Crosscut

Okay, so I went over the problems of the old prototype, and explained how I intend to address most of them. What I left out, was the syntax and evaluation model, because just as I started to write about that, my plans changed. I tried to go with a tried and true approach but then... couldn't.

I think that this is, at the same time, a big weakness and a big strength of mine. I just don't know how to do the boring thing, if I think there may be a better way. In this case, I saw how the tried and true approach wasn't great for Crosscut. Meanwhile, I had come up with an alternative model. So now I'm working on that.

Maybe I'll look back at this in a few months, and call it over-ambition. It wouldn't be the first time. But the fact is, my new model looks promising. There are no open questions at this point. It's not clearly a bad idea. So I'll try. And if it doesn't work, I'll try something else.