I don't see the appeal; it sounds like it would devolve into white noise
*: for the sake of argument
ironically when I used Mastodon, while dealing with these issues, I was unable to filter out other languages. So in addition to extremely questionable content, a lot of it was simply in another language.
ActivityPub is a really half baked protocol and the sooner we realize that and move on from it the better. Personally, I didn't feel that defederation was an adequate defense against those MissKey instances and I decided running an instance is a very big liability.
Like I get that moving instances or between applications isn’t really possible on AP and there is concerns with moderation and so on but it’s been the best internet experience I’ve had. It’s a bubble but I easily just come here or to 4chan or reddit to see outside that bubble.
Defederation is not a huge issue if you assume and embrace a segregationist view and cut off likely major fractions of the organically formed Fediverse out of itself. After all it's porn and scripts you don't even recognize, what's the point in having them? My insistence is, that's a fresh dead canary in cage.
I don’t understand how if I host my own AP client on my own hardware and choose only to federate and subscribe with a small subset of sites and people who post using AP that this is a bad thing. I can use other websites like Hacker News to see other opinions and views.
Each instance chooses to adopt defederating lists.
If you don’t like that make your own instance.
Eg if I run a small server I have a difficult time getting my updates federated quickly because other servers have a lot of fire hoses to manage. You end up low on the priority list and less likely to be seen. In my experience from last I tried, at least.
It’s always at the instance level. They own the machine.
The difference between federated networks and decentralized networks is that the main control is with node operators vs cryptographic key holders.
You’re looking for a decentralized solution, not a federated one.
The point is that there shouldn't be any "the" machine for users to get locked into as a chokepoint. If you want to block someone, you block them, or delegate it to someone in a way that you can later change at no switching cost to yourself.
> The difference between federated networks and decentralized networks is that the main control is with node operators vs cryptographic key holders.
Fully decentralized networks have to solve a difficult technical problem: If your device is offline, who is hosting your stuff? How do you make it fast and reliable?
Federation solves that by hosting your stuff on an always-on server somewhere, which you get to choose and should be able to trivially switch at any time without affecting your social graph or account name or who is blocked by anybody in any way. Instead of your stuff being hosted nowhere, each person gets to pick, which can and should be independent of any moderation or other considerations. The benefit, and goal, of federation here should be to make the hosting node a fungible commodity.
You can also federate moderation by, for example, choosing a moderator who publishes a block list that you can subscribe to.
But these two things should not be linked together. Doing so is a mistake. As many things of this nature should be made separate as possible and with the lowest achievable switching costs, to inhibit forces that tend toward market concentration.
Federation works when there are thousands of federated instances that integrate seamlessly with one another, not when there are four that are significantly isolated from one another and you need state-level resources to spin up a fifth.
Why would you force the provider to support objectionable (for them) content? It makes sense for the instance to be aligned with its users on moderation rules.
> The benefit, and goal, of federation here should be to make the hosting node a fungible commodity.
Communities aren't fungible! And your insistence on having federation completely seamless will result in "what's the point anyway, let's centralize it, more efficient"
This is the correct answer though, as much as we don't like it "users" as a whole do not care about privacy or centralization
You’re, again, looking for a decentralized system.
> Federation solves that by hosting your stuff on an always-on server somewhere, which you get to choose and should be able to trivially switch at any time without affecting your social graph or account name or who is blocked by anybody in any way
Yes, the ability to change home servers is missing from mastodon.
But even if they had such a feature, the content you see and likely your ability to change servers would be controlled by your instance owner because they literally own the machine your data lives on and which serves you content.
This is the defining quality of a federated network vs a decentralized one.
> Federation works when there are thousands of federated instances that integrate seamlessly with one another
I don’t think this is a useful definition as it also fits decentralized systems.
Federated networks are networks where independent instances of compatible software are able to exchange information without being owned by a single entity (think email, mastodon, lemmy, etc)
User decision making is still very much an entrenched thing.
Most decision making on blocks in small-to-medium instances is democratic, in my experience, with users voting on them together. Also, as pointed out there's the obvious "vote with your feet" of switching to a different instance if you don't agree with its policies and/or how other instances don't agree with its policies.
There's definitely a risk of large instances trying to strong arm smaller instances with blocks and/or threats of blocks. But so far it's more a philosophical risk than a real risk from what I've seen. At least in my parts of the Fediverse small instances are "the norm" and it's is more likely the blocks are against the larger instances because with size they are more likely to allow spam registrations, they are more likely to have users that don't respect cultural norms like CWs or Alt text/Image Descriptions and don't feel a need to respect them because their mods won't enforce them, or yes they set up an ideologue as a mod/admin and shift to a gross direction. As a "telegraph network between a lot of small villages that mostly ignore the big cities", ActivityPub can be rather nice.
Let's start with "no one has heard of mastodon" because no one is spending money marketing it to joe public. Sure it'll spread by word of mouth, but honestly that's not terribly compelling (because most of the current mouths are, um, the same people ranting about the incumbents. )
I don't disagree that the same process leads to the same outcome. I personally don't think bluesky will ultimately be any different to the rest.
But the no-money approach of mastodon means its a very very slow burn, which will take a decade or more to succeed, and even then may not be what we expect when a billion people show up.
Which is a good thing from the spec point of view but maybe bad from a user adoption point of view. Even for the later you'd be wrong, as Threads is supposed to be an ActivityPub application.
While I think Mastodon's irrelevance is deserved, let's also be fair to the "incumbents": Facebook, Mysterious Twitter X, Reddit, et al. gained and maintain their critical mass from word of mouth.
Many other would-be upstarts in history also usurped thrones by word of mouth, foremost example being Firefox against Internet Explorer.
Mastodon's problem with becoming relevant (and also BlueSky's problem with upending Mysterious Twitter X) is far more fundamental than lack of awareness.
Anyway I have checked several social medias today (HN included) and everywhere except one place there was too much noise about TikTok - only place that my feed was without it was Mastodon - it is quite slow there but i consder it to be good thing. However I think that there is no good social media - Mastodon included and my days would be improved without any of them. RSS feeds feels like more then enough. Discussion seems to be mostly point-less. Maybe even this one, but those enhanced with algorithmic engagement and endless scroll are net-negative.
It's for the same reason I don't recommend Signal to anyone in my circle. I don't want to trade one walled garden for another (Signal still refuses third-party clients for example). I use Matrix which is truly open.
Stay with me...
So: federation is very cool in principle, and it's extremely cool in that it in theory means we don't have Just One Batshit Master of all our content... but in the way it's being done with Masto, it IMHO makes for a weak proposition.
Why?
Mainly because people (normal people, not us lot) don't understand or care what "federation" is. They expect (because it's been the norm for every other service), a SINGLE place where they can go to find their mates and celebs and politicians.
What they instead get is a thing where:
1) They can't search a global place and find all those people they want to find (why the Mastodon team don't have this as the #1 thing they are working on, who knows)
2) They find someone on one "instance" (not understanding what an "instance" is) and then can't (easily) follow them from their own instance without having to think about namespaces and all that
3) They naturally gravitate towards the biggest one - probably mastodon.social - and then we're right back at the beginning, with everyone on a single instance, beholden to the possibly loony who might shut it down / monetise it / etc
Moving between instances is much harder than it is claimed to be (you lose all sorts of stuff like your history, or at least you did when I tried it).
Federation also brings all manner of hard things to those trying to run an instance - I tried, as "medium level nerd" and ended up walking away from the complexity of just not understanding why some content didn't seem to be getting from my instance to others, etc etc.
If I was the Mastodon team, I'd be focusing all my attention on global search, and on never using the word "federated" in any of their marketing ever again. It might well be the coolest thing, but it's a non-marketable thing.
Of course all this is predicated on "a good outcome" being "everyone on Mastodon" and I do appreciate those who don't want that. It's definitely the case that less people tends to make for better online social spaces, and maybe small niche groups leads to better things all round.
Amen and hallelujah! This is why I gave up on Mastodon. I read that not allowing full text search across instances was actually a design decision in order to discourage brigading. But, more crucially it undermines discovery.
If someone want full-text search across the entire fediverse, who is going to pay for it? Bluesky has a business model that makes that attractive to them as a value add to pay for. ActivityPub has instances running on RPis in a shoebox. (This is a great thing, it's part of where ActivityPub federation works well.)
Of course, that's also before you get into the sociopolitics that many instances don't want full text search and are concerned about brigading and pulling old microblog posts out of context for nefarious reasons; some of which is why some instances left services like Twitter and have no interest in services like Bluesky.
The people you describe wouldn’t use Mastodon in a web browser and this is a solved problem on the apps.
The only person I know who regularly uses Twitter says she has never visited the site in a browser and is quite sure that everyone in her circle uses the app. But that's just anecdote
You mean, it lacks centralization?
It’s called “marketing” and “paid-for articles”
Mastodon has many MANY MANY issues.
The first is that instance operators regularly abuse their users as hostages in personal petty fights. I don't care too much about drama, but there has been a lot of it regarding Israel/Palestine or Ukraine/Russia and instances defederating from each other as a result of said drama.
The second one is instances can go down for whatever reason - the admins just being unable/unwilling to cope with moderation, running out of money, getting into trouble with the legal system, ... - and users can't move their post, DM and media history to another instance.
And the third one is it takes them forever to ship updates. Bluesky is so much faster moving when it comes to implementing new features, but Mastodon ships even slower than Twitter which is an "achievement" in itself.
Mastodon is a non-profit with a handfull of engineers. How can you compare their resources to something like Bluesky or even Twitter, that has thousands of engineers, is beyond me.
Put another way, they were structured in a way that doesn't allow them to compete.
It’s understandable in ancient protocols like email where storage was at such a premium that universal replication was out and cryptography was primitive. It’s not forgivable today.
I am ignorant of AT — does it have this problem? I know that Nostr doesn’t and it’s always struck me as technically superior. Problem is there is nothing on there but Bitcoiners and all the topics adjacent to that subculture.
ATproto allows data to be hosted off-site but account lookup goes through the Bluesky owned centralized infra. Just my hunch but maybe its "federation" aims is just a sugarcoated version of "it's a carbon copy of late 2010s Twitter microservices, but we're building it on public IP with intentionally minimal authentication".
There is zero mechanism for the funding or promotion of anything that's not a lock-in play or a data play (or both).
I didn't realize Nostr had such scaling problems but I think it makes sense now that I consider how it's a client-server system with a network of servers. Making all traffic go through it that way is going to cause scaling issues or require scale-up of infrastructure that will break decentralization. AFAIK they intentionally passed on P2P because "it's hard," which is true, but it's also how you don't pay for bandwidth.
IPv6 has enough penetration now that you could probably get away with easy mode P2P where IPv6 is required. You still have to hole punch there but it works about 100% of the time because no port remapping. (Even the few areas where V6 NAT is deployed, it's usually 1:1 NAT without port remap.) If you don't have V6 you get a slower experience because you have to relay.
There's much more interesting innovation waiting for adoption on that lower part of the internet stack I suppose. As someone for whom that's a too specialist area I would love to have more overview of what are the promising technologies and upcoming standards to place early bets on.
The other day by accident I found out about Named Data Networking networking [0] via a paper [1] "Exploring the Design of Collaborative Applications via the Lens of NDN Workspace", and saw that NDN still sees active development after many years, so I wondered about the extent the technology still is considered promising for mass adoption today.
There is also DID:Web. This one has the downside that you need to continue to control the domain name in question indefinitely, and it can be argued that the domain name system is still a form of centralization. Like PLC users can theoretically change handles to another domain name with this scheme (but must contrinue to control the original domain name). Users can freely move to another personal data server.
AT Protocol can add new DID schemes in the future to avoid these downsides, with the caveat that users cannot change from DID type to a different one seamlessly, and adding new DID types may potentially require updates by multiple other parts of the ecosystem.
I’ll take this all seriously when people admit what the real drivers are. Admit why people are actually looking for an X alternative.
It’s a mix of ideologues, performative outrage, foot stomping, and wanting the 2020 status quo.
That has not been my experience at all.
I tried X again at some later point, I think even twice.
A fresh account bombards you with far-right propaganda and outright lies. At the time at least this included hateful and incoherent rambling by Musk himself, which you couldn't unfollow (or ignore, it simply didn't work).
Call it ideological, but I'm not going to spend my free time with this "content", especially when the platform clearly disregards repeated signals that I don't want to read hateful ideological propaganda comment no 73646445 by some alt right shill.
I'm all for open discourse and dealing with other peoples differing opinions.
But at this point, the "ideology" accusation by the far right against any other opinion is nothing but laughable. Well, it would be, if people didn't still pay it credibility.
And no, I'm not a "leftist", "transgender activist", or whatever group gets to be public enemy of the day for these people.
I did notice that Bluesky seems to have more politically left people.
This kind of content is not very interesting to me, we already have Reddit.
But for Bluesky, the platform bubble phenomenon didn't seem that strong to me.
Appreciate the new features to build my own (interest) bubble.
For political content, I think actual journalism and real-life discourse are most valuable.
1) terrible experience: a lot of people just don’t have fun on twitter, it’s just an awful experience. why spend your free funtime in a place that you just don’t enjoy? we don’t go to restaurants that we hate, why on earth would we go to a website that we don’t enjoy?
2) too crowded. take a music concert for example, a lot of people absolutely prefer a music venue with 2,000 people over a concert where there are 100,000 people.
Over the years I've come to the conclusion that there are people who say they are in favour of diversity but underneath only want their kind of diversity, not genuine diversity.
Diversity of opinion would definitely be a feature, not a bug.
Cutting that off and "just walk into the brightest place" couldn't happen at the same time, so the core devs and ops switched to the "Mastodon's strength is in small servers, pick any of the right one" narrative.
[0] https://en.wikipedia.org/wiki/Child_pornography_laws_in_the_...
I thought they were super strict there (after all they even require that blocky censor thing on the functional parts in normal adult pornos making it basically useless).
IMO, human aging vary too much to solve this "looks kids to me" problem. Age by appearance is clearly regional; there are plenty reports of East Asians abroad barred entries at morning and beer at evening. So any substantial East Asian content feed is destined to include tons of "kids/age unreliable" content, unfortunate opposite being "middle aged/unknown". Other factors as cultural norms and effect of modes of consumption compounds on it.
Long term, the solution to this has to be algorithmic: the mix of content must be artificially manipulated so to suit the need of consumers. The fraction of first(primary) language English speaker is below 9%[2], less than third of Mandarin Chinese at 21% or about same as Spanish at 11%. So a viewer application for a truly global feed of all contents will have to mark and remove ~90% of content for it to be palatable to an English speaker. "Endless Japanese CP" problem is just but an early sampler.
1: https://en.wikipedia.org/wiki/List_of_languages_by_total_num...
Though I have to say I sometimes wonder how sites like Onlyfans do this with male actors. Often the only thing you see is their dick and I can't remember having to submit a dickpic for my passport :D So how do they know a submitted passport matches the dick in question?
But really I think it's better to know for sure than to guess.
Sites like Onlyfans, or any website for that matter, just delete contents based on appearances and user metrics. Otherwise escalations won't stop; users would leave, protest will occur, new laws will be created, state committees will be formed and blacklists are made, and credit card processing suspiciously starts failing. Recordkeeping is for honest people to be honest if they want to be. Some numbers on a driver's license do not stop the horde from banging the walls.
So a truly global feed has to hide inconvenient contents from users until users are ready for it. Ideally user editable. Which is what everyone is fat-finger doing at server side anyway.
If you wanna delve into the details of why people so often avoid the platforms that FOSS enthusiasts tend to recommend, that's an interesting question, but we gotta be clear here, we already knows who's successful.and who's not.
No they don't. If this was true, my wife's family would be on Telegram or Discord, haha. We actually did go into Telegram briefly, but they all dropped out. What got them to switch from SMS to WhatsApp was her parents temporarily moving to Austria.
No True Scotsman is Exhibit A for why technology-first platforms tend to fail.
I think part of the issue is that you can’t do full text search across instances. You can only search on hashtags, and people don’t always use hashtags.
You could tell me that, but don't. The answers are very much besides the point if you can't explain it and the advantages of the arrangement in five minutes to somebody on the opposite end of the normal distribution of technical understanding to your average HN account. Network effects dominate most social network pros and cons, so if you can only educate five percent of the general population there isn't much point.
The same can be said about the atproto as different apps are being developed for specific users like picture video only feeds and 3p clients for bsky.
> And unlike Bluesky, Mastodon has put federation into action; as an anecdote, even for posts with lots of replies, I've rarely seen more than two people from the same server comment on a given post. The diversity is astounding. Mastodon is already everything everyone wants from Bluesky in this regard.
I'm not sure I get your point. I run my own bsky PDS, and federation is working fine for me and my users. I see plenty of posts from others.
> Is it because choosing a server is to hard or stressful?
Are you asking if perception of difficulty is really impactful to people's choices?
W/ bsky the hardest choice was the domain name purchase. And I didn't have to host my own server to do that either.
I think one of the major parts that resonates with people is the focus on data portability. AFAIK, ActivityPub doesn't help with wanting to move your data somewhere, without having to manually perform a async migration.
> which is promising federation but has yet to deliver
They seem to surely but slowly make headway on the federation stuff, there are more and more successful experiments of people hooking into the network.
> it is backed by a for-profit company that has all the incentive to enshittify much like Facebook and Twitter have.
That is true, hopefully the network will be resilient enough by that point, because I'm wary of that too. I guess time will tell how that goes.
On a more optimistic note, there seems to be more and more efforts of trying to fund other non-Bluesky projects, including for core infrastructure. Example, not an endorsement: https://bsky.app/profile/freeourfeeds.com
Compare this to Mastodon (which unlike Bluesky) is just one service in a sea of many others using ActivityPub (Pixelfed, PeerTube, etc) which overall makes for a much more vibrant and promising platform.
And unlike Bluesky, Mastodon has put federation into action; as an anecdote, even for posts with lots of replies, I've rarely seen more than two people from the same server comment on a given post. The diversity is astounding. Mastodon is already everything everyone wants from Bluesky in this regard.
To me, it just looks like everyone is getting set up again to shoot themselves in the foot much like what happened with Twitter, and I don't understand why? Is it because choosing a server is to hard or stressful?