Quick note on the Ink stuff: I realized that the behavior changes I want can't be introduced in the interpreter, and I should instead work on hacking the compiler when I feel like it. This is helpful, because it means that my goal with the interpreter port is entirely to get it functioning to spec. The most complicated thing I might end up doing is having separate branches for mutation testing coverage, and profiling and optimization. But none of that is relevant right now, since I still haven't implemented all of the opcodes.
Anyway, as far as conlanging goes, I've got all of the glosses put into tables now, which renders hideously, but I'll look into that later. I've added sections for all of the sections from my original outline, and started copying stuff over. My task now is to collate and categorize all of the disorganized thoughts I've had so far, and fit them into the outline.
In the process of inputing the glosses, I found a few more errors in the glosses I posted, but I kind of don't feel like correcting them in the post.
I don't have much more to say now, because this has mostly been just taking things I've already talked about somewhat, and putting them together so they can usefully interact.
Anyway, I am sleepy and should go to bed.