Nostr Archives

It's interesting how the creators of the *multi-platform* social client @0fe0b18b…d16bd6ac disappeared from Nostr. The last activity I see from the team is in August.. I am more and more convinced that bridging and connecting different protocols in the social sphere doesn't work very well. Why? Maintenance becomes exponential - Every protocol has its own update cycles and breaking changes. You're not building one client - you're building several. When one protocol changes, everything breaks. Lowest common denominator - To work everywhere, bridges strip out what makes each protocol unique. Nostr's zaps, Mastodon's content warnings, Bluesky's feeds - all lost in translation. Users get a watered-down experience. Identity doesn't translate - Nostr keypairs, ActivityPub domains, and AT Protocol DIDs are fundamentally incompatible. Bridges create confusing mapping layers instead of real unification. Culture clash - Each protocol has its own norms and expectations. Bridges create an awkward middle ground where nobody feels at home.

💬 25 replies

Replies (25)

SondreB74d ago

@Melvin Carvalho has been working on DID support for Nostr, which would make any Nostr key pair pluggable into the DID/VC ecosystem. I just read a blog post regarding the cost of moving social networks, and how Facebook integrated a "backdoor" into MySpace allowing users to migrate more easily back in the day. Mastodon did something similar when people started exit from Twitter, until they stopped it by closing down their APIs. The bridge with Mastodon that @Alex Gleason built works fairly well, it does increase the amount of content on Nostr. You're pointing out a major issue though and there is no good solution to the situation.

000

codearchy74d ago

I might be missing something, but all of these seem like solvable problems. They are mostly data transformation problems, AFAICS. For example, it seems fairly trivial (especially relative to the payoff of doing so) to map a Bluesky feed to a key pair in Nostr to make it easy to consume that way. Indeed, this could be a monetisation vector for relays able to offer more advanced interoperability features. It’s not like the problems of integrating with, say, Twitter, Reddit, or Facebook. There are far more obstacles to that than merely transforming data from one shape into another: violating terms of service, lack of APIs, KYCing, etc.

0000 sats

Globe9974d ago

Yeah it's interesting to look at this in the context of legacy social media: Facebook and X literally *can't* interoperate, not necessarily on any technical level -- although as these platforms are constantly introducing and taking away new features that becomes fraught -- but because of legal and fiduciary issues. Meanwhile, I think there are certain common "nouns and verbs" to microblogging social media that at least allow for some "floor" of interoperability -- posts, likes, comments, reacts, boosts, quotes, etc -- these things in their basic form are pretty universal. Obviously Zaps aren't going anywhere outside of Nostr, but signalling an "intention to Zap" could be a powerful way to attract people to the network.

000

Globe9974d ago

Actually, Mostr is pretty broken, as I've described: 📝 d0c360d0… Agree that this is a pathway to what @4b461234…7953bde6 has described as "adversiarial interoperability." Personally I think this is a great way to jumpstart Nostr's growth, but I don't think it's a long-term solution for the reasons @tree 木 describes.

000

tree 木74d ago

My comment was specifically about bridges and multi-platform clients. That's what I think doesn't work and is a waste of time. One-time migration tools make sense, but that's a little different story for me. For example, Matrix protocol has a lot of bridges, but that hasn't helped with adoption much. Similarly, on Bluesky, there is Bridgy Fed to connect to Activity Pub, but I don't think that has had any major impact on growth either. I feel like bridges like this are helping more the larger network become even more dominant, instead of helping smaller networks grow... so maybe it's counterproductive in some way.

0000 sats

tree 木74d ago

> @Melvin Carvalho has been working on DID support for Nostr, which would make any Nostr key pair pluggable into the DID/VC ecosystem. Do you have any details about that? The fact that Nostr identity will have a DID variant is useless in itself if other protocol (such as the AT Protocol) does not implement it. Moreover, the idea that you can create a DID variant of Nostr identity and it will be "pluggable" is very naive, ignoring all dependencies, identity is just the tip of the iceberg. Identity (actually the public key) is tied to human-readable identifiers, to profile information, content... and each protocol works completely differently in this.

0000 sats

Melvin Carvalho72d ago

Hello yes, working alot of did:nostr of late https://nostrcg.github.io/did-nostr/

0000 sats

Globe9974d ago

I don't think we even know what the "numbers" are. Nostr.band has been frequently criticized as inaccurate. How do you even estimate how many DAUs there are here, and what direction that's trending?

0000 sats

The Daniel 🖖74d ago

Openvibe, previously Plebstr, previously Tweetoshi was a repeated attempt at creating a product for the wrong audience. It never served its purpose, even if the concept was novel at the time. Tweetoshi was a great idea for its time. They had zaps on Twitter before Nostr existed. Then they focused entirely on building a Nostr client, but it was so full of bugs that it was unusable. When Bluesky launched they pivoted to try to build a Swiss Army knife of decentral-ish protocols, but there was no appeal for it because social network users are tribal and bridging has never succeeded, not once that I can remember, and I have seen it tried for at least 15 years.

000

tree 木74d ago

> posts, likes, comments, reacts, boosts, quotes It sounds simple, but when you think about the details, it's a very complex problem full of problems that you have to solve... for example, when a bridged user quotes a post of a user who is not bridged, and there is dozens or hundreds similar problems. And whats worst - It usually leads to centralization, that someone has to run those bridges and have the private keys to all those "pseudo" identities. For me, it's just not worth it.

0000 sats

tree 木74d ago

Bridgy Fed on AT Protocol is a typical bridge, connecting Activity Pub and Bluesky. Sure, it's probably a bit useful.... but I also see the danger - that there's one enthusiast who runs it and who now has the power to manipulate all those thousands bridged profiles. Or turn it off, in short, it's all dependent on one intermediary. The complexity of bridging clearly points towards centralization.

0000 sats

tree 木74d ago

It's not 15 years, but it seems like :) I think Matěj started this whole thing sometime around 2020, it was a very young team Yeah, the only interesting thing about Tweetoshi were the zaps, otherwise it was pretty stupid - building a client on a proprietary Twitter API 🤦‍♂ From the beginning, it was, from my point of view, quite clear inexperience and naivety. Now they are probably focusing on AT Protocol/Mastodon, that could bring them finally the growth they are probably expecting, but I don't know.

0000 sats

The Daniel 🖖73d ago

I’m talking about in general. I remember many attempts to launch apps that would let people crosspost between Facebook, Twitter, Google Circles or whatever the hell it was called, the list goes on. I mean, good luck to anyone who’s still trying to find the holy grail, but I don’t think it will ever work.

0000 sats

Globe9974d ago

>The complexity of bridging clearly points towards centralization. As it's always been throughout history... The operators of bridges, the original "trolls taking a toll", have always enjoyed this sort of privileged position. Society responds by (a) building more bridges to compete, or eventually (b) obviating the need for the bridge in the first place by migration to whatever side of the bridge has "greener pastures."

0000 sats

tree 木74d ago

You're missing the most likely option (c) do nothing :)

0000 sats

Melvin Carvalho72d ago

A bit about bridges here: https://socialdocs.org/docs/ecosystem/bridges

0000 sats

CXPLAY71d ago

Bridgy Fed's Nostr Bridge is under active development, and much of the groundwork has been completed. 📝 18fa87ee…

0000 sats