But what is the actual PROBLEM with Mastodon?? Said one hundred dozen people over several months of threads here. So I wrote it up.

It's too long, but we got bad doors, stilt-walking French shepherds, water bugs, a bunch of @/shengokai@zirk.us quotations, What The Fuck Is Up With Bluesky, and more.

https://erinkissane.com/the-affordance-loop

I would bet actual money that I've got some nuance of Mastodon's functionality wrong in there so when you see that please do lmk. I thought about asking instance-running friends to preview it but I've taken up enough of their time this week.
@kissane I think you got the tech nuances pretty right on actually. Maybe worth a note there that quote posts, and full text search, and filling in replies that are missing now have pretty much full consensus that they all need to be fixed, and added and are all in some part of the roadmap to add.
@tchambers I thought I'd kind of gestured toward that, but maybe it's on the cutting room floor, there are so many words Tim
@kissane @tchambers The main things impeding quote posts and filling in replies, as I understand it, are:
1. They require changes to ActivityPub (for quote posts, the ability to opt out, being notified when you've been quote posted, limiting who can reply or quote post). This can’t be done unilaterally since ActivityPub is a W3C standard.
2. The Mastodon dev team is extremely short-handed. There is literally one full-time dev (Eugen is part-time since he has many roles). They're swamped.

@dgoldsmith @kissane

Yes, both. The first the important first impediment…the second more solvable by programmer help….

@tchambers @kissane Finding the programmers seems to be a problem.
@dgoldsmith @tchambers One of the Mastodon official dudes was saying that the organization/product management stuff was the real bottleneck there. I don’t have the link in front of me tho.
@kissane @tchambers
Here's a recent post by a member of the team: https://oisaur.com/@renchap/110742907279209554
Renaud Chaput (@renchap@oisaur.com)

@taylorlorenz@mastodon.social There is currently only one full-time developer working on the core Mastodon software (server and web app), and this is barely enough for keeping the lights up (bug fixes, security, maintenance…). Big feature work takes a lot of time and unfortunately Mastodon gGmbH does not allow more at the moment. Volunteers (like me!) are always welcome, but such a big work as Quote Posts require experience that few people have, and time that even fewer people have.

Oisaur
@kissane @dgoldsmith saw that: He’s not wrong from a maintainers and direct development standpoint - but volunteer programmers could create that implementation (once standardized) and submit it for their review, if coded well, would help push it downfield greatly….

Actually the quote post story is fun from somebody who understands Activitypub perspective:

  • It's trivial to implement on the ActivityPub level, see e.g. FEP-e232.
  • Lots of people have implemented it, see fiefish
  • People don't seem to care and continue to talk about it like a missing feature.
fep

Fediverse Enhancement Proposals

Codeberg.org

@helge Thanks @helge is that #FEP e232 the consensus & final one defining quote posts? My understanding was like this:

There are 3 competing standards for quotes: _quote: _misskey_quote , fedibird:quoteUri , and as:quoteUrl

Is that incorrect?

Does FEP-e232 pick one method (say “as:quoteUrl”) as the now standard way to do this?

FEP-e232 is the only proposal with a specification. So it is the only standard vs. ad-hoc solution.

Also FEP-e232 picks none of these options. Instead it adds the quoted object as a tag. This is preferable as it is similar to how mentions, hashtags, emojis are already being handled. Furthermore, it allows for multiple quotes.

@helge OK, my understanding was that contention between those other implementations was the main hesitation for mastodon or others implementing quote posts - worried that another implementations might become the final standard and then the code implementated before that would be wasted or actually hurting the final standard.

I would expect the reason not to implement quote posts is a form of decision paralysis. Most implementation (including taking a screenshot) provide the ability to quote post. They don't provide the ability to stop people from quoting something nor do they provide notifications.

Providing either of the two missing features is actually quite hard. Implementing either in a way that cannot be defeated by just posting a link, would require quite a bit of innovation. FEP-e232 provides for neither of these requirements.

I would personally encourage people to implement quote posts with the two features mostly relying on trust.

@tchambers @h I think Mastodon devs want to have a quote control mechanism, hence they are considering making quote-posts a special type of reply and using FEP-5624: Per-object reply control policies. All other implementations just add a reference to a quoted post.

There's a nice writeup by @trwnh: https://socialhub.activitypub.rocks/t/disambiguating-various-interpretations-of-a-quote-feature-pre-fep/3426

fep

Fediverse Enhancement Proposals

Codeberg.org

@silverpill @helge @trwnh

This is true: making a way for users to limit unwanted quote posts is a must have feature.

The problem with making "limiting who can quote post" a MUST HAVE feature is that it basically means: Quote posts won't happen soon.

Being able to limit other people's actions is an antithesis to openness principles. Bringing these things into harmony will require both social and technical work.

@helge Well: we could standardize both solutions now, no?
Or you could standardize quote posts NOW and suss out user limits on reply, quote posts, etc, as that works out. But I’d think any quote post solution standard should allow for near term user defined limits on who they want to allow in order to avoid harassment.
@helge @silverpill @tchambers i don't think it has anything to do with "openness". it's more akin to a stamp of approval that you can use to verify that the other actor accepted something. you're free to make an "unverified reply" if you wish. by the same principle, we might decide we don't care to verify the sender and therefore drop http signatures, leaving us with "unverified posts". this is no different than any normal web page which can make whatever claims it wants to.
@helge @silverpill @tchambers so the thing about "x control" for various types of links, is that you're never actually preventing the link. you're just verifying it, and possibly moving to require verification after some transitional period.