After reading through https://vox.ominous.net/@occult/117031746774508558 , it seems cat's out of the bag now and pixelfed may officially be considered unmaintained. Not that this has been any different the last couple of years, given for how long some rather nasty issues in even the very core feature set have been unaddressed and unfixed. I've been having certain issues with the project for quite a while now, but maybe to sum it up: 

• As a single developer, you aren't set to compete with a multi-million-dollar company like Meta in building a fully-fledged, easily usable application ready for the mass market.
• ... let alone building this application and running it reliably on a larger scale, including administration, content moderation, machine maintenance.
• ... let alone building it and running it on a larger scale in a federated heterogenous environment where you're supposed to talk to a wide range of other systems and incompatibility is not a nasty side-effect but actually a conscious design principle.
• ... let alone doing all that for multiple applications and platforms (pixelfed, loops, sup, fedidb) at the same time. 

All of this is far too much for a single individual to even handle from a responsibility point of view, not even talking about actual doing that work. Developers claiming otherwise probably should reconsider assessment of their own skills and abilities, and also should stop telling people out there otherwise for it's misleading and treacherous at best. Users, though, might be off better comparing software solutions or online services other "complex" pieces of technology, software, service they use on a daily basis and wonder whether it really could be designed and built and maintained altogether by a single person. We're a society of shared cooperative work with myriads of specialists joining forces to get things done. In every field. All the time. One person handling projects, tasks this heavy is just not going to work. On no level.

And maybe it shouldn't. Maybe, it hasn't to. Maybe the solution to that big tech problem isn't to mimick what they do best - starting and building large scale projects to fiercely compete with each other and everything else - but rather focusing on a really different approach. Like: Do things at a smaller scale. some.pics in omg.lol is a great example how this could be done. Or maybe subversive.pics. Keep complexity low. Keep bells and whistles out unless absolutely needed. Don't reinvent the wheel if there's prior art to build upon (like, in case of pixelfed, I'll never understand why this hasn't "just" been a great, image-centric frontend to whichever ActivityPub backend people might want to use). But also: Community over individualism. Focus on fostering a small crowd of like-minded souls to work together collaboratively on building and running stuff, rather than making it more or less a one-man-show with an individual programmer standing out. Get people enthusiastic about building and using this particular piece of technology. Get people to, also, think about sustainability, about making sure (from both a technical and a structural perspective) that thing can still be around at least years from now. Maybe this would fix a bunch of problems we currently see with some of the FLOSS projects out there. And maybe, too, it ... could help turning structures that still are essentially big-tech capitalist (a very few people "providing" something and a by far bigger crowd of people "using" something without really having a chance to contribute or add to it) to some sort of community-built, community-owned infrastructure. It might not be a bad thing after all.

Filed under: