Foreword

This article is born out of several threads I've written on Mastodon (relevant references: here, here, here, here and here at least, but also see this and this for additional arguments) after the announcement of a Meta/​Facebook project codenamed Project 92 (P92) and also known as Barcelona that would be “compatible with Mastodon” (common speak for supporting the ActivityPub federation protocol).

(Update: If you got here from the Fediverse, you will probably also be interested in my call for a pre-emptive Fediblock on Meta's platform.)

(Update 2023-12-13: the platform has been launched under the name Threads on 2023-07-05 in the USA, and is scheduled to launch in the EU as well tomorrow, and has apparently started testing ActivityPub integration today.)

The article tries to present my perspective in a more organic way, and includes some additional considerations that I've more recently had the opportunity to make not only about what the Fediverse may expect from companies like Facebook “joining” it, but also from others like Bluesky Social choosing to not be compatible.

(And yes, as we're now in June it's getting late, and it's possible P92 will release before I finish putting this in place, but most of my predictions are for much later, 3–5 years from now at least, so they may still count as such rather than as a retrospective.)

One of the most important aspects to take into consideration is what I mean by some piece of technology or protocol getting “killed”: some people interpret this in a very strict sense, considering something dead only when literally nobody uses it anymore, while I (and several others) prefer a much more lax interpretation, where assassination is achieved when something is successfully removed from the awareness of the general public.

In the latter sense, it may even be argued that the “app mindset” pushed by modern mobile device (started by Apple) is on the path of killing not just the open web, but the World Wide Web altogether, shifting content distribution to native “apps” from browser-accessible web pages.

But this is a discussion beyond the scope of this article.

When Google killed RSS

There's some debate on what exactly killed RSS, or even if RSS is dead in the first place, as it is still extensively used —although mostly behind the scenes— on several platforms, but it's generally understood that the main blow to RSS came from Google discontinuing their Google Reader.

A relevant thread started by @eniko​@peoplemaking.games does a pretty good job at collecting the different opinions.

The OP stance is that RSS isn't really dead and even if it was, it wasn't Google's fault, as anybody could have created an alternative to it —that nobody did shows there just wasn't enough interest in RSS in the first place.

My readers know that I agree on the fact that Google wasn't the only party contributing to the RSS downfall —for example, I also blame Mozilla for pulling their Live Bookmarks feature from Firefox without adequate replacement— but Google's role cannot be overestimated.

Two key weak points in @eniko's post were addressed in the comments to the aforementioned thread.

The first pertains the semantics (that I've addressed in the introduction) about what does it mean to kill something like RSS. As mentioned by @elrohir​@mastodon.gal here and by @luigihann​@mstdn.social here, what Google shutting off Reader achieved was to kill off the mindshare that RSS had accrued, turning it from “something every site should have” into “well, if Google is looking elsewhere, it mustn't be that important anymore”.

Several other comments highlighted the second aspect, that the Reader shutdown also removed the “social” aspect of RSS perusal, just as the general public attention was being corralled to centralized, ad-monetizable social networks —which probably also explains why the Reader was sunset: it went against the prevailing “attention economy”.

So I would argue that not only the shutdown of Google Reader was the primary cause for the RSS demise, but also this was actually intentional, to wipe out from the general consciousness the awareness of the possibility and existence of decentralized, user-controlled forms of content distribution.

Several users have also remarked that RSS and Atom feeds still exist, citing podcasts as primary application (in fact, it could be argued that it's not a podcast if it's not available via RSS). I'll add to that that there's another form of media that still heavily employs RSS, and those that follow me won't be surprised by my mention of this: webcomics. A lot of self-hosted webcomic sites build on WordPress, so RSS availability isn't a big surprise, but I appreciate that (or rather, when) dedicated hosting solutions also offer the feature. (Comic Fury does it, and even WebToons, while AFAICS Tapas does not.) I use a self-hosted Tiny Tiny RSS instance to follow feeds, and the largest active (and still growing!) category is comics. The vast majority of the blog feeds I (used to) follow are either dead (as the hosting site shut down) or inactive, last updated some 10 years ago, with few exceptions (among which, the feeds from this site).

(And for what it's worth, I still visit the original posts, even when the full article is available in the feed itself.)

And of course, Mastodon and other Fediverse platforms also support RSS, as producers, as consumers, or both (Friendica being the best example of the latter). Several news sites still have feeds too, even though they don't always link to them in any user-visible part of the page, or even at all (secret URLs FTW).

Hence the importance of stating what exactly is meant by “Google Reader (or whatever else) killed RSS”, and why the mindshare aspect mentioned above is so important.

The technology is still alive, it's still being actively (and passively) used. Feeds are still being produced by a surprisingly large part of the Internet, and still being consumed, but the general public has been made unaware of this, to push the centralized social network paradigm of the attention economy. Unsurprisingly, I'm convinced there might be an opportunity now, with the Twitter takeover bringing attention to the Fediverse and rekindling interest in POSSE and PESOS, to reverse the trend.

And yes, as individuals this can be extremely frustrating, because the feeling is that even knowing what would be best to support the open web, there's the feeling that our choices have little effect if any. And in this context, the non-ubiquitousness of RSS production and consumption “where people gather” often means a need to double or triple the efforts to reach out and connect with other people with matching interests, especially for those whose livelihood depends on it.

I do have a feeling that we're approaching another paradigm shift: as the GAFAM enshittification progresses at increasingly rapid pace, times are getting ripe for the emergence of new contenders to the search engines and social silos that dominated the last 15 years, and ironically the strongholds of the old web are more likely to survive this change than anything that moved to the now sinking giants.

I'm not as optimist as others about the open web and indie web making a return comparable to the days of glory of the blogosphere, but I do have feeling that it will be a bridge, and those finding their space there will have an edge in whatever the new paradigm is going to be, even if it currently seems like a cause lost in the mist of unfriendly User Agents and single-user apps.

I'll take this opportunity to renew my plea to browsers that claim to be in favor of the open web that it's way past the time you put your engineers where your mouth is, and get back into leading the adoption of open technology, instead of blindly following the “leader” in their destruction of the same —and not just when it's a matter of hype.

(And yes, that was really aimed at Mozilla as an invitation to bring back RSS support in Firefox, and to openly promote JPEG XL instead of keeping it hidden in Firefox Nightly —and more in general to go against everything that Google's chokehold on the web currently represent. Support user choice instead of crippling it. Prove that you are different.)

The Fediverse is a threat to Big Tech silos —and they're preparing to counter

Shortly after posting about how Google killed RSS (as discussed in the previous section) I was made aware that Meta (Facebook's and Instagram's parent company) has been working on a “Twitter competitor”, codenamed P92 and also known as Barcelona (more recent news seem to indicate that the actual name of the product will be “Threads”) designed to “interoperate with Mastodon” (which is “common people” speak for “will support the ActivityPub protocol).

The timing of the news and my original thread was quite serendipitous, as it allowed me to recall that Facebook played a role in killing of RSS, through strategies better known from Microsoft's fight against open protocols and FLOSS: Embrace, Extend, Extinguish (EEE). Luckily, the memory is still fresh enough for some of us to realize that Meta's actions represent a threat for the Fedivers.

The obvious recommendation here is to defederate with extreme prejudice any and all Facebook-associated instance. We know how it will play out, we've seen it happen again and again. It doesn't matter how nice and cooperative they will be in the beginning, this is a power move against the Fediverse, a move against which it has no chance of resisting if even an inch is given.

(You want tags for that? We have tags! , , )

The only measure in which this news can be viewed positively is that the decision confirms the Fediverse as a threat to the social silos. Let's keep it this way. Do not trust Facebook. Do not forget that just last week Instagram tried to block the tag (relevant thread), lest people became aware of the existence of the world outside, a world that still has the appeal of early-days Instagram, and not subject to the threat of enshittification.

I'm quite confident that many (most?) of the people on the Fediverse (especially the old-timers), and most of the admins, will know better and avoid Facebook affiliation (pardon, federation). But here's the million-dollar question: will Mastodon's own founder and creator Eugen Rochko resist, or will he fall for the lure, trustly welcoming the Trojan horse of a Facebook-managed ActivityPub instance, believing it a sign of “victory”?

I would have loved to claim “most” in the paragraph above, but as the news circulated I found that the amount of people deluding themselves into thinking we should welcome the upcoming Meta instance with open arms was much higher than I expected. Truly people never learn anything from history. Seriously, how can we ever hope to find a solution to things like anthropogenic climate change when we can't even cope with the threat of surveillance capitalists siphoning off all the energy from the opportunity of a breakout?

The problem with awareness is, again, frustration: knowing how things are going to turn out, being powerless against people with stronger voices or with more power dismissing the warnings and insisting on pursuing their delusions of impermeability to the laws of physics (for climate change) or surveillance capitalism (on the Internet), so that the only thing that's left in the end is being in the position to say «I told you so.» when things inevitably go as you predicted.

(Relevant XKCD)

It gets worse: unless the Fediverse starts guarding against the Facebook torpedo (and by the looks of it, I'm starting to doubt it will) now, the people that don't believe the warning now will come up with excuses why all energy will have been siphoned off the ecosystem, blaming everything but the elephant that they let in thinking it couldn't lay waste of the china shop. They'll add insult to injury.

There's one thing I've been thinking again these last couple of days: would the situation be different if the client-to-server part of the ActivityPub specification had found the same level of adoption as the server-to-server side? One of the ugly things in the Fediverse today is that each platform has its own API to communicate, which is burdensome for clients, and fragments the ecosystem.

I know there are historical reasons for this (most major Fediverse platforms either precede the ActivityPub spec and later adopted it for federation, or tried to emulate the existing ones also client-API-wise), but I still wonder: is the C2S protocol sufficiently specified to allow the same functionality currently implemented via ad-hoc APIs in existing platforms?

If so, it would be good for platforms to start moving towards it (possibly initially supporting it side-by-side with the existing APIs). If not, it would be good for the C2S protocol to be improved (and/or better specified, if it's a problem of underspecification), so it could be adopted as above.

(In fact, I'm not even sure there are ActivityPub servers that use the client-to-server part of the protocol. I should probably look into the more obscure ones, like GoToSocial or Vocata.)

The sad tale of XMPP: a cautionary tale

There's actually a better example than RSS for the threat that Meta (and Google) pose to the open web and the indie web, one that is much closer to the Fediverse in spirit and scope, and that these two companies almost single-handedly managed to kill off (in the usual sense of siphoning all energy and mindshare off): XMPP, a federated, open, extensible protocol for chat and instant messaging.

When Google and Facebook adopted XMPP for their messaging products, a lot of people in the ecosystem were thrilled. Open source had won! The Big Ones were finally joining the FLOSS community with an interoperable, federated product that allowed any client to be used with their system, and communication between people with account on other servers!

Yes, the rhetoric at the time was exactly the same that we're seeing from people looking at codename Barcelona with enthusiasm.

The rest is history. The XMPP protocol at the time had some glaring inadequacies in some areas, so it had to be extended to cover those use cases. Instead of working with the community at large to converge towards common extensions for logging, device switching, audio/video and conference calls, Facebook and Google went with the most classic of rug-pulls, first by defederating their servers (lock users in), and then by switching out of the protocol altogether.

(Interestingly, Facebook was much more successful in locking people in than Google, probably partly due to the very schizophrenic approach to products that the latter has, but Google also kept their XMPP compatibility for longer: access was shut down only on June 2022, years after the defederation and transition away from Chat.)

I'm sure you can see now why I am less than thrilled about the enthusiasm with which people are looking at Facebook “joining” the Fediverse with their ActivityPub-compatible Twitter clone, despite there being literally no indication that their attitude today is any different from the one at the time of XMPP —and if anything, the opposite, as we've seen with Instagram blocking the tag.

Nostradamus: how Facebook will torpedo the Fediverse using P92/​Barcelona

So I'm going to make a prediction now, about how this will go if their server isn't treated exactly like Gab, the Nazi social network. Feel free to bookmark this for reference and set a reminder for a couple of years from now to tell me (and everyone still interested in listening) how right I was (and rest assured I'm not the only one: everybody with a good memory thinks along the same lines).

Prediction: Meta's P92/​Barcelona/​Threads joins the Fediverse. Many instances (especially large ones and ones with an obsession with “growing the network at any cost”) federate with it. The Fediverse sees a growth like nothing before, with the Twitter migration in November last year paling to a glitch by comparison. Smooth sailing for a couple of years, but nearly every new account is created on Meta's side. Meta possibly brings Instagram in. @dansup​@mastodon.social, PixelFed creator, rejoices.

Prediction, part 2: a couple of years in, with over 90% of the Fediverse on Meta servers, the company goes for the rug pull, either on bogus technical claims on the protocol, or on the basis that “everybody is here anyway”. They defederate from everything. The Fediverse idea loses momentum, participation falls back to diaspora* levels. Tankies explain to us that this was due to ActivityPub not being good enough and not due to Meta killing it because it was a real threat.

I think that the main delusion that shines a positive light on the Barcelona thing is the illusion that this kind of connection can help bring people out of Facebook or Instagram, despite there being no indicator that this would actually happen, white there are several ways for Facebook to extend their control on the Fediverse through that same gate (see for example this write-up by @darnell​@one.darnell.one on a possible scenario for the takeover).

Striking a balance

Among the comments to my first run of these thoughts on the Fediverse, one of the key point that emerges is that the hardest task for the Fediverse is to find the right balance between moving beyond the “cool tech for nerds” phase towards a more general public acceptance, and avoiding falling into the hands of a small set of centralizing players that would ultimately get to decide what to make of it.

In many ways, the parable of the Fediverse is replicating that of the World Wide Web, that failed to deliver on its promise as a tool for the democratization of knowledge (both production and consumption) as Big Tech found a way to monetize it concentrating power at the expense of the open web.

This is also the reason why I believe strongly into the objective of a “Fediverse of all”, but I do not believe that this can be achieved with major players that have nothing to gain from it entering the playing field. This is also why I'm skeptical that the best (or only) way to counterbalance the Meta threat is for other major players to get into the field, Automattic's Tumblr being the best candidate if Matt's promises carry any weight: I do believe that having more than one “silo-derived” server on the Fediverse would be less dangerous than having only one, but I do not believe it is enough. Again, the XMPP history teaches us something, since both Facebook and Google being onboard did nothing to avoid them from killing it —and in fact probably made things worse in the end. But even if any of the majors would join in good faith (something that I would trust Automattic to do more than Meta), they would still carry too much weight, and could easily choose to expunge all the minor servers from their choice of federation, making it nearly impossible to successfully self-host an ActivityPub instance similarly to what's happening with email.

There are technical solutions that could alleviate this issue (first and foremost nomadic identity as supported by HubZilla or Streams), but I'd honestly put more hope on regulation such as the EU's Digital Markets Act forcing the giants to keep interoperating.

Meta Fediverse: validation or trap?

It's not that I don't get the enthusiasm about P92/​Barcelona —that's not the reason why I warn against federating with it. I do get it. There is something elating, validating, empowering even, when some{one,thing} Big & Famous (seems to) adopts “underdog” tech. I know because I've been there, both as a user and as a developer. But because of that, and for having been burned already not once, not twice, but three times at least, I know what to look out for.

The single most important thing to look out for is the difference between actually adopting a technology and “adopting” it. And this must be looked at not from the perspective of the user, but from that of the Big & Famous: how can they best use it to serve their own interest?

And here's the thing: Meta/​Facebook has literally nothing to gain from actually adopting ActivityPub and federating with the rest of the Fediverse, exactly as it had nothing to gain from RSS or from XMPP.

In fact, all three of them go completely against everything that company and its products stand for. So why would they even bother adopting them if they weren't dragged into it kicking and screaming? Because “adopting” them gives them a unique opportunity to destroy them thanks to the company market size. And this is exactly what they've done with RSS and XMPP, and what they will do to ActivityPub if we let them get even just a sliver of an inch in.

One of least-sensical objection I've had to my threads on Mastodon, worse than the uninformed “oh but we still use RSS/​XMPP” or “that's not the reason why they failed”, is “the products Facebook and Google used to allegedly sink XMPP just flopped”.

This objection completely misses the point that the purpose of these products was never to succeed. Their entire purpose was to die and bring down the competitors with them. And this is absolutely the case with P92/​Barcelona too.

“There's no guarantee that P92/​Barcelona will succeed” isn't a reason to not obstruct it in every possible way —in fact it's worse than that: not only it's not guaranteed to succeed, but it's pretty much guaranteed that even Meta doesn't care if it succeeds, because that's not why it's being developed. This is a golden opportunity for them to take a sweep at all of the competition at once: they can give a finishing blow to Twitter, trip BlueSky and siphon out all the energy from ActivityPub/​the Fediverse, all in one go!

Now you may consider that the Twitter/​BlueSky tripping combo might be worth rooting for, for people on the Fediverse, but are you willing to sacrifice the future of the Fediverse on that? I can safely say that I'm not.

And again: the Fediverse as it stands now has no chance to resist the Meta EEE torpedo. It is not large enough, widespread enough, or resilient enough to withstand it. The bridge will not help people move from Facebook/​Instagram to the Fediverse, it will only suck many of those that have made it here back into the silo.

I mean, seriously, the Fediverse can't even protect itself from the opinionated, crippling, overreaching influence of Mastodon, how can you delude yourself into hoping it can withstand intentional sabotage from one of if not the single most unethical social media giant on the Internet?

To get an idea about what I'm referring to, consider this. As I mentioned, much of the optimism comes from a belief that, from the federation, Meta/​P92 users would get a “window” on the Fediverse, which would tickle their curiosity and provide them with the opportunity to explore this connection to experience social networking outside of the giant's grasp, find it superior, and migrate out thanks to the alleged lack of risk of losing their network when moving out. Yet even most Mastodon users aren't even aware of the wider Fediverse, thanks in part to the platform being (intentionally?) designed to hide this, by not rendering rich text formatting, providing inadequate support for non-Note object types, no indication of the originating platform for contacts and posts coming from non-Mastodon angles of the Fediverse (compare with Friendica proudly showing its extensive federation support), etc.

So why would the tens or hundreds of millions users Meta can instantly put on P92 ever be made aware of the diversity of the Fediverse, when their only window on it will be controlled by Meta, who can do everything to make it as non-obvious as possible?

So, again: federation with P92 will not show the richness of the Fediverse to Meta users: it will only give unsatisfied Fediverse users the opportunity to jump back into the corporate silo before the door closes again.

And there are plenty of such users on Mastodon, people all but forced to jump ship when Twitter went to hell that would be more than happy to go back to anything resembling their idealized memory of “old Twitter”, people rightly disgruntled, unsatisfied by the innumerable idiosyncrasies of federation (and especially the way it's handled by Mastodon), people that would be easily corralled again, to their own future detriment, by a sufficiently “shiny new thing” with a promise (soon to be disregarded) of interoperability.

I've seen several people proposing already in advance to “fediblock” P92/​Barcelona because, being a Meta product, it poses a threat to the privacy of uses across the entire Fediverse. See for example this comment by @smallpatatas​@mstdn.patatas.ca and @atomicpoet​@calckey.social response (which coincidentally is also an excellent example of how the much-aspired but controversial “quote toot” feature requested for Mastodon and implemented by CalcKey destroys the flow of discussion even when not used maliciously) for two very different views on the matter.

And the privacy issue is quite legitimate. But honestly, to me this is a “second order” worry, Now, if the Fediverse had the resilience to plow through Meta's attempt at torpedoing it with P92/​Barcelona, or if Meta actually had good intentions regarding federation, the privacy angle would be something to think seriously about (whatever decision is taken ultimately, I assume different instances will have different takes on the matter). But if the EEE strategy works, privacy will be the least of our worries, as the Fediverse will crumble back to “also ran” numbers.

For what it's worth, I do wish the Fediverse had that kind of resilience. Especially since I wouldn't have to write this many threads (or this ginormous article that I'm editing now, collecting them) on the topic of the dangers of federating with P92 —because the dangers I'm talking about wouldn't be there in the first place. But until and unless the Fediverse becomes as pervasive as email, until links to Mastodon and the Fediverse become as common as links to Twitter, Facebook or Instagram in the “social media” sections of websites, until every person with a following on Mastodon will sport such a link proudly on their website (and there's still a lot that don't), the Fediverse simply doesn't have the kind of mindshare that would allow it to resist the P92 torbedo.

The Fediverse has not even reached critical mass for the network effect to bring more people in at a significant rate (outside of the bursts of the Twitter migration, the last of which was months ago), let alone to keep the ones already in from jumping back out.

So let's stay away from it. If you really do care about the good people you know on Facebook, Instagram, WhatsApp, tell them, help them come to the Fediverse now, and block P92. Don't wait until the damage has been irrevocably done.

Prevention is better than cure.
A stitch in time saves nine.
Better safe than sorry.

And yes, I know, it's true, it doesn't matter what I say. It won't be treated like Gab, it'll find lots of friendly instances willing to federate, especially among the larger ones (mastodon.social anyone?) I guess the only thing left for me is to keep this article (and the Mastodon threads it comes from) as future reference, when I'll get the opportunity to spam “I told you so” under every optimistic comment in a couple years' time.

Forewarned is forearmed.

Not just corporate hate

There was an interesting poll by @matthieu_xyz​@calckey.social about the preference to block a number of other platforms if/​when they federate: the list mentioned Meta, Tumblr and WordPress, all presented as “closed source, corporate” platforms.

I found the list a bit curious, because WordPress is not even closed source, and while its development is controlled by a corporation (Automattic, that has since bought Tumblr too), it has been in many way a champion of the open web, always supporting open standards and even helping define new ones and pushing for their adoption. It wasn't a surprise that the “block WordPress” option in the poll was the one that received the least votes —especially since WordPress can already federate using a dedicated plugin.

The situation with Tumblr is definitely different from that of WordPress, despite them being both owned by the same company now, but it's also different from that of Meta. So while Tumblr joining the Fediverse would pose a problem (especially since, in contrast to WordPress, it's not a myriad of self-hosted websites, but a single huge entity that would be perceived a single server on the ActivityPub network), it would be largely for different reasons than P92/​Barcelona.

And yes, this may be considered a bit naive, but it is due to the enormous difference between how Meta does things and how Automattic does things. While I'm wary of both, Facebook has already played the EEE game with XMPP and RSS: there's no reason to expect it will be different for ActivityPub. WordPress and Tumblr, on the other hand, are managed by a company that has always played “by the rules”. And yes, this doesn't make them immune to the risk that would come from a catastrophic change of hands in Automattic, but at the least this isn't in the horizon now, and while it may be an excellent reason to not create an account with them, the federation aspect may be considered with less worry than Meta's.

Even now we can see the difference: Automattic hired the developer of the WordPress ActivityPub integration plugin, so that he could work on it full time and improve the existence for every WordPress installation out there. What would Barcelona bring to the table, outside of a massive instance people will flock to because “it federates anyway” and then be rug-pulled out of their contacts when enough people are there?

I would also highly recommend reading @atomicpoet​@calckey.social's take on whether it's a good idea to block P92/​Barcelona or not, even though I (obviously) disagree with it on several points.

For example, whether or not (or how many) people will use Barcelona shouldn't really factor in. In fact, precisely because Meta can populate that service simply by giving an account to every Instagram user (which the last time I bother to check was kind of the plan) it should be defederated en masse. The point of blocking P92 isn't to make it fail, it's to prevent its usage as an EEE torpedo against the Fediverse. And again, not even Meta cares if P92 survives in the medium/​long term: as long as it manages to sink Twitter, BlueSky and the Fediverse, its work is done. And in the process they may even gain that non-insignificant fraction of the current Fediverse users that don't actually care about freedom, privacy and federation, and are on the Fediverse only provisionally, waiting for a “less Nazified, but similar to the old Twitter I remember” platform to come around —which is exactly what Meta is going to give them.

And yes, corporate-controlled media and corporate-brainwashed drone will look at the fediblock negatively. It wouldn't be the first nor the last instance of them being catastrophically wrong. But most importantly, not blocking will not bring positive presentations of the Fediverse in such media either: even now a lot of “tech” journalists have been disparaging it (and Mastodon in particular) for being too different from Twitter (but mostly for it not being corporate-controlled). Worse, just as most of them are “blissfully” unaware that Mastodon is just a small (albeit significant) part of the Fediverse, with Barcelona joining in they'll be guaranteed to completely reframe it all in terms of Meta's platform: no corporate outlet will give the Fediverse as a whole any serious weight, it's not how they reason, and most importantly, it's not what they'll be paid to do. You can be sure Meta will strive to make sure the entire federation discourse be reframed in terms of their platform, to help the Fediverse (and Mastodon in particular) disappear from the collective consciousness.

And as I mentioned already, I agree that it's important that people migrate off Meta services, and the Fediverse is the best there is now to help them to do that. But federating with P92 will not help achieve that. P92 is neither Facebook nor Instagram, it's a new thing, and when (not if, but when) it will defederate, it'll push a lot of people to move from the Fediverse to P92 to remain in contact with the people there. I predict that federating will not “buy” the Fediverse any new user migrating off the Meta platforms, but it will cause it to lose a lot of active users when Meta pulls the plug.

One of the aspects where @atomicpoet is (IMO) exceedingly optimistic is that Barcelona federating would give Meta users the opportunity to experience the Fediverse, which could make them interested in migrating off Meta's platform, since with P92 in the federation they can do this without losing their social graph. But interaction with users from other platforms does not raise awareness of the existence of these other platforms, as we see with Mastodon even now: a lot of people interact with users on other platforms and remain blissfully unaware, because Mastodon hides this information, both by hiding the platform of origin and by crippling content that isn't just short notes with no formatting. Unless someone actively bombards them with the information, the fact that other platforms exist and interoperate remains a “behind the scene” detail for most, and often a source of issues due to subtle differences and incompatibilities between these platforms and Mastodon. If we can't fix this for Mastodon, we have no hope of making it work against P92, that will go to extra lengths to hide this information from the users on purpose.

On a technical level, the Fediverse platforms competing with Meta's largest horses, Facebook and Instagram, even when reasonably mature, are still far from giving potential migrators all that they would expect. People have complained about Twitter feature missing from Mastodon, despite Mastodon largely offering a better feature set than Twitter. It would be even worse with Facebook/​Instagram versus Friendica/​PixelFed.

And even if the Fediverse platforms were mature enough to work as drop-in replacements for Meta's, there still aren't enough servers out there to absorb a potential mass migration —and especially for media-heavy platforms, this is unlikely to change any time soon.

Finally, even if there were enough hosts, people still wouldn't migrate, couldn't migrate actually, because Meta is not going to federate Facebook and Instagram: they're setting up P92 as a separate platform, so they can control what its users will see. P92 users will not wonder why PixelFed users have reels, because Meta will make sure that incoming posts will be crippled enough to not show anything that may make them wonder. Again, even Mastodon often fails a lot in managing correctly inbound content —and that's not out of malice. Now image the same, but with the express intent of making the Fediverse experience as crippled as possible so that people stick to Meta's platform.

What's next?

I conclude here the part about P92/​Barcelona and how (most likely) Meta will try to use to take a jab at all the competition, both corporate and open/​decentralized. I've posted an appeal to Fediblock their instances on Mastodon, in a most probably misplaced hope that it will be heard enough to avoid the worst-case scenario.

Next up (when I finish writing it) will be some (hopefully shorter) considerations on BlueSky, the new brainchild of the former Twitter founder Jack Dorsey, and how it's likely designed to be a different approach to corporate capture of the idea of federation, and thus the second prong of a coordinate attach on the open Fediverse.