Three Dollar Quill

Okay, let's see if I can figure out what I "should" be working on here.

• I want to design an Old-Norse-analogue
• Which is going to be descended from a Proto-Indo-European-analogue
• Which could plausibly have other descendants, if I go that direction
• I'm handling the combination of "common elements and specific elements" by creating a single source repository, and cloning from it

With all of that in mind, it seems to me that I could get sensible behavior by creating a generic "descendant of this protolanguage" repository, and then cloning a repository for Cosmopolitan from that. That deals with some concerns I had. Next up...

• The source lexicon will be generally stable, and each root in it will be transformed by a set of rules that is consistent within a single repository
• It seems to me that I should like to verify the form of a root at specific checkpoints, and handle stuff like sometimes synthesizing roots together somehow; somehow

That isn't resolving out as obviously as the previous list. Let's let that sit, and consider the issues I'm dealing with currently:

• It's really obnoxious to construct phoneme tables in HTML
• I'm not sure what's a good way to handle sound changes

I've got some ideas partially sketched out for the first one, so let's focus on the second one (assuming that the whole intermediate descendant repo idea works out). There are command-line tools for applying changes, and Python code for applying changes. I know I've got some kind of attempt lying around...

Oof, that code sure is code.

At this point, I should see how shelling out to Lexurgy works. Just stop thinking about trying to find, update, or port code. See how it all works, and then only mess around with other stuff if there's some kind of specific problem I can't see how to address.

It's late, so I'll see how this comes together in a few days. For now, I really need to get ready for bed.

Good night.