Unexpected Help With Coding Projects

Fair warning, this is a post in two parts. First a project idea and second musings on the tools I used to create it.

I really do like to write code for fun. Nothing complicated (been there, done that, got the T-shirt - literally) but just little things to "scratch an itch" as they say.

Lately as I played Wordle I was wondering which letters appeared most in each place in the five letter words in my word list. A couple of nights ago, I wrote some code to find out. I had my code output a comma delimited file so I could use Excel to look at the results. That’s what the image to the side shows.

Now this sort of thing is highly dependent on the word list of course. But for my list, S is the most common letter in the first and fifth location. Not surprising as S is used to make plurals. Wordle doesn’t use plurals so I note that the second most common fifth letter is E with Y a close second.

The letter A is the most common second and third letter. The letter E is the most common fourth letter.

If I were ambitious, I could probably use this information to make a smarter Wordle solver. I’m not quite that ambitious though. I am toying with gathering some other statistics though.

I develop using Visual Studio – the full blown version. That means that Copilot jumps in to help. That’s not something I anticipated when I started but I confess that I found it surprisingly helpful. I did specifically ask Copilot to write one specific method – generate a string array of two character combinations – but it jumped in on its own with a couple of small bits of code. I was surprised at how well it anticipated what I wanted.

The implications for teaching programming are something to think about. On one hand it’s scary that AI tools can so easily write coding solutions to simple programming assignments. That turns our process of evaluating learning on its head a bit. At the same time, I am not ready to blindly trust AI generated code. I do not want students to blindly trust it either. So asking students to test generated code seems like a reasonable thing to assign. Yes, I suppose some students will ask AI to generate test cases but if we can’t trust AI to write the code in the first place we can’t trust the generated test cases.

We could ask students to explain the test and related tests. Could be quite a recursive rat hole.

We can also ask students to explain the generated code. We should probably ask them to do that either verbally or by writing manual in class so they can’t ask AI to do it for them.

What I keep coming back to in my own thinking is a focus on abstraction and top down design. Can we ask students to break the problem down to component parts and have them prompt the AI to implement various methods and code pieces. A focus on design rather than writing code. We could have students submit the design document and the various prompts that they use. Add to that some serious examination of testing and verification.

Students are going to have to work with artificial intelligence. They can’t let it do all the work for them because AI is not I enough yet. I don’t think it ever will be either.

Coding For Fun and Mental Exercise

Recently, I started reading a book about the mathematics of cryptography. (Mathematics of Secrets) Fascinating. A lot of the math is hard (to put it lightly) for me but the stories that go along with the development of it all are fascinating.

So far I have learned a few things and made my Caesar Cipher program much more efficient. I really need to revisit my Vigenère cipher program and make it more efficient and more interesting. The other night I coded up a quick transposition cipher. It was fun and once I got the algorithm down in code it was easier to get to sleep. Writing the code does help me understand what I am reading and that is a real benefit.

We’ll see how I am feeling when I get to the chapter on public key encryption. So far, I am sticking with the easy math and playing with ciphers that are far from modern cryptography. I'll leave that to the professionals.

Caesar and Vigenere are common enough programming assignments but I may write up the transposition cipher as a project for future use. If not for my own classroom for a project book I have in mind. Miles Berry pointed out that teaching ciphers by having students write a little code and experiment with different variations is much more fun and engaging for students than exercises away from the computer. Doing this stuff by hand can be a bit tedious.

For now though I am finding some pleasure is writing some not very complicated code as a way of exploring ideas that I am learning. I’ve had a chance to play with some libraries and methods that I haven’t really gotten to use before. That’s been fun. So much of my coding the last couple of years as been limited to the stuff I teach in a first programming course. I’m using this time to stretch myself a bit. Perhaps get my coding “muscles” back in shape.

In any case, for me,  coding == fun

AI-assisted Programming

As I said in yesterday’s blog post,Microsoft has been running their big MSIgnite event this week and making all sorts of announcements. One interesting announcement what improvements to what they call Visual Studio IntelliCode.  The idea is that they are using machine learning to help programmers with writing code. In the latest update are whole line completions and refactoring. Interesting.

I haven’t seen IntelliCode in use before and it is probably because it is not turned on by default. I did some searching and found that it can be turned on as an Extension (I had to search a bit under the Manage Extentions option) and then you can customize aspects of it from Tools –> Options –> IntelliCode. Oh, and you need Visual Studio or VS Code 2019 for this. I am using Visual Studio 2019 Community Edition which is free.

SO far I am not seeing obvious help from it but it may be that I have to do something more complicated in my code or write more code first. It does apparently take context into account for its suggestions and I may not have given it enough yet. But I’ll keep playing. And maybe I should look for some video demos online.

There are other things to think about though regardless of if it has been useful already. What is the training like? The write up I have seen says that it learns from examples on GitHub. Initially that scared me as literally anyone can upload to GitHub. One reference I found said  something about the 500 best repositories on GitHub. What ever that means. And what was the criteria for “500 best?”  What sort of standards will this learning suggest?

Some of the documentation talks about options for using other databases of code to train the system so that businesses can train the AI with their code and standards. As a teacher I wonder if there could be a training set for AP Computer Science A? I think Java is supported with IntelliCode in Visual Studio Code.

No doubt some will complain that it is “too much help” for beginners but I’m not sure about that. You still have to know enough to take or reject suggestions for one thing. And having the computer tell a beginner something like “you should make a method to reduce this redundant code” might be better received than the same comment from a human teacher.

Will this help students learn more or dumb down the process? I’m an optimist and think it will be a good thing. What are your thoughts and concerns?