"there is a recurring refrain about implementers deciding they don't care to implement AP as specified, and that this indicates a problem with the spec, not a problem with implementers."
-
"there is a recurring refrain about implementers deciding they don't care to implement AP as specified, and that this indicates a problem with the spec, not a problem with implementers."
It does. The point of a technical standard is to document the best practice of a group of implementers. It is, by nature, a living document that needs to be updated, as the experience of implementing reveals problems and finds better approaches.
-
strypey@mastodon.nzoss.nzreplied to strypey@mastodon.nzoss.nz last edited by
"when ActivityPub was being standardized alongside AS2 it basically had two compelling reasons for what would become the fediverse to adopt it"
The people standardising AP were already active in fediverse implementations. The network existed under the name "the fediverse" long before that. Originally using OStatus, with some apps also supporting Zot, or a reverse-engineered version of Diaspora's internal protocol (forked from OStatus).
-
trwnh@mastodon.socialreplied to strypey@mastodon.nzoss.nz last edited by
@strypey i think i don’t explicitly state this in the thread but i don’t see it that way; it is not a problem with either spec or implementers, but rather a mismatch between what the spec tries to do vs. what implementers are trying to do. i also disagree that standards necessarily document best practices of implementers — part of it is“which came first, the implementation or the standard?” and the other part is again that mismatch. (in a later draft this’d be a footnote)
-
trwnh@mastodon.socialreplied to strypey@mastodon.nzoss.nz last edited by
@strypey this one was poor wording, what i meant to say was “what would become the [modern] fediverse”. i am aware of the history of the term, i just accidentally a word
-
strypey@mastodon.nzoss.nzreplied to trwnh@mastodon.social last edited by
@trwnh
>“which came first, the implementation or the standard?”The implementers. Otherwise it's not a standard, it's an academic paper on a potential approach. In the case of the fediverse, the people the standard was drafted for were all in the room in some way, shape or form. Or at least that was the goal.
-
strypey@mastodon.nzoss.nzreplied to trwnh@mastodon.social last edited by
@trwnh
> what i meant to say was “what would become the [modern] fediverse”OK but that still carries a subtly but significantly different meaning from saying;
"AP basically had two compelling reasons for the OStatus fediverse to adopt it"
OR
"basically had two compelling reasons for the OStatus fediverse to join the AP standardisation process and adopt it"
I realise I'm labouring the point, but with people floating around rewriting the history to push their own agendas, I think it matters.
-
trwnh@mastodon.socialreplied to strypey@mastodon.nzoss.nz last edited by
@strypey noted, i’ll incorporate your feedback into the edited work
-
strypey@mastodon.nzoss.nzreplied to strypey@mastodon.nzoss.nz last edited by
Having said that, I agree there's...
@trwnh
> a mismatch between what the spec tries to do vs. what implementers are trying to doBut the paragraph quoted implies the solution is to change how implementers relate to the spec. When it seems self-evident to me that the solution is to update the spec to meet the needs of implementers.
Particularly in ways that make it more useful for unfunded tinkerers like @andrew from Takahē or @rimu from PieFed @piefedadmin. Rather than for big companies.
-
trwnh@mastodon.socialreplied to strypey@mastodon.nzoss.nz last edited by
@strypey @andrew @rimu @piefedadmin I obviously haven't gotten to writing/publishing the "part 3" of this but the solution is neither for implementers to change or for the spec to change. I think that it's important to recognize that the needs and intentions diverge, and that the worldviews can't really be reconciled. It's a question of whether you fundamentally believe in the paradigm of "social networking" or "social media", or if you want to transcend that with a paradigm shift.
-
trwnh@mastodon.socialreplied to trwnh@mastodon.social last edited by
@strypey @andrew @rimu @piefedadmin Which is to say: if you want to build a "social network" then you need a "social networking protocol", and ActivityPub is not that. But what "that" is, or "that" looks like, is something that is the subject of that unwritten and unpublished "part 3" in the series.
(I'm not sure when I'll get to it, because I have other priorities than writing thinkpieces... I also need to get my website updated before I feel comfy pushing the final version of that article!)