@tasket@infosec.exchange an official protocol handler would help a lot. Today there is the option of introducing a web protocol handler but the UX for it is pretty dogshit (Piefed recently implemented it, and the number of dialogues was too damn high!)

That said I don't know if PWAs can register against non-web protocol handlers. That would be useful for sites like NodeBB.

@julian There are several dozens of actively maintained ActivityPub implementations, I think it is not difficult to find two implementers among them, especially if they will be paid to implement a proposed change / extension (as we have seen with the E2EE proposal).

@slyborg @evan @connected-places @fediversereport @ArneBab @alexchapman

@julian IMO there's no reason why a web browser should understand where to open fedi links, without having any other type of app properly address those links as well.

What if someone in an instant messenger or email app sends you a link to fedi content?

Defining it at the system level (again, as is done with email) removes critical uncertainties.

Fedi has other big UX issues as well. Celebrities don't like it here because the TL mechanics make them unintentionally annoying... users follow then later mute them because their posts are popular for a while and we have to see them each and every time they're boosted (or manually silence those posts). Allowing the selection of some transparent algorithms could fix this.

@silverpill In a hilarious twist of fate, this gives (streams) and Forte an unfair advantage. They're nearly identical, they have the same maintainer, but they're two separate implementations, also seeing as Forte uses ActivityPub for nomadic identity, and (streams) doesn't and still uses its own Nomad protocol for it.

Since Mitra appears to implement (streams)/Forte features one by one and cast them into FEPs, that's three implementations already. Two if nomadic identity via ActivityPub is involved. And if Hubzilla happens to have it, too, we've got up to four implementations.

Yes, ActivityPub is only an optional add-on on Hubzilla and (streams), but an implementation is an implementation. And whatever they do on Nomad that federates has to get out through ActivityPub one way or another.

It'd be even more hilariously skewed, hadn't Mike discontinued the five apps between Hubzilla and (streams) on New Year's Eve 2022.

CC: @slyborg @Evan Prodromou @Connected Places @ArneBab @Alex Chapman

#Long #LongPost #CWLong #CWLongPost #FediMeta #FediverseMeta #CWFediMeta #CWFediverseMeta #ActivityPub #Hubzilla #Streams #(streams) #Forte #Mitra

@julian @silverpill @slyborg I will fight pretty hard against breaking changes in ActivityPub. We have an active network with tens of millions of people and tens of thousands of servers. It's too late for breaking changes and it has been for a really long time. Expect changes like: describing required properties of activities better. How `replies` (and maybe `context`) work. References to OAuth, Webfinger and HTTP Signature.

@julian @silverpill @slyborg it's also worth noting that all discussions of the WG will be on a public mailing list. People can join the meetings, comment on drafts on GitHub. People interested in making more substantive contributions can become invited experts, even if they're not from a member organization.

@julian @silverpill @slyborg most importantly: no protocol is mandatory. No protocol revision is mandatory. If the work the WG does isn't useful, nobody has to implement it.

@julian @silverpill @slyborg the issues I have marked for the next version are here.

https://github.com/w3c/activitypub/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22Next%20version%22

I know there are some on there that Silverpill won't like, such as supporting IRIs for object IDs. I think it's worth having that conversation.

@julian @silverpill @slyborg I wonder, though: what would be some changes that would worry you? I'm having a hard time imagining what they would be.

The best I can come up with are features that are too complex for small development teams (e.g. oodles of mandatory APIs), or too resource intensive for small instances to support (e.g. required to handle terabytes of big data).

@julian @silverpill @slyborg

The only other thing I can think of are forced anti-features, like mandatory advertising, mandatory algorithmic feeds, or forced participation in LLM training.

Are there other things I'm missing?

@julian @silverpill for you two especially, I wonder if you think there could be Trojans inserted into the ActivityPub 1.1 spec -- something that seems innocuous on the surface, but would actually EEE the Fediverse? I just don't think the standard is complex enough that anyone could hide anti-features in it that you and I couldn't find. Maybe, I dunno.

@julian @silverpill I think a heaping dose of skepticism is healthy for standards efforts. I'm glad to know you're keeping your eyes open.

@evan @julian @silverpill @slyborg What about "breaking" bug fixes in the spec? Many parts of spec are used by ~0 people on ~0 servers so the impact is only positive to do those fixes. Required properties is an interesting topic. Adding a required property beyond `id` (conditionally), `type`, and `input`/`outbox` for actor types *would* be breaking and potentially have a large negative impact (unless they are only associated with optional new features).

@slyborg I think this is a great point.

@evan @julian @slyborg

The best I can come up with are features that are too complex for small development teams

This is probably the biggest risk.

Another risk is changes that prevent the development of important features in the future.

@jupiter_rowland The two-implementation requirement sounds totally inadequate to me. Does it really work like that?

I think nothing new should ever be added to the core spec unless it is supported by 51% of implementers.

@fediversereport @alexchapman @ArneBab @evan @slyborg

@silverpill 'Implemented' is not boolean, some FEPs have partial implementations, implementations of a prior draft or implementations that do the opposite of a MUST

@eyeinthesky One of such "bug fixes" has already been proposed:

In section 4.1 "Actor objects", the definition of "inbox" uses the imprecise term "reference" and is different from the definition of "outbox", giving the false impression that the range of the "inbox" property is different than that of "outbox". A possible correction is to make the definition of inbox parallel with that of outbox: "An OrderedCollection comprised of all the messages received by the actor; see 5.2 Inbox."

-- https://www.w3.org/wiki/ActivityPub_errata

Previously, inbox was a reference, but now an embedded inbox collection will be considered valid:

{
  "type": "Person",
  "inbox": {
    "type": "OrderedCollection",
    "items": []
  }

It's a breaking change. I don't actually mind the change itself, but the way it was made. When I pointed out that this change affects existing implementations and asked to amend the erratum, other participants literally started to gaslight me with "there is no change" and completely ignored my objections.

@evan @julian @slyborg

@dimkr I mean implementers of ActivityPub specification. I think if a feature doesn't rise to the level of "most projects should support it", it shouldn't be included in the specification.

FEPs are a different story, In many cases 2 independent implementations of a FEP is enough.

@silverpill I agree, huge parts of the spec don't have enough interoperable implementations to justify them. However, this rule also means it's super hard to 'upstream' a FEP into the spec because it's hard to count implementations that do exactly what the FEP says.