(1/2)
@evan
> fun work is on extensions and applications, not on word-smithing the description of the `proxyUrl` property
Sure. But for a WG to change AP federation in practice it needs buy-in from implementers, yes? Otherwise all it can do is change official specs, which will continue to be treated as suggestions not standards.
I'd like to help with getting more AP devs excited about the spec update, and giving feedback. Whether that's by actually joining the WG or otherwise.
(1/3)
@laurenshof
> W3C membership structure concentrates formal power ... while the implementers who could counterbalance that power have largely opted out of the process
My impression is that like a lot of things, this is caused by a combo of;
* lack of time and energy among projects devs, most of whom still work as volunteers
* the tense atmosphere that persists in many dev collaboration spaces, like SocialHub (not pointing the bone, I'm no saint, and I recognise I've contributed to this)
(2/3)
My hope is that the chartering of a formal W3C Working Group to update ActivityPub will motivate devs to participate more in standards work. The promise of making concrete changes to the protocol specs might make it feel like less of a Boulder of Sisyphus. Plus, the prospect of working in a formally facilitated process, with fewer rubberneckers (like me), might mitigate any anxiety about discussions getting unpleasant.
(3/3)
I wonder if there's a way to get grants to cover membership fees for fediverse projects to join the W3C? Especially those with formal legal structures. I wonder if CodeBerg, for example, could become a W3C member and nominate the devs of fediverse software hosted on CB to represent them on the AP Working Group?
Maybe the SWF could use some of the money they got from Meta etc to pay W3C member fees for other fediverse projects?
@strypey anyone who works on ActivityPub and wants to be part of the Working Group can become an Invited Expert. The chair, @darius , is an invited expert. I was an invited expert when I chaired the first wg. All of the authors of ActivityPub and half the authors of Activity Streams were invited experts.
@evan
> anyone who works on ActivityPub and wants to be part of the Working Group can become an Invited Expert
Fantastic. What's the process here then, are people working on AP implementations being actively contacted and invited? Or do they need to contact someone to get an invitation, and if so, who and how?
I ask because I want to be clear on what the situation is, and I'm wondering if there's any way I can help with making sure AP devs know what the process is : )
@strypey @darius here's the big question: do you really *want* to work on an iterative revision of ActivityPub core? Most of the fun work is on extensions and applications, not on word-smithing the description of the `proxyUrl` property.
I'd advise people to make sure they come to joint WG/CG meetings and definitely come to the weekly AP issue triage meetings. If you find yourself energized, ask about becoming a WG member.
#LOLA is NOT »ready for developers to start testing implementations«.
The main problem to solve with moving post content is NOT how to do it technically, but to find consensus on some of the "Trust decisions" (3.2) to be made:
Bulk import (3.2.1), aka "We trust your source server", or "We just click 'OK'"?
"Copied content may undergo moderation" (3.2.2), aka: "Yeah, sure, we'll check all of your 2,871 old posts and will tell you which ones we'd accept on our server" (»We don’t make any recommendations for implementations how to do this, as many options are available.«, lol)
Saying LOLA was »ready for developers to start testing implementations« ignores, as usual, the real pending nontechnical questions and pretends there were already technical "solutions" to them.
»The main bottleneck since then has been getting organisations like Mastodon interested in actually building it.« That's not a bottleneck, but a more than obvious indicator of the nontechnical questions not being resolved, and of being reluctant to invest time into the technical side, just because some techie is chomping at the bit.
https://swicg.github.io/activitypub-data-portability/lola#trust-decisions
@katzenberger
> Saying LOLA was »ready for developers to start testing implementations« ignores, as usual, the real pending nontechnical questions
Protocol development is a place to define technical approaches, not adjudicate social questions. So if the remaining questions are non-technical, then the proposed protocol is ready for testing. It's during this testing process that solutions to social questions can be explored and tested, to improve the UX of the final implementation.
Protocol development is the place for respecting the results of an upfront resolution of these matters.
I can't help you with your ignorance, grow up and learn to accept responsibility, or stay out of my mentions.
@katzenberger
> I can't help you with your ignorance, grow up and learn to accept responsibility, or stay out of my mentions
Physician heal thyself. Muted for a week.
