Thank you, yes I agree with that. I was thinking the model at least needed some standard relationship types to have well-defined meaning and behavior when encountered. Other relationship types are then app or domain-specific and can be ignored when encountered.
Other than that I wonder if permissions and permission discovery mechanism - and some other common dimensions - should be another AP extension altogether. Separately defined and not tied to the Groups extension. So that all these extensions become like ‘micro building blocks’ that can be combined together or build on top of one another, to add some capability. For instance, for my own purposes I’d like to build on Groups and Permissions to enable Governance models.
Groups vs. Communities?
Something else I wanted to highlight. I find our overall approach to adding ‘Groups support’ to be overly technical. We look at adding as:Group in our individual apps in a standards-compliant way, rather than imagining how universal group support more broadly enriches the fediverse. This while we all recognize that empowering people and their social interactions is key. We strive to model real-world concepts and make them work online. And be innovative with that: Social networking reimagined.
I see my draft diagram not as ‘Groups support’, but rather as (the MVP of) a Community concept. In more technical terms it is a sub-domain. While I did not name my Community AP extension topic as such, I imagine it enables a paradigm shift for the Fediverse which I call “Community has no Boundary”. Not just the community is important, but the space between communities that allows them to interact. This is where the Fediverse shines. Our community model should break barriers between app / instance silo’s, and be more versatile than just offering follow and join capabilities.
Looking with this mindset shouldn’t mean that adding groups support suddenly becomes much more work, and we get stuck in ‘analysis paralysis’. It would allow us to lay the proper foundation to the paradigm shift, an MVP if you will, and maybe a maturity model to gradually expand on that. And I think it would be greatly inspirational to adopt this broader perspective.
In this Community extension post I already mentioned the work of Aldo de Moor, focusing on relationship types in his community vocabulary. But I highly recommend reading the entire paper and imagine how the Fediverse could be supportive to realizing its goals.
And see also:
(PS. I tooted a link to this post in: Humane Tech Now: "I just wrote a plea on #SocialHub to think broade…" - Mastodon )