Pubky is legit

It's on another level compared to Nostr / Primal, IMO

Reply to this note

Please Login to reply.

Discussion

Were you able to test it at all? https://github.com/pubky/pubky-app says it's depreciated, linking to a dead repo. and https://pubky.app/ wont' let you join without an invite code, hardly premissionless...

Yeah, I've been trying it out the last few days. I think only the web app is currently live.

It's very well done, especially for a beta.

John previously said the reason for invite only was to prevent spam, bots, etc. Seems reasonable to me for a beta app that's in the testing/feedback stage.

I assume that once the apps and homeservers are ready for primetime, it will be fully permissionless and open to anyone.

It's currently a MVP - an excellent one, IMO. Some really great features - edit/delete post, cool tag system, audio clips, excellent embeds from X, YouTube etc. (see image below). Even the X videos play within the Pubky app.

I think Nostr is great, and has a great community, but Pubky appears to a superior platform from a technical / architectural perspective.

Request a code and check it out.

The only code that isn't already public is App 2.0, which we will open next month... it just isn't really usable yet!

It’s just not really saying permissionless protocol when all the docs lead you to apps that require invite codes. Generally I’ve found if folks are working in a closed by default way they tend to keep that mindset.

If your goal is prejudice, I can't stop that. I don't know why you think you have enough info about our work or our behavior, or why you're making predictions about us, but I understand this can be a tribal environment.

If your goal is to learn about our work, and how it might be a step forward, then we are here to answer any questions, and we have provided as much documentation as we can so far.

Everything is free, open source. We started public dev calls, we're hiring a developer advocate, etc, etc.

I think judging us ignorantly would be a mistake, but it's your time!

I’ve heard good stuff about pubky but was disappointed that it didn’t feel like it lived up to the promised values. I’ll take more time to look at it.

Why do you compare Pubky as a platform with Nostr (which is a protocol)?

I think of the whole ecosystem as "Pubky" in the same way I think of the Nostr ecosystem as "Nostr."

Perhaps it would be more semantically correct to compare the Pubky app to the Primal app. I use the Primal app (which is excellent), but I think of it as using "Nostr."

Pubky also has an underlying protocol - pubky core. There's other components as well. It seems to have been really well thought out from an architectural perspective.

My general understanding is that it aims to solve many of the critical limitations of Nostr and other decentralized social networks/platforms/protocols.

The UI/UX of the MVP is great. One thing that really stands out is how fast it is, but it's not really a fair comparison because it's not yet decentralized. Also, the embeds are impressive and it handles a lot of different media types very well.

I have zero tribalism instinct when it comes to various decentralized technologies. For me, the only competition that matters is between centralized and decentralized.

This allows me to be objective, and from what I've seen so far, Pubky (the protocol / ecosystem) is objectively superior to Nostr (from a technical / architecture perspective).

It doesn't have much of a community yet, as it's just a small group of beta testers at the moment. It will be interesting to see how the community develops after the actual launch.

[For the record, I have no affiliation with Pubky, Synonym, etc. - I'm just a rando lurker rooting for decentralization to win]

When I follow your words, they are following a similar path as BlueSky with ATproto. That’s a very centralized path.

From a technical architectural point of view you can’t create a pragmatic judgement about Pubky, cause it’s a closed / permissioned / controlled environment. Only when it’s open and permissionless, it can be battle-tested.

There's a lot of information available regarding the architecture:

https://medium.com/pubky/pubky-the-next-web-3287b35408f1

https://pubky.org/

It appears to have a credible decentralized design.

However, your point about ATproto/Bluesky is a good one. That was supposed to be decentralized but, in practice, it's extremely centralized and even more censorious than Twitter/X.

ATproto is essentially just ActivityPub with DiDs. Everyone ended up on a single instance, Bluesky. Just like in the Fediverse, the instance operator calls the shots.

Pubky has a radically different architecture that emphasizes mass distribution of data, control, and discovery across a large number of servers.

If everyone ends up using Synonym's homeserver and nobody spins up their own homeservers, then Pubky might end up looking more like Bluesky than Nostr.

However, in that scenario, it would be the userbase, not the architecture, that failed.

And, that same centralization can happen to Nostr too, and there is already evidence.

I could easily make an argument how Pubky is the only design positioned to actually allow for both decentralization AND scale in a self-sovereign-compatible way.

Why do you think these things, they are not true.

Pubky is entirely open source already, and people are already building their own apps.

How can you comment on the architecture if you do not know it?

The invite codes for the beta are just because our server isn't ready to handle abuse yet and there needs to be a first server right?

Please get your information from better people, or get it yourself, it seems like you're learning through podcaster vibes or something?

So can we run our own homeservers and have them fully work as part of the larger network? Can I easily point the clients at my own home server?

This aren’t gotcha questions. I’m genuinely interested. Pubky is interesting.

This is a chaptered version of our most recent public call, where we explain the topography of the system, PKDNS details, community updates, and our intended roadmap for the rest of this year.

https://www.youtube.com/playlist?list=PLPl4ngNy_cl-LGQ9oPdrNXItjCv42ISLX

This link has only the roadmap, broken down by each repo/sub-project:

https://pubky.app/post/gujx6qd8ksydh1makdphd3bxu351d9b8waqka8hfg6q7hnqkxexo/0033SP9JAKR0G

Our blog has great posts about the censorship resistance of PKDNS/PKARR/Mainline:

https://medium.com/pubky/

To answer your questions:

1. Can we run our own homeservers?

Yes, that is the github.com/pubky/pubky-core repo.

2. ... and have them work as a "part of the larger network"?

I guess this requires nuance in that there is no "larger network" aside from Mainline DHT, used in place of ICANN DNS. Your identity key is used to create your own signed DNS record that tells anyone that knows your key where to find whatever you want them to find. I will assist and reframe your question as...

3. "when will data from my homeserver show up in your app?"

The Pubky.app beta only reads from our single homeserver and indexer right now. Part of coming out of beta (Q4) is updating all aspects of our homeserver, indexer, as well as our own Pubky.app product, to support external homeservers, and to provide basic apps anyone can run to mirror their data. But please note that people can already use all of these things without relying on our infra, if they run it themselves, and some already are, (despite us warning them that it's still a bit early!).

4. Can I easily point the clients at my own home server?

Yes, the pubky-client, sdk, etc, and ALL working aspects of Pubky are already open source and being experimented with by others. Some are using our homeserver for their own app. One is feeding directly from our indexer. Iroh doesn't use any of our app-level stuff, just pkarr and mainline repos. Now... "easy" is of course subjective, and I'd say that we are at the stage where you would need to be particularly motivated to use our code, mostly because we are rewriting a lot of it, and we have a lot of work to do in docs.

Before the end of the year, things will look much better, we're just grinding through the roadmap and trying to scale our team from 20 to 30 people (includes a wallet app team).

nostr:npub13ndpm2hm9hud4azsq5euhf5mv3d05r90wymwxsd7rdn29609hhvqp60svh thanks!

Those last questions do not feel you are inviting me to get my hands dirty. Please keep it gentle.

I'm just a person sharing info, don't expect love from people on the web.

You should embrace abuse and the friction that comes with it. Just a tip, it will prioritize the things from the bottom-up.

I was just looking at this https://www.youtube.com/watch?v=X3ogGRuv7qY

This is just as complex as what I've seen (and learned) from ActivityPub and ATproto.

https://world.hey.com/dhh/merchants-of-complexity-4851301b

You are talking about Pubky.app in the sense of Primal.net. Which is not what he is talking about.

You can right now develop whatever you want on top of the underlying tech over there. You could make your own Pubky.app. Or weave PKARR into your stack like Iroh does. Or do something with PKDNS. Whatever you like. Ask nobody. Just do it. It's all open source.

Whose permission do you need and to do what, besides needing something akin to Primal's permission to use a beta app from Primal? So far nobody in the "Pubky is permissioned" camp has been able to spell this out.

I just onderstood there is a protocol and app with the same name. The app is permissioned, the protocol is not.

What is Iroh?

https://www.iroh.computer/ tech behind delta.chat

Because Pubky is both protocol and an app. Nostr use cases are actually a subset of Pubky's, not the reverse, because Pubky also introduces discovery via PKDNS.

That makes thing even more complicated. Using the same name for the protocol and an app… Why?

I think Pubky has a marketing problem there (yea Nostr has marketing problems too).

The mktg problem you describe appears to simply be that Nostr people are not interested in other technologies, to the point where they will invent reasons to hate them and speak badly of them, while knowing nothing about them at all.

Real users of apps don't care what the protocols are called.

Real users of protocols don't care what the apps are called.

Neither of them genuinely cares if they are the same name.

The app repo is the live beta app, but is undergoing a rewrite, so we warn about deprecation so people don't build on it. The new version will be ready next month.

The invite code is because we are the only ones running a homeserver right now, and because it is a beta.

Most importantly, all of the protocol is already open source, The parts you want to pay special attention to are the PKDNS/PKARR aspect, and how we are applying an open "semantic social graph."

I'm happy to explain more, generally, it's more like Bluesky than Nostr, but distinct in various ways, including the two mentioned here.

Pubky isn't just a social media app, it is part of a larger design to build out a free market society.

The app repo is the live beta app, but is undergoing a rewrite, so we warn about deprecation so people don't build on it. The new version will be ready next month. The invite code is because we are the only ones running a homeserver right now, and because it is a beta. Most importantly, all of the protocol is already open source, The parts you want to pay special attention to are the PKDNS/PKARR aspect, and how we are applying an open "semantic social graph." I'm happy to explain more, generally, it's more like Bluesky than Nostr, but distinct in various ways, including the two mentioned here. Pubky isn't just a social media app, it is part of a larger design to build out a free market society.<|begin▁of▁sentence|>