Three Dollar Quill

Hm.

I got all set up in Flet, and after a bit of struggle, managed to preview my app on my phone, but as I look at and tinker with the preview, I'm skeptical that the app preview is telling me what I really want to know. Like, if using the wrong version of a package causes import errors to pop up on my laptop instead of my phone, then that doesn't tell me anything about whether the Python packages I want to use can be cleanly packaged into an apk, right? I guess the point is to make sure that the UI isn't just totally unusable on my phone.

...

And from the radical step of reading the documentation, I see that compiled packages are a problem, they did anticipate it, there is a solution, it's not turnkey for what I have in mind, but it doesn't look like bashing together something for my own use is a fundamentally unreasonable proposition. I won't have this all ready for when I want it all ready, but it seems likely I can get something sort of workable soonish, at least as far as Flet goes.

So, that's some brief investigation, and I think I want to get off my laptop for tonight, so I'll focus on paper prototyping or reading.

...

I did things in the wrong order. So, that's some good barebones prototyping done, and a bit of research on Flet. The one thing I'm concerned about is the whole "Oh, we layered imperative on top of Flutter's declarative, so you don't need to worry your silly little head about defining the UI as a function of the state" thing. Like, I look at that, and I've got the twin dæmons of Spite and Pragmatism shaking hands and smirking over the concept of writing a function to regenerate the UI from scratch every time the state of the underlying data changes. Like, I'm going to have enough trouble getting the drag-and-drop behavior (which I kind of need) to propagate down to the data, I don't want to be manually optimizing the updates to the representation of that data.

Anyway, really should rest my eyes now.

Good night.