Crosscut

I'd like to conclude this long series on the new prototypes, by saying that I'm fully aware that I'm setting myself up for some of the same problems that prototype 11 had. Then, I hadn't built a code database, despite considering that a central feature. This led to a situation where all design decisions I made where either suboptimal or preliminary.

I'm going to run into similar problems again, for example with the type system. My plan is to have a static and linear type system, and yet I've decided to go with a dynamic one for now. I'm sure this will cause the same kinds of problems that I had with the code database previously.

But it's going to cause those problems in an area that's less central to my vision (although still very important). And I don't think that can be avoided. I can't make everything perfect from day one, so I have to draw the line somewhere. But I'll have to prioritize this soon. Even before other work, that might have a more positive impact in the short term.