Notices by Dmitri | 🇺🇦 (dmitri@social.coop), page 2

@panos @kainoa @syuilo @Gargron Hi Panos, would it makes sense for the Calckey devs to do a writeup on how they're implemented Quote Posts, on somewhere like the FEP repo (https://codeberg.org/fediverse/fep ), so that we can point implementers at it?

@silverpill Would you be interested in attending the first SWICG call?

@silverpill Are you saying the FEP community would not want to work with the community group? Like, we can still orient around the FEP repo, and also provide additional infrastructure / calls, etc..

@silverpill Make sense, ok. (That someone was me, @rhiaro and @bengo, incidentally.)
Ok, so what can a W3C CG offer to the FEP community? Off the top of my head:
1) Regular voice calls for any sort of topics that require higher bandwidth communication.
2) Visibility to, and liaising with, the larger W3C community. Basically this means: easier to start a Working Group when time comes, and we represent the SocialWeb/FEP community to other CGs/WGs.

@silverpill 3) (and here, we get into the whole topic of - so why do anything in a standards group like W3C rather than outside of it) IP protection. Basically, W3C exists to protect standards from patent trolls. Which hasn't been an issue so far for FEP, but becomes more important as the Fediverse gains more mainstream visibility and adoption.

@silverpill Like, I'm not saying the FEP community NEEDS the W3C CG. But surely, more hands and eyes and voices on some of the Fediverse spec problems, might help?

@silverpill All good points.
> Paying instance admins means supporting bigger instances

::shrugs:: One, instance size doesn't really mean anything. And two - hey, if social.coop wants to provide hosting and moderation for me, why wouldn't I want to pay for their labor?

@silverpill
> Standards work is done here: https://codeberg.org/fediverse/fep, but you're creating a competing group

SWICG is not a competing group :) It's the continuation of the original WG that created ActivityPub, etc. Also, I'm a HUGE fan of the FEP community (and of your work there, too), and I hope to work with them closely.
There's room for both communities without conflict.

Nobody:

Me: Let me tell you all about my thoughts on the recent rivalry (read: fairly one-sided grumbling on the Fedi side) between Bluesky and the Fediverse.

What I think of the "conflict": Wholly misguided. But I'm glad it's happening, because it's galvanized a bunch of people and projects on the Fediverse side.

(Thread continues, summarized over at https://medium.com/@codenamedmitri/fediverse-vs-bluesky-62afcc2745eb )

1/

I have no doubt that a bridge between Bluesky and Fediverse will go up in very short order (already being worked on at @activitypubblueskybridge). Some instances will block it, most others won't, and people will continue posting and interacting as usual.

2/

What I think of the Bluesky data model: Siiiigh, ok, so, do I wish they at least started with the ActivityStreams2 data model for posts and tried to extend from there? Yes. Do I understand why they studied it and decided to not use it? After a bunch of discussion, yes :/ Regardless, I think we (the builders of the protocol bridges, and the larger Social Web) will be able to translate between the two data models fairly easily.

3/

What I think of the AT Protocol: Do I wish the team didn't re-invent JSON-RPC and JSON Schemas? Yeah, my heart always leans towards interop- and existing spec-first. Do I understand why they did it? Yeah, maybe, grudgingly. Change is bad, but also, they wanted to play around with some new options and technologies, and I highly suspect it will help improve the Social Web overall.

4/

What I think of the Bluesky architecture: Ok, so, this is where I'm kind of boggled by the current Fediverse discourse. That architecture? Is almost exactly what the Fediverse will end up with, as it scales. Third-party search indexing (opt-in, of course) is a general tool of a decentralized Web, there's nothing AT-specific about it. Same with all the other pieces.

5/

What I think of the source of Bluesky funding: Heeeere's the thing. If a billionaire wanted to give me a million $ or ten, and I saw it as an opportunity to further the cause of the Open Social Web, would I do the same thing Bluesky is doing? Yes, IN A HEARTBEAT. (So, if you know any extra $ burning people's pockets, hook me up.) Except I think I'd re-use existing tech more.

6/

Just whose team am I on? Uh... I'm on team Plurality, Queer Post-Scarcity Space Anarchism, and Libre and Liberatory Software. I'm against fascism, and _for_ trying to survive the current Multicrisis/Jackpot with the least loss of human lives and least species extinctions. (Except for deer ticks, those can seriously go to hell.) So in that regard, I think (until proof to the contrary) that the Bluesky team is on our side.

7/

But so, where am I going to be spending my time and effort? On the Fediverse side, of course.

8/

Btw, I'm not thrilled with the term "Fediverse". It's better than "Mastodon", of course, for obvious reasons. And better than "ActivityPub" (since that's just ONE protocol of the whole stack). But the thing is.. there's nothing sacred about federation. It's just one (albeit important) tool in the distributed system toolbox. To name our whole scene after it, is like naming us "Array-ists", or "Linked List-ists". It's just a low level detail.
Let's just call us what we are: Open Social Web.

9/

Anyways, what do I think of the current Fediverse: Well, it's my people. I love it and use it, and am hoping to help it in whatever ways possible (paying instance admins, reviving the W3C Social Web Incubation Community Group for standards work, building bridges). Do I think it needs some improvements? Oh, hells yes -- and I'm'a tell you all about those, shortly.

So, we have a lot of work to do. And remember, we're aiming for the Pluriverse. And lots of bridges.

10/10

@silverpill I don't suppose we can convince the spec authors to rename it to the 'Authentication' framework, rather than authorization? Cause DIDs are just for authn; for authz you should use a combination of Solid-style ACLs and capabilities like UCANs.

@cuchaz @grishka ok, but also, use DIDs (Decentralized Identifiers). Like, start with did:web (which is just a well-formatted JSON object that lives on a domain).