A clear victory for the free fediverse: Meta now says integrating with ActivityPub is "a long way out"

Join the discussion on Lemmy and infosec.exchange!

When Meta launched their new Twitter competitor Threads on July 5, they said that it would be compatible with the ActivityPub protocol, Mastodon, and all the other decentralized social networks in the fediverse "soon" – and  Mastodon's Eugen Rochko welcomed it as a "clear victory for our cause."

But on July 14, Alex Heath of the Verge reported that Meta's saying ActivityPub integration's "a long way out".  Hey wait a second.  Make up your mind already!

Then again, Meta is Facebook (they changed their company hoping people would forget that Mark F***king Zuckerberg Is Not Your Friend), so it's not surprising that they gave a false impression originally.  Also, Meta is Facebook, and for some strange reason some people in the fediverse see Facebook, Instagram, and Meta's track record – welcoming hate groups, making money from disinfo and discriminatory ads,  backing authoritarian politicians, experimenting on their users, privacy abuses, contributing to genocides, etc – as problematic. So even though guys like Rochko, long-time decentralized social network Evan Prodromou, and comedian John Gruber are eager to welcome their new surveillance capitalism overloards, others disagreed.  For example:

• Hundreds of instances (servers running Mastodon or other software) signed the Anti-Meta #FediPact pledging – and as the week went on more and more instances preemptively blocked threads.net.  
• Polls consistently showed most people want their instance to block Meta, with many are willing to move if that didn't happen?
• Opposition from trans and queer people and LGBTQIA2S+-focused instances was overwhelming especially after Meta's refusal to moderate anti-LGBTQ+ hate groups on Threads?  

And from the perspective of the "free fediverse" that's not welcoming Meta, the new positioning that ActivityPub integration is "a long way out" is encouraging.  OK, it's not as good as "when hell freezes over," but it's a heckuva lot better than "soon."  

In fact, I'd go so far as to say "a long way out" is a clear victory for the free fediverse's cause.

The strong and broad resistance to Meta reinforces the point I made in In chaos there is opportunity!: Meta's potential arrival is a likely to be a good thing for the fediverse no matter whether or not they actually go forward with it. And as I predicted in A (partial) queer, trans, and non-binary history of Mastodon and the fediverse. queer, trans, and non-binary people played a big role in shaping the future. Let’s take a moment to appreciate how different this is from the situation on any of the the big commercial social networks!

That said, opinions differ; a lot of people in the fediverse do want to welcome Meta.  So Meta and their supporters aren't going to give up on their plans to embrace and extend the fediverse just because of the pushback.  And this is Meta (aka Facebook) we're talking about here, so don't assume that a "long way out" means anything more than "this is the story we think it's useful to tell today."  I'll talk about their likely next steps below, along some good learning from the situation so far.  

But first, congratulations are in order!  Kudos to

• everybody who spoke out and talked about why they saw Meta's arrival as putting people at risk and clashing with the fediverse's values;
• every instance that signed the #FediPact and/or announced they were blocking Meta
• all the activists who got word out about reasons not to welcome Meta
• and #FediPact creator vantablack, for drawing a line in the sand and mobilizing people to take action.

Well done!!!!

The next step in embrace-and-extend, coming soon to a standards working group near you

"This week a representative from Meta, a software engineer named Ben Savage, joined the ActivityPub working group in the W3C."

– Threads Adopting ActivityPub Makes Sense, but Won’t Be Easy, Richard McManus, The New Stack

Meta certainly made a big deal of "the fediverse" in Threads' launch, and they have lots of good business reasons for taking a decentralized approach, so they're not going to give up just because they're not wanted.  I can't remember the last time that Facebook, Instagram, or Meta actually created a successful new product, but Threads' initial launch went well in terms of usage: 100,000,000 users signing up in less than a week, including lots of brands as well as white supremacists and anti-trans hate groups – key market segments that bring in a lot of revenue on Facebook and Instagram.  Twitter is so horrible these days, and other commercial social networks aren't offering good alternatives yet; so if Meta executes even halfway competently they should see some signifcant usage. Then again, that's not a given – Threads' initial functionality is remarkably weak and (for many) depressing.  Time will tell.

As Richard McManus points out in Threads Adopting ActivityPub Makes Sense, but Won’t Be Easy, ActivityPub is a natural choice for Meta because it allows them to control the identity layer once they federate. It wouldn't surprise me if they're currently testing ActivityPub integration with one or more friendly Mastodon admins – or with brands like Disney, who are natural partners with audiences far more valuable to Meta than the current fediverse. Still, ActivityPub has some major weaknesses, so it also wouldn't surprise me if Meta's current position of "a long while" turns into "ActivityPub isn't ready for us yet so here's what we're doing in the interim".  

One obvious next step for Meta is to implementing proprietary ActivityPub extensions that address real problems in the current spec (just as Mastodon and other implementations do).  As well as putting them in control, this gives Meta leverage to pressure the World Wide Web Consortium (W3C) SWICG (the "ActivityPub working group" McManus referred to, although technically it's not a working group) to make sure the next version of the standard is Meta-friendly.  After all, if it doesn't meet Meta's needs then they'll have to (gasp) rethink their "commitment" to ActivityPub.  Oh noes!!!!!!!!!

It's a classic embrace-and-extend strategy, and Ploum's How to Kill a Decentralised Network (such as the Fediverse) – describing the dismal outcome in a somewhat-similar situation with the XMPP protocol – is a cautionary tale. That said, there are also other perspectives on what happened with XMPP.  Christina Warren, for example, highlights how XMPP advocates denial of key weakensses in the spec contributed to the situation, and the ActivityPub spec has some key weaknesses as well; Prodromou highlights the responsibility that he and others in the XMPP community had in not addressing the spam and harassment problems, and as I pointed out in An interesting dynamic, "some of the loudest voices for working with Meta ... don't actually have a  good track record of prioritizing user safety." Any way you look at it, though, the parallels shouldn't be particularly comforting for the people welcoming Meta's embrace.

Of course, Meta's a dues-paying member of the W3C, so of course they're allowed to be on the SWICG – and whatever working group gets the charter for the next version of ActivityPub (which as I understand it is beyond the scope of the SWICG).  And a lot of anti-Meta people (including me) see the cards as so stacked in favor of Meta that we're not likely to spend much time in the process.  As a result, Meta's very likely to get what they want in future versions of the "official" ActivityPub standard, although they'll no doubt still offer proprietary extensions as well.  

ActivityPub advocates are optimistic that they'll be able to resist embrace-and-extend.  That seems optimistic to me (and quite frankly the free fediverse approach of just saying no to Meta's embrace seems a much better path to resisting) but we shall see.

Good learning all around!

One of the interesting dynamics of the discussion so far is so much of the resistance to Meta has come from queer and trans people, and that most of the loudest supporters of Meta in the fediverse are cis guys.    Of course, Meta doesn't care what queer and trans people think – like I said, they're currently allowing anti-LGBTQ+ hate groups to run amok on Threads – so maybe they expected the pushback they got.  

Then again I also have to wonder whether Rochko, Prodromou, and Meta's other supporters in the fediverse had set corporate expectations properly.  Comedian John Gruber, for example, was candid about his view that it's only "misfit loser zealots" who don't understand why the arrival of a genocidal racist anti-LGBTQ+ surveillance capitalism company like Meta is such a great thing. So if Meta actually was foolish enough to think that Gruber et al knew what they were talking about, they might well have mistakenly believed if only a few extremists who had reservations, and that most of the fediverse would indeed eagerly welcoming the creepy embrace of their new surveillance capitalism overlords. If so, they now know better.

And if anybody in the fediverse was foolish enough to think that the cis guys welcoming Meta's embrace are prioritizing user's safety, or are looking out for trans and queer interests, they also now know better as well.  

Good learning all around!

Things are looking good for the "free fediverse"

"Depending on how things evolve, it's quite possible we'll see a schism (or partition), with instances in the anti-Meta "free fediverse" blocking not just Meta instances, but also any instances that federate with Meta.... Threads' entrance to the fediverse actually highlights the advantage of "free fediverse" instances – which are likely to be much more appealing to the increasing number of people people who want as little as possible to do with Meta, surveillance capitalism, and all the harms and abuses that come with it"

– There are many fediverses -- and a lot of people will prefer them to Threads

Meta's revised timeframe means that any potential schism won't happen for a while, and it's possible that the software could evolve in ways that make it less likely.  One good outome from Meta's potential arrival would be for Mastodon and other software to start making progress on privacy and safety issues that have been ignored for years.  If that happens before Meta starts federating, them there might be less pressure for transitive blocking.

Still, whether or not a schism happens, this initial skirmish highlights the tension between different camps in today's fediverse.  Back in 2017's  Lessons (so far) from Mastodon for independent social networks, I noted that "Policies against racism, sexism, discrimination against gender and sexual minorities, and Nazis are extremely appealing positioning these days", and it's still just as true. Meta's advocates in the fediverse eagerness to ignore or downplay Threads' coddling of white supremacists and anti-trans groups throws that advantage away.

And by prioritizing their desire to be embraced by Meta over queer and trans people's safety, Meta's cis advocates undercut their claims to be allies in ways that may be hard to recover from.   In chaos there is opportunity!  suggests that one of the good outcomes from the Meta situation is that it will "almost certainly upend the power structures that have hindered progress," so it'll be interesting to see how this plays out.

Meanwhile, things are looking good for the free fediverse. There's clearly a critical mass of people who are in the fediverse to get away from Facebook, Instagram, and other commercial social networks – and if some of the fediverse's other problems can be addressed, that's also an extremely appealing positioning to more and more people these days.  

Of course, there are quite a few issues that need addressing in the fediverse.  In chaos there is opportunity! discusses several more, and highlights that Meta's potential arrival is likely to catalyze progress.  It's an optimistic take (and it's still not clear where the funding will come from), so we'll see whether the optimism is justified. But the results so far are certainly encouraging!

Update log

July 18: updated description of W3C groups, including getting the acronym right and highlighting that the SWICG doesn't have the charter for the next version of the spec.