Thank you for your feedback. I'm glad you think it's stupid.

I make some exceptions, and Canvas may be one of them in the future. It's not a hard limit, more like a guideline.

I want to make stuff that lasts at least as long, and that means I should use tech that's already been around that long, to take advantage of the Lindy Effect.

But can't that backfire? Python 2 just meets your criteria, but you shouldn't use it, vs Python 3 which is only 12 years old.

In what way is that a backfire? Python 2 was retired after 14 years. It began to show signs of impending retirement after about 9 years, in the form of fork of "new version", parallel to diminishing support of 2.x.

I saved myself a bunch of time on a rewrite.