided policies like the small schools movement take a further hit so as to get CS everywhere.

Just some food for thought.