The State of the Software: Glitch Frontend

[marrus-sh]

It’s bad.

I’ve been working through the glitch flavour in the interests of #21 and trying to knock out a piece from #5 as well, and it is a mess of pasta 🍝.

Largely this is my own fault for ⓐ hard‐forking the glitch frontend from the vanilla one, ⓑ getting halfway through some pretty big refactors of how things worked, and then ⓒ peacing out forever ✌️✌️. Since then Mastodon has made its coding style more strict / consistent, while React has updated in some pretty major ways. Claire has done an admirable job of keeping the thing working, but as far as getting it to a “nice” place goes, I think it’s going to take a lot of work.

Considering this, my recommendations are to prioritize one or both of the following things :—

1. Port over all the changes from glitch back into vanilla, prioritizing maintainability and contemporary React best‐practices. Then, trash glitch lol. This is effectively the same result as Remove the vanilla flavour #22, but should be easier to maintain.

2. Develop a new Kibicat frontend which does not suffer from these problems. Discuss this in develop a GlitchCat flavour and make it the default one #16. I would estimate this as taking a month or two of dedicated work for an initial product, and maybe twice that for full feature support. Of course, depending on other priorities, it will take longer (and if we can bring in more people to work on it, it could be done faster).

In the meantime, we can maintain glitch as what it is. I would consider this blocking for significant new frontend work, however.