Caterpillar

Once Caterpillar is capable enough to support writing a simple game that runs in the browser, that's arguably a useful result. But it won't mean that I'm done thinking about risks. They will just shift.

A simple game is nice, but will Caterpillar's performance be good enough to run a less simple game at 60 frames per second? Will the previously implemented approach to interactive programming still work under those constraints? Can I improve the language to the point that it has any advantage over more established ones?

All those risks will have to be addressed, one by one. I think it will take quite some time before I can say with confidence, that Caterpillar works.