Thread by @ShriramKMurthi

This is an interesting thread by Jon. Having been (sort of) on both sides of this, I have some thoughts. Jon is super right that making changes to teaching languages is really problematic. Books get printed, materials consolidate, updating is not easy. ↵

https://twitter.com/jonmsterling/status/1460872586318032902

I've extensively supported both secondary school and university educators. It's tricky at all levels. You're a prof, you have your lecture, it's all set to go, the evening before class you re-run the code to make sure it's all good, and … the language has changed on you. WTF. ↵

In @racketlang, of course, we have #lang, a general, universal mechanism that can be used for versioning and soooo much more. Curiously, it seems to be used for soooo much more but NOT for versioning. Maybe it's too powerful. Some of it is also cultural. ↵

In @PyretLang we've long thought about what a #lang-like thing would be in our world. We aren't ready for (nor think we need) full #lang. The #lang-inspired Grace dialects have been a promising approach (@jameskjx). In general, there are two kinds of issues. ↵

One is you want to configure libraries (i.e., the default namespace). The other is you want to configure the language itself (e.g., remove :=). The latter is especially hard around parsing, documentation, etc. But we now have a solution for the former. ↵

When you fire up the Pyret IDE now, you see "use context essentials 2021". It's grey and not editable. This configures your namespace. You can change your context from the menu. (Thanks @asolove.) Legacy code is essentials2020. ↵

Someday, the context may provide the ability to configure the language constructs too; we're not ruling that out (and indeed want it eventually). But for now, there's a nice, clean, stable way for educators to not suffer night-before-lecture panic. All ed languages need it.