Wow, thanks for the great in-depth feedback! ๐คฉ
Yeah, there's definitely other areas I could've delved into, like Public Clips or MFM or pagebuilding. These in-depth reviews are challenging, due to trying to strike a balance between features and actually getting something published. Most articles of this nature takes me weeks, sometimes even a month or more.
The UI definitely has a learning curve, too, but as a veteran Fedi user, it suited me just fine. I've dealt with far, far worse ๐
The instability really bums me out. I'd like to think that things are slowly improving, but the lack of transparency (and frankly useless error messages) make it really hard to triage where the problem is and forge a path forward. The lead dev has also been sick recently, and suddenly is not very active online.
Finally, I think Firefish takes part in a long tradition of Misskey forks, where a half dozen systems all branch off of each other. It's a shame that more of them don't collaborate on the same platform, leaving many devs to cherry-pick across forks. I wonder sometimes whether this hurts development more than it helps.
Yeah, I get that it's not that new, especially since it's a rebranded fork. But, as a fediverse project with its own brand and design sensibilities, it's relatively new, especially compared to Mastodon.
Unfortunately, I've used the Antennas for like two months, and they're janky. The "non-retroactive data fill" point that you make is only partially true, and seems to mostly apply to filters pertaining to a collection of users, or a collection of servers as a data source. It's a confusing UX papercut. Worse yet, my Fediverse Devs antenna example has been around for two months, and barely produces anything most of the time.
I don't believe that an "official branded app" is strictly necessary. What I meant was more in line with "Firefish could really use apps developed for its features specifically." Sadly, Misskey compatible apps continues to be a wasteland compared to the plethora of Mastodon ones.
The whole "the flagship instance is a sandbox test instance" is kind of a sometimes-true sometimes-not situation. It's definitely less stable now than it was two months ago, but that kind of messaging and expectations management didn't seem to really happen until sometime after the CalcKey migration.
And yeah, proper group / community federation with Lemmy is a huge deal! I'm looking forward to what Pixelfed is doing with Groups, as they look somewhat similar, and aim to maximize compatibility while providing good management tools.