Lots of exciting #decentralization protocols and technology out there. Some are not ready for usage, others are not following the paradigm I prefer, I love that we're spoiled for choice.
IMO I still love #SecureScuttlebutt, for me it is still the best offline-first local-first gossip protocol out there. Yes, it has dangerous corners and design issues, but it works and I can build apps with it for my friends.
@cy
> What was that non-blockchain network... Briar I think?
Briar is a neat experiment, but they've never shipped apps for anything but Android. The problem with depending on one proprietary OS ought to be obvious, Goggle's recent decision to start farming Android app devs is a good example;
So until it's cross-platform, Briar is a fun toy, but not suitable for production use.
@cy
> Dunno what one would be good though
Depends on your use case/ threat model. Ask yourself questions like; who do I want to communicate with and why? Are you looking for software for an existing group/ network of people who can make and action decisions about where to communicate? Are you wanting to adopt an app to make new contacts among its current network? How sensitive are the communications? Etc, etc.
@soapdog
> the fediverse is indeed cool, but it is not the p2p I aim for
Pure P2P networks have been the holy grail of every new generation of cypherpunks since the 90s. They've never worked out. Everything that's turned out to be practical for use beyond dogfooding has some kind of supernode, and that's not even a bad thing;
https://bridgeseat.substack.com/p/in-defence-of-servers
> It is very costly to run an instance in terms of bandwidth
If you use Mastodon, sure. There are much more efficient fediverse servers.
Just read the thread and there are many good points in it. As I said before, I'm not here to attempt to convince any of yous or have a flamewar.
I disagree with many of the view points expressed in the thread around terming SSB as dead or extinct and claiming it is only valid if old time developers are still present. Don't forget, I wad there from the start as well. Find it mildly annoying to see it called hobby even if it was not intentional.
SSB worked back then and still works now. It has more features now than it had back then.
I also think that people in this thread are pushing requirements and constraints in SSB that were NEVER a part of SSB protocol.
SSB is a network for connecting friends. It follows the same semantics of real life social interactions and groups.
It has never been a privacy oriented protocol for adversarial situations.
SSB p2p connections have secret handshake but SSB is not private or deniable.
Messages are a chain because it is important for the designers that they can attest that a given message was posted by a given identity.
You don't want such chain or need more privacy, there are other protocols.
This thread feels like a king of the hill where people are throwing unrelated protocols against each other trying to see who comes up on top regardless of that protocol objective.
Federation is a whole different ecosystem. ActivityPub servers are servers mostly on ther web and that comes with other needs.
SSB can work without the internet or the web. Tiny SSB can work over BT and LoRa.
This is not a competition, these are the decisions made by the protocol designers and they were made for a reason even if the reason was "yolo this is cool".
Certain things only become apparent in hindsight though, and that's life.
(1/3)
@soapdog
> This is not a competition
If you're talking about protocols as a fun toy to tinker with, it's not. Let a thousands flowers bloom.
But when it comes to getting broad adoption it is, because of a thing called network effects. A protocol that only one person uses might be technically brilliant, but it's also useless in social software until they can convince others to use it. The same is true to a lesser extent, of protocols used only by a handful of hobbyists.
(2/3)
To be socially useful beyond that, a social network protocol needs broad adoption. And people can only manage so many social apps on our devices, and only so many accounts/ IDs.
So fragmenting the potential audience for a decentralised social protocol, across many incompatible and non-interoperable protocols, runs directly against the goal of achieving a critical mass of network effects. Sufficient to replace the platforms that are wrecking democratic discourse and thus our environment.
(3/3)
At this level of concern, social protocol choice is not just a matter of personal style. It's not an exaggeration to say it's an existential issue for humanity.
So yes, it *is* a competition, and the stakes are *very* high. In 2026, Presenting hobby software based on obsolete protocols as being interchangeable with social software based on robust and evolving technical standards is irresponsible, at best.
yeah, all roads lead to dat.
its the oldest and also the best system we have.
...but just like bluesky, ppl first need to go through all the distractions before they finally ran into all the problems that make them want true peer to peer and then they try all the p2p systems until they figure out the only one with universality and performance is dat.
..its fine - i can wait. we are early in the adoption curve 🤷♀️
@cy @serapath @strypey you are doing nothing wrong, this is a limitation.
It uses a lot of disk space so it can be offline first. You have all the data. It shouldn't become super slow forever. On the new version, I added a better search engine but it takes a while until it indexes everything and also it takes longer to load on launch, but once it is running it is very good.
You can't delete cause you can't modify the signature chain. That is the secure part of secure scuttlebutt.
(1/2)
@cy @soapdog @strypey
//A.js
swarm = new Hyperswarm()
store = new Corestore()
feed = store.get({name:'ssb-feed'})
console.log(feed.key)
swarm.join(feed.discoverKey)
swarm.on('connection', socket => feed.replicate(socket))
feed.append('hello')
feed.append('world')
//B.js
swarm = new Hyperswarm()
store = new Corestore()
const [,,key] = process.args
feed = store.get({key})
swarm.join(feed.discoverKey)
swarm.on('connection', socket => {
feed.replicate(socket)
console.log(feed.get(1)) // world
})
@soapdog
> images and other attachments ... can be deleted
Good to know.
> I will add a blob manager in the next version
That sounds useful : ) Even more useful if it has the ability to set data retention policies, eg delete all blobs older than X, or not viewed for X days. As well as a sane default, eg delete all blobs that haven't been viewed for 1 month. What counts as "sane" here is a balance between saving space and keeping the app convenient to use.
i guess feature and UI wise, it would be possible to make the same app on top of dat, but the p2p stack as such is incompatible and i dont have a good idea how to make them interoperable or if that wpuld even be useful... i just wish ssb and dat could somehow join forces.
i perceived them as "cousin projects" since forever, because both communities had so much overlap 🙂
@serapath
> i dont have a good idea how to make them interoperable
I presume this is impossible, because in protocols like SSB, the network layer is inseparably welded to the data layer, which is inseparably welded to the interface layer. All of which are fundamental design mistakes, which Dominic probably wouldn't have made if he'd followed the Kalashnikov Principle before he started hacking. He certainly wouldn't make the same mistakes if he was designing a protocol now.
Strypey,
Please stop with FUD.
> the network layer is inseparably welded to the data layer,
No it is not. The data is saved in flumedb with attachments, it is just an append-only log. Some clients don't use it at all and have other solutions.
> which is inseparably welded to the interface layer.
Different clients have different interface layers. There are git clients, npm clients, dropbox-like clients, social networks, genealogy managers all built on top of SSB.
yes. dominic tarr is actually a brilliant developer and all about modularity. in that sense ssb is the only ecosystem that follows this approach as radically as dat does - hence xommunity overlap and i consider them "spiritual cousins" 😁
they both use libsodium as well.
they both are implemented in javascript too.
i wish there was a way to join the ecosystems it would be powerful. this wpuld mean to somehow use adapters between pull-streams and streamx and more
@cy
you can delete branches from the tree and then nobody can get that data anymore. if you just evict it from your cache, you can re-download it from the p2p network given somebody still seeds the data.
you can also build applications where you rewrite history (feed.truncate()) ...that will increase the hypercores `.fork` counter and its detectable. use cases are when you use `autobase` to have multi writer hypercores
@cy the latest signed root verifies all data of the log, because after appending, the new signed merkle root still includes the entire older merkle tree too.
it is immutable, because change can be detected. it would change the merkle tree hashes.
@serapath
> dat has nothing to do wih cryptocurrencies or bitcoin
Something can be a blockchain without having anything to do with those, eg;
https://jami.net/the-jami-blockchain-switches-from-proof-of-work-to-proof-of-authority/
Also, as I've said before SSB is *not* a blockchain. If it was, it wouldn't be possible to delete old media ("blobs" as @soapdog called them). They would all be immutable parts of the chain, as they are in social apps that do use blockchains;
i am not sure everyone would agree with your definition of blockchain.
you are saying "git" is a blockchain.
i'm not saying its not a possible (very literal) definition, but imho most ppl think of "global consensus" mechanism where everyone can contribute blockchain inputs permisionlessly based on the rules.
...that definition makes neither dat nor ssb a blockchain
yeah, there is no byzantines general problem to solve in dat, because dat is not about consensus of global state - imho thats why its not a boockchain.
git, ssb, dat
...chain blocks ... git and ssb by literally including hashes in blocks to point to previous blocks. dat doesnt, so its even less blockchain. dat has merkle tree metadata to order log entries, but thats not part of the independent log entries as such.
nine of them solve global state consensus or byzantine general issues
@cy
> Byzantine Consensus thing is also terrible. In addition to blockchains
AFAIK blockchains were defined by the BitCoin whitepaper (which I read not long after it was published). The consensus mechanism is part of the definition. If an append-only log doesn't do consensus, it isn't a blockchain. Git preceded blockchains. SSB is not one. Neither is HoloChain.
You seem to think have a much more expansive definition, and I'm really curious to know where you're basing that on.
@cy
yes, given root hash always leads log version.
deleting post33 means, if you try to download it you'll wait forever or timeout, because its not available. there is no new hash.
if you actually `fork` and override post33, the `feed.fork` counter increases and yes, you trying to download 33 will always know it was changed, because hash doesnt check out.
for html pages i'd use hyperdrive instead of hypercore directly. ...a hyperdrive base web makes internet archive trivial 🙂
its not about nothing has been deleted, its just if somebody has the data, you can verify it is part of the log (=hypercore) ...but doesnt mean the data has to be there `feed.clear(...)` drops data
@soapdog
> claiming it is only valid if old time developers are still present
That's a strawman. My argument is that vast majority of former devs departed for *good reasons*, due to inherent limitations of the protocol that break UX for most use cases, and that they were unable to hack around despite trying their darnedest.
> I was there from the start as well
That's important context, thanks for clarifying. Good on you for sticking with it, I guess.
given dat works very similar to ssb, but without the shortcomings - has sparse verifiable replication and also supports seed phrase based multi-device identity - wouldnt it be cool and/or easier to hack together a UI a la patchwork and migrate?
it means also big files can now be shared trivially using hyperdrive 🙂
if there was interest, i'd be very happy to help
SoapDog
Cy
Strypey
𝓼𝓮𝓻𝓪𝓹𝓪𝓽𝓱【ツ】☮(📍🇨🇦)