from the the-decentralized-world dept
Today is six months since Elon took over Twitter and began this bizarre speedrun of the content moderation learning curve in which he seems to repeatedly… not learn a damn thing. Over and over again he makes ridiculous choices that have made the entire platform less welcoming for speech, more willing to obey government demands, and even when he finally comes around to realizing that what Twitter was doing before was a sensible approach, he reimplements it in the dumbest possible manner. It’s uncanny that one guy could be so bad at this.
There have been a bunch of attempts at filling the void left by an unstable and untrustworthy Twitter, and it’s been fascinating to watch how it’s all played out over these past six months. I’ve actually enjoyed playing around with various other options and exploring what they have to offer, so wanted to share a brief overview of current (and hopefully future options) for where people can go to get their Twitter-fix without it being on Twitter.
The Decentralized Leaders
Back in December, as I got comfortable with Mastodon, I was somewhat confused as to why anyone would spend the time and effort to invest in building up a new social graph on another centralized social media platform, which would just become subject to all of the same issues that Twitter and other centralized social media platforms had faced.
I still stand by that today, though now I think there are three really interesting decentralized social media protocols worth paying attention to. A decentralized protocol-based system is interesting for all the reasons I outlined in my paper a while back, so I won’t repeat them all here. But, it creates much better incentives, and avoids the possibility of one random jackass controlling everything. Some of them may have to deal with jackasses who have influence, but none can control the overall system, and just that fact alone is incentive for anyone prone to jackass-tendencies not to go full jackass, or people will… just route around them.
Here are the three most interesting decentralized projects:
ActivityPub/Fediverse/Mastodon: Over the last six months I’ve spent more time on Mastodon than anywhere else, and the community there is fantastic. I understand why some people complain about the onboarding process, or the lack of some features (text search and quote tweets being the two biggest). But, honestly, if you spend 15 minutes playing around with stuff, and follow a reasonable number of active accounts and (most importantly) start interacting and actually talking to people, it quickly becomes a very fun place.
Obviously, that only works if the communities you want to interact with are there, and for me, there’s definitely a critical mass of the kinds of accounts I find most interesting. There’s plenty of tech news, tech policy folks, and computer security folks. And, the conversations are much more engaging (it sucks up way more time than I used to use on Twitter because something about it seems to encourage more conversational setup than Twitter).
That said, the limitations are real. The learning curve aspect of it really seems to anger some people, and some of the early decisions, while there were reasons given for them, clearly are limiting some ability of the fediverse to reach that next level. It still can get there, and I’m excited about the companies who have embraced it, as well as the variety of really amazing clients that are now being offered, such as Elk and Phanpy both of which provide a more Twitter-like experience for those who want that, or Mastodeck, which provides a Tweetdeck-like experience.
There are also ActivityPub-compatible platforms that are not “Mastodon” but connect with Mastodon and actually fix some (but not all) of Mastodon’s limits. But, some of the concern is that people so associate the wider Fediverse with “Mastodon” that they don’t even realize you can sign up for services that interoperate with Mastodon, but which include tools like quote tweets and search. So, services like Calckey, for example, are building out some more user friendly features. It’s not Mastodon, but it feels like Mastodon with a better UI and interacts seamlessly with Mastodon.
Another “limitation” to Mastodon is its cultural norm against “algorithms.” I think this is somewhat misguided. I get why people don’t like the algorithms on centralized social media platforms like Twitter and Facebook, but when done right, algorithms are super useful in making it easier for you to find more signal over noise. The problem has been who controls the algorithm and what are they tweaking it to do.
With Mastodon there are some 3rd party algorithms, and some of them are really useful. But a few others were shouted down, and shut down, by people who believe that there should never be any algorithms in Mastodon at all, and that’s unfortunate. In the early days, I would talk about some of the cooler Mastodon algorithms I’d been finding, but after a few them then were yelled at by a bunch of Mastodon users, I’ve generally decided it’s not worth promoting those useful tools, for fear that people yell at them to shut them down.
Bluesky: I started writing this post last week, and at the time, I was talking more about the theoretical possibilities for Bluesky, and why I’m really excited about where the project is heading. But over the last week, really, it went from being a tiny obscure platform a few people were testing to “the new hotness” and has been repeatedly trending on Twitter, as lots of people are checking it out, including some big names.
I’ve written a bunch about the project from when it launched, and when Elon completed his takeover of Twitter, I had jokingly sent someone at Bluesky a note asking if they might be able to code a little faster…
There’s almost no learning curve, and it does provide a very Twitter-like experience, though with some limitations, as you might expect from an early beta offering.
There are also a ton of misconceptions about Bluesky, which can be a little bit frustrating, but that is part of today’s ecosystem. Contrary to what many people seem to believe, it’s not a blockchain and is not built on a blockchain. Also, while Jack Dorsey provided the seed funding for it, Bluesky is not run by Dorsey, and both Dorsey and Bluesky CEO, Jay Graber, have been somewhat public about where they disagree on where they expected Bluesky to go, with Graber sticking to her vision and Jack focusing more on a different protocol (that one’s up next).
Also, while there are a growing number of folks joining it, it’s still an invite-only beta, and that’s deliberate, not to “build buzz” as some people assume, but because there still is a bunch of functionality that hasn’t been built yet, which the team knows they need to have before really opening the doors.
The team has been working like crazy on those features, including its federated setup, so that it’s truly decentralized (right now you can only use it through Bluesky’s own service). I’ve seen some people complain that making it federated will “ruin” it, but that’s almost entirely based on their experiences with ActivityPub federation, which does have some limits and is presented in a manner that confuses at least a certain segment of people who encounter it.
But there are ways to build federated systems that aren’t confusing. Anyone who uses email already recognizes this, perhaps unconsciously, because that’s a federated communication system.
Most importantly, the team at Bluesky know that this was a serious weakness of Mastodon, and are designing to make the federated part be less pronounced and something that the average user won’t need to care about, but which will allow for some really interesting developments.
The Bluesky team has also been quite open and upfront and willing to engage on these issues, and trying to find the right path forward. I know that some people remain annoyed that Bluesky built their own protocol, the AT Protocol, rather than adopting and building on ActivityPub, but the team behind Bluesky has been quite clear from the beginning that they considered all options on the table. Before she became CEO of Bluesky, Jay Graber wrote the definitive exploration of decentralized social media networks back in 2020, and has laid out the pros and cons of nearly every approach.
While people may not understand why certain decisions are being made, it’s often because they haven’t deeply thought through all the pros and cons of every approach. Bluesky, from the beginning, has tried to build a setup that takes the best concepts of a distributed social media system, and build them in a way that minimizes the problems, limitations, and pain points that some have had with something like Mastodon.
Just as an example, Bluesky has been clear why there were limitations with ActivityPub that would limit how useful Bluesky could be:
Account portability is the major reason why we chose to build a separate protocol. We consider portability to be crucial because it protects users from sudden bans, server shutdowns, and policy disagreements. Our solution for portability requires both signed data repositories and DIDs, neither of which are easy to retrofit into ActivityPub. The migration tools for ActivityPub are comparatively limited; they require the original server to provide a redirect and cannot migrate the user’s previous data.
Other smaller differences include: a different viewpoint about how schemas should be handled, a preference for domain usernames over AP’s double-@ email usernames, and the goal of having large scale search and discovery (rather than the hashtag style of discovery that ActivityPub favors).
Finally, I’ll just make quick notes on two “controversies” that have already hit Bluesky, and both of which I think are misguided. First, the service does not yet have a “block” feature, which is generally considered an important first option for safety of users on a social media platform. I’ve seen a lot of anger over this, and even people insisting to me that it proved that Bluesky was ill-prepared to run a social network.
But, that’s not true. Graber explained in a long thread that blocking will be introduced very soon, and the reason it wasn’t yet had nothing to do with them not prioritizing it, but rather because the underlying system is designed to be decentralized and federated, there are a ton of questions about how block works across multiple servers, that couldn’t be answered until the federation setup was ready, and that hasn’t been released just yet (though it will be soon).
They’re figuring this stuff out, and trying to do it right, rather than rush something half-assed. It’s a good approach and one that I find encouraging.
The second “controversy” that has popped up a few times in the last week is that their terms of service are full of scary-sounding boilerplate legal language that people regularly misunderstand, and then attribute the worst possible motives to the company. I mean, we’ve done posts on this exact thing before. It always happens.
No, Bluesky hasn’t banned screenshots. It’s not planning to take your artwork and sell it. Admittedly, some of the terms of service are clunky (and I think there are a couple of the terms that, in their present form, will run into trouble with the EU if they’re left in place). But, the company came out publicly last week and admitted these were quick boilerplate terms, and they had already begun working with lawyers on a complete rewrite that will be user friendly.
nostr: On this one, I can already hear some people groaning. If people thought Mastodon was too confusing, they’re not going to like nostr, as it seems to confuse people a lot more. In some ways, this feels ironic, because the amazingly cool part about nostr is just how freaking simple it is. From a technical standpoint, nostr is kind of beautiful in its simplicity. In talking about it with friends, I’ve had multiple people say that they were tempted to build tools (clients or otherwise) for nostr as an afternoon project, because it was just that simple.
Nostr isn’t federated like Bluesky or Mastodon. There aren’t distributed servers. Instead, it’s just a bunch of clients and relays, and using standard public/private key infrastructure (I know, I know, but keep reading), you can just publish and the content travels across relays to various clients.
It’s elegantly simple. And… it works. Mostly. It’s heavily dependent on what clients you’re using, but as noted above, it’s so easy to build a client that lots of people are, and some of them are proving to be quite featureful.
Of course, the differences in clients can also be confusing. Some let you delete, some don’t. Some have “likes,” some don’t. Some let you upload images, some don’t. Basically the simplicity of the protocol means that all of the features have to come at the client level, which can be a blessing and a curse. For a few months I was using one client, which looked nice, but really didn’t have many features, and it was confusing. Then I started playing around with some others, and began to realize why some people really like it.
And, there’s lots more development happening these days on it, including from some very thoughtful people who have been thinking about how to build a good decentralized social media app.
For most (especially non-techie) users, it’s probably not worth figuring out nostr yet. But the simplicity of the protocol, combined with the pretty active development going on, and some fun features I’ve seen on some of the clients, makes me think there could be something pretty cool there that shows up in the future.
This is the protocol that Jack Dorsey seems most interested in these days, and he spends a lot of time there (and, also just convinced his parents to join nostr as well), so even as he helped kick off Bluesky (and Twitter!) if you’re following where his interest lies, it’s clearly on nostr right now.
All three of these are fascinating to watch, as we’re seeing real efforts to build a decentralized/distributed social media ecosystem. I don’t think one of them has to “win.” I’d like to see all three (and others?) thrive, and they can learn from each other. Eventually we’ll see bridges linking them all together as well (some early versions of that have shown up with things like nostr-ActivityPub bridges, and there will be more).
There are, of course, a bunch of other attempts at such systems out there as well, with a bunch taking the “web3” approach with cryptocurrency and tokens being a big part of the deal. I’ve yet to see one that’s been all that interesting to me. There was Secure Scuttlebutt, which always was interesting, but also had some serious limitations, and it seems that nostr (which was partially inspired by Scuttlebutt) may just be a better overall approach to such a system.
I’m not big on prognostication as to where all this goes, but at a first pass, Bluesky seems to have the elements to become the most “mainstream” of these options, and the team does seem focused on that as a goal. Mastodon has been going through some changes, and I think that enough folks involved there realize that some of the earlier decisions may have turned out to have been wrong, and more hostile to new users, but I think they’re trying to fix some of that, not by copying what other services are doing, but implementing them in a more Mastodon-like way. Nostr is the most “out there” approach, but the enthusiasm and development still has me excited to see what unique solutions come out of it.
Centralized offerings: As I said, I’m a lot less interested in putting in the effort for another centralized offering. There are now a bunch, but they… all have real limitations. Post.news got lots of attention early, but… just feels too focused on news content to actually be all that useful. T2, by some former Twitter employees, is nice, and works, and looks like Twitter, but… is just another centralized clone. If I were them, I’d be looking to make use of Bluesky’s AT Protocol as soon as they can. People keep talking about Spoutible, but… the company has made some very odd early choices that make me not trust them to handle a social media system. Substack has its “Notes” feature of course, but I’ve written about that already.
Just sticking with Twitter: This has to be at least mentioned, as many people remain on Twitter, and certain communities don’t seem motivated enough to move. It still remains possible that somehow Twitter will stop making ridiculous decisions. But, the last six months is… well… not encouraging. Still, inertia is a powerful force and it can win.
Either way, this is a rapidly evolving space, and I’ve been really fascinated and encouraged by what’s happened over the past six months on Mastodon/ActivityPub, Bluesky, and even nostr. For all the talk of a lack of competition, we’re now seeing a ton, and competition leads to experimentation, and experimentation leads to innovation.
When we hit the one year mark of Musk’s Twitter takeover, I imagine the world of decentralized social media will have continued to evolve and improve, and I’m really excited to see where it all goes. For folks who are still on Twitter, you obviously can do what you want, but I will say that none of these services are that complicated, and it’s worth exploring the world out there. You just might find you like something.
Filed Under: activity pub, at protocol, decentralized social media, mastodon, nostr, protocols, protocols not platforms
Companies: bluesky, twitter
Source link
