As a Fediverse advocate who is also a (technical) product owner I notice that many AP projects:
Deep-dive into technical stuff very fast, and stick there, often with less attention to product development.
Speak about AP support as a way of “being part of the Fediverse” rather than consider federation support more broadly.
Think of federation features too much as alignment to the Microblogging business domain, that dominates the Fediverse.
These three things limit creative thinking and make it harder to convince other project teams to add ActivityPub support to their product. To raise awareness of this, I created some SocialHub topics on how to change our advocacy for ActivityPub:
And on Discourse Meta forum a brainstorm on what federation support can bring to forums:
Fedeproxy: A case study
Like Forgefed the Fedeproxy project is a great example of an entirely new business domain, where AP federation support opens up a whole range of interesting use cases.
I would like to invite and encourage you to contribute to Fediverse Futures discussions, by posting in the same-named category on SocialHub forum, to inspire AP devs to think more out-of-the-box and beyond microblogging-like social features. It may help bring attention and contributors to the project too.
I’m very grateful that you reached out and your insightful comments here are most inspiring. I was on the verge of writing some code (because I can’t think for too long without writing code) knowing that it would be mostly wrong and to be thrown away. I now realize that reading the Fediverse Futures topic will bring me clarity. And save valuable time as well. Thanks
I had encountered it in the past, but forgotten about it. Thanks for the pointer, this is a great resource and I am happy it is in my ‘attention bubble’ again
Yes, I know about it. The SemApps ActivityPub + Solid project has an implementation. I am hesitant towards using the standard, but following from afar. I feel the Solid project has followed a go-it-alone approach so far, and there are other partially overlapping or competing standards in the works. The ActivityPub community spent much time trying to reach out to the Solid project for coordination, each time without any response from the core team (until very recently with a bit of acknowledgment).
Anyway, all interesting stuff to follow, but not yet matured enough to make choices for implementations, imho. Important that Christopher has some serious reservations to Solid’s approach. For the future of the Fediverse I am most eagerly following the Spritely Project and Object Capabilities (though they are for authz).
Other than Identity I think Solid still has a challenge in positioning itself, something I discussed in a number of topics on their forum:
I am trying to set up an initiative where people think more broadly and from the start of a project about all the advantages that federation might bring. Currently most Fediverse apps are created like this: Create app --> Add federation between app installs --> Think about supporting Microblogging-like federation (thought of as integrating with / “being part of the Fediverse”).
This is shortselling the capabilities of ActivityPub, which based on Linked Data / JSON-LD, is able to federate any business domain (based on closed AP vocabularies of said domain… we’re not talking full Semantic Web and its complexity). The initiative is called Fediverse Futures and the objective is to bring people together with a variety of backgrounds, not just devs.
Very interesting! You may want to have a look at TerminusDB. I created the delightful project on Codeberg (which is similar to awesome project on GH, but only for FOSS, Open Science and Open Data) and have a curated list there for Linked Data resources. You may want to create an issue for inclusion.