they kinda already did
through the initial media coverage and the people involved a lot of current still-somewhat X/Twitter user think of it as Twitter 2.0 and plan to go there ones it's more finalized/ready
mainly as far as I can tell there are a lot of people stuck on Twitter which do not like it anymore for various reasons but need something with decent discoverability which mastodon seems to not have so well
so if they manage to deliver in that area I would think they have a "not so small" user base already secured
but we will see
Well, this is a positioning of Bluesky in the market: they position Bluesky as an attractive product for users who are interested and knowledgable about technological topics. I guess if they have an initial user base of this group, and this group loves the product, they will take the next step and analyze how to grow from this to a larger group of (potential) users.
The lesson of the past 15 years is that creating a "global town hall" with hundreds of millions of daily users doesn't guarantee profitability. So why not decentralize and try to go small and local, like the phpBB forums that are still chugging along?
Centralisation is a feature in social networks: being able to find your friends is the very first thing. But don't then pretend you aren't.
Every now and then I post to Bluesky asking: so what's the loadout for Bluesky? How many servers does it take to run this thing? What would it take to run my own?
These questions never quite seem to get answered or documented that I could find.
A "PDS" is a reasonable Ubuntu box. They seem unable to admit what it takes to run the rest, though.
I thought it was already positioned as a Twitter competitor with the aim to make it harder to censor. Its aim is to replace Twitter.
This problem has already been solved by Telegram.
You will never get censorship-free anything if there's money to be made by making it not so. Bluesky, Nostr, all these non-Twitter Twitters will also cater to advertiser needs as soon as VC money enters the picture.
But you're right however this is HN not WSJ.
Bluesky is dead as a product.
Mind you, per capita France and Germany also have way more servers than the USA. Interesting.
That's the main reason I'm not on it personally: It feels lame to just join the biggest instance, but none of the smaller instances really seem to "click", and I'm not really so keen that I want to run my own instance.
Having One Obvious Choice to join to begin with is the best thing federated services can do to encourage wider adoption.
Last I checked, the Western or European parts of Fediverse had since been trying hard to distance, obscure, dilute and fight Japanese anime content, without much success. I suspect that's where your felt barrier comes from.
I think people exaggerate how massive a decision this is. When the news that Twitter might be bought by a naughty melon broke, I revived an old mastodon.online account that I'd created back during a _previous_ Twitter scandal, used that for a bit, then moved to mastodon.ie. Moving is, generally, reasonably easy. Now there are edge-cases; your instance might just _go away_, say. But generally you can't go too far wrong by initially selecting one of the big ones and then moving as and when.
The decision isn't really that massive, a lot of users have alt accounts on different servers and it's not that difficult to migrate between them.
It'll be interesting to (if that ever happens) see whether bluesky, once it actually becomes a decentralized system instead of a closed one, shakes out differently due to protocol differences, or broadly follows the same patterns.
If*
Everything must be copied over to your server first. Let me say that again: You only see data that is stored on your local instance.
So likes, profiles, replies, etc are all very limited, since your server only stores them from people you follow (since you’re the only user).
Not only this, but this works in reverse aswell. Only servers that know of your existence will see your posts. If no one is following you, only people (and people on their server) you directly reply to will see your posts.
As such, being in a self-hosted instance or small community can be extremely isolating and make it hard to really network with other people outside of the circle that is already around you.
So you can just use your own domain for example.
What I don't understand is why the bsky.social service (which holds most user accounts at the moment) doesn't redirect http requests to ones profile. Then you could just say "I'm https://gnod.bsky.social on bluesky". Instead of "I'm https://bsky.app/profile/gnod.bsky.social on bluesky".
Maybe they don't want to make bsky.social very convenient so people have more incentive to use their own domain. Or there is some other thought behind it. I don't know.
A way around this issue would be for users to create a specific subdomain for their Bluesky handle (like @bsky.example.net instead of @example.net) but that defeats the purpose in term of "branding", shortness of the handle, etc.
We should just stick to public keys and forget about the domain name system if we ever want to actually improve the situations instead of just reinventing Twitter and Facebook by another name. It's endlessly frustrating how long public keys have been around and how little use they still get. Finger crossed that Nostr gets some traction, since that so far sounds like the most sensible approach towards a social network that is under the controls of the users themselves and nobody else.
Compared to what? IPv4 addresses are more rare and more expensive, for comparison.
Besides, public keys are great for uniqueness, and that they're relatively cheap, but that also means they're hard to remember (compared to domains) and they're easy for spammers to get.
Basically, neither domains nor keys solve Zooko's Triangle by themselves, and probably the ideal solution uses both and more to enable something that is decentralized, easy to remember and also secure.
or simply gnod.bsky.social as https is the default protocol on browsers.
Mastodon is a bit like email. Not great but it works right now and is reasonably easy to get into and integrate with. The feddiverse is bigger than Mastodon. But it's mainly Mastodon. And Threads soon apparently. Millions of people daily active on it. Not quite critical mass yet but closer than most other things have gotten since Twitter started imploding. Good enough?
IMHO mastodon with signed content and per user key pairs would make a lot of sense and it should not be that hard to add it. Why is this not a thing? Just like using pgp makes a lot of sense for email. Except that never happened either. Already when I started using email in the mid nineties, some people were insisting on using pgp. That's thirty years ago and it was only some people. It remains very uncommon for people to use that. Key management is an obstacle. Techies get it. Everybody else doesn't.
Just to be clear, nostr doesn't directly require the blockchain for anything. It has a nice integration with lightning payments, but that's optional.
> Bluesky is mainly a thing that people talk about rather than join. And it has the Twitter founder backing it
Pretty sure that's not really true anymore. Jack Dorsey has apparently moved on to nostr, interestingly enough.
> IMHO mastodon with signed content and per user key pairs would make a lot of sense and it should not be that hard to add it. Why is this not a thing?
There's some work in this area[0], but I agree it's surprising this wasn't baked into the protocol from the beginning. Same thing with signed requests. Things have de facto evolved in that directly, but we seem stuck at a local optima using a 5yo version of the HTTP signatures spec (see discussion here[1]), which is really confusing for new implementors as well.
Ultimately I think a lot has been learned from implementing ActivityPub in the wild, but I feel it badly needs a version 2.0, and even then I'm not convinced all the most important lessons have been learned yet. For example, it's very concerning to me that Mastodon still doesn't provide an easy way to bring your own domain to any instance.
[0]: https://codeberg.org/fediverse/fep/src/branch/main/fep/8b32/...
[1]: https://socialhub.activitypub.rocks/t/state-of-http-signatur...
For what?
I mean, that's basically the story of the internet as a whole; a teetering tower of "eh, not great, but it'll do for now", since the 80s.
There are active efforts to bring other groups to nostr- check out nos.social for example.
Two of my main grievances with the communities both have to do with identity:
ActivityPub as implemented in the wild doesn't make it easy to migrate your identity between instances. This has caused endless problems and drama with instances shutting down. You also have to use multiple accounts if you want to interact with different formats, ie Mastodon vs Lemmy vs PeerTube. As a user that seems pretty ridiculous. You're losing a huge portion of the advantages of federation. IMO a better model would be for you to create a single account on a universal ActivityPub server that supports all data types. Apps like Mastodon would act as clients, not servers. My reading of the specs makes me think this was the original intention (there is a client to server "C2S" protocol), but then Mastodon got big and no one ended up implementing the C2S protocol, treating Mastodon as the de facto C2S API.
Note there is some work to create these universal servers[0], but I wouldn't hold my breath on Mastodon/Lemmy/etc implementing the C2S protocol.
As for nostr, private key management is a complete non-starter for identity, barring some major new breakthrough. DNS is the pragmatic solution to identity.
(Since then the network attracted a userbase that laughed Jack off the site for vaccine denialism, so lol)
If you want to see my personal views on the matter, you can watch a talk I gave prior to joining the project: https://www.youtube.com/watch?v=ULFj714_Vvo
But, I don't think you guys are right. You can have a distributed system that also has authenticated posts. It's been done before, and it will be done again.
And you can delete things. You just gotta trust nodes to also delete them and they probably won't.
And I'm totally willing to help you guys build it. I know, you don't want me to work for you guys because I work at California-based company at the mall. But I assure you, I can get you guys over this little obstacle you've discovered.
Call me! I'll answer 773-510-8601. Or the usual channels.
I still have to see a distributed social network which works well enough for end users. It's not simple if you're centralising everything, no wonder distributed alternatives are hard to come by.
Currently the BlueSky search engine is pretty bad; it only goes back a few hours. If you don't post for a day, none of your posts will be found. Searching on obscure terms doesn't work very well.
No, you really don't.
If you have 10 million people all spread out across small servers, how do you suggest I find everyone posting a particular hashtag?
Naturally I'm very biased, but I wouldn't say they were all developed merely so companies could have a blockchain strategy. Certainly just keeping up with the Joneses was a motivation for some of the middle/upper executives who sponsored some of the apps built on those platforms, just like it is for AI now, and was for XML back in the day. But there were also actual real business problems worked out in detail, with people who genuinely wanted to solve them and saw a chance to use blockchain hype as a Schelling point to coordinate large numbers of companies to collaborate better. And real apps have been developed and shipped on these, that provide at business value.
It'd probably be interesting to write about some of these systems from a purely technical perspective, as a decentralised database engine, because the underlying protocols and designs are pretty interesting and (at least in Corda's case) quite powerful too.
Do you have a blog/twitter/bluesky (hah)/anything that I could use to get notified if you write anything on this topic?
Last i saw it still relied entirely on a single central relay.
They're really talented folks so if you have any interested in p2p comms, I'd give them a look. https://keet.io
Systems to defeat these exist: ICE/WebRTC, ZeroTier, Tailscale, dozens of other protocols I can’t recall, but they all add overhead and require coordination. There are also many of these systems which means there is no standard. There is no equivalent to TCP/IP that is universally available that works around the ugly hacks we have used to destroy the potential of the Internet.
It would be doable if everything were a desktop or a laptop, but most people use casual apps like Bluesky on phones. Phones have huge limits mostly around battery life. An always on P2P app on a phone will keep radios from sleeping and reduce battery life up to 50%. It’s also a problem for people who have crummy metered mobile plans. Firewalls require keepalive every 15 seconds or so or traversal must be done all over again, which means continuous chatter just to keep peer to peer links available.
If phones could just call each other without executing what amounts to a coordinated attack against middleboxes in the way, low power peer to peer might be feasible. You could do things like time synchronize so peers sync at designated times and the radios sleep the rest of the time. But we can’t have nice things.
All the heterogeneity and complexity in the Bluesky protocol exists to work around this. Without NAT and firewalls all you would need is maybe directory servers to find people and something to do the equivalent of dynamic DNS for when IPs change. That’s it.
You can take your data to another server only if the original server is up. When the admin forgets to pay the bill and ISP nukes the box and the backups admin said they had don't exist and the coadmin turns out to have no access at all, you lose everything. Ask me how I know.
Self hosting in theory should fix this but it doesn't: it's a resource intensive afterthought because the developers only really care about their Big Instance.
Couldn't we make this with existing tech ? Having your own user id as your own domain is easy when everyone is on Bluesky but can already be done today. What's missing ?
A couple of things then happened all at once:
DJB's NaCl became widely available, and, with it, more compact public keys"
NaCl was widely available in 2011. (Maybe as early as 2008 TBH.)
What came in 2013 was that someone took djb's concise presentation of the code and added some cruft to make it palatable to a wider audience.
