Is anyone aware of a document/podcast/video that compares all of the various “web5” protocols?
I think this is a great question, I would love to see such a write up.
Here's my limited understanding of how each of these stack up. Maybe someone who knows better can help fill in the gaps or correct my mistakes:
Nostr
  • Identification - public/private key pair: npub and nsec
  • Identity validation (tying your key to some real world identity) - NIP-05 (DNS)
  • Data exchange protocol - Nostr protocol - clients send signed messages to relays
  • Data storage and metadata - relays store user data (I think??)
Synonym
  • Identification - public/private key pair: Slashtags
  • Identify validation - ?
  • Data exchange protocol - Hypercore - users send signed messages or data to append only logs
  • Data storage and metadata - Hyperdrive (I think??)
Bluesky/TBD
  • Identification - public/private key pair: DIDs (Decentralized Identifiers)
  • Identify validation - I believe the spec stores some info on the blockchain. They seem to also be focusing on allowing individuals to verify their DID with some organization (eg. a government, a company, an educational institution)
  • Data exchange protocol - Decentralized Web Nodes (???)
  • Data storage and metadata - Decentralized Web Nodes (???)
The obvious (at least to me) step, in terms of compatibility or interoperability, would be for the projects to agree on the nature of the public/private key pair. I'm not sure if they all use the same spec currently, but if they did, I would think it would be trivial to move from one protocol to the next because the public key can be used as a bridge.
reply
AFIAK
Nostr uses bitcoin taproot keys and schnorr signatures
Slashtags use ed25519 private keys and a custom derivation of public key
DIDs ("web5", bluesky) use did methods to define identity (there's 100+), verifiable claims for signatures (ongoing work in progress)
Nostr benefits from being finished and easy to implement.
There are over 100 nostr projects on github alone. With more than one new project daily:
reply
worse is better?
reply
My guess is someone will create tools to translate keys to other protocols. The private key will be the same, but the public keys will have different values due to the differing EC groups, and then they can add support for providing public keys from other protocols and the apps can bridge them. It's just a matter of adding fields to the user information packet.
reply
Interoperability in this context is a red herring. Specs and protocols compete, and eventually there is capitulation to the winner that gets the most support and thus most tools, in the most coding languages, for the most OS's, with the most end users, skin in the game, etc.
Interoperability is not easy even when you cooperate. People use different languages to program in on both the base/native level, and then in implementations at app level, and then syncing over protocol/specs is another issue.
Even LN implementations are not fully interop.
Interoperability just means people using the same thing, but these are all different things (Slashtags, Web5, nostr) despite them solving some overlapping use cases. So what is really needed is some combination of migration, resolvers, aliasing, conversion, support for multiple formats in clients, etc, etc.
You probably won't see much of that at this stage. This is not to say that Synonym is closed-minded or stubborn, so much as all of our teams have to manage our priorities. Synonym has actively researched, contacted, and collaborated as often as possible, but this environment, so far is honestly one of the more hostile and collaboration-disinterested I have encountered. Very competitive.
I personally think managing keys across formats, resolving/aliasing, is an interesting problem, but until there are actually two popular formats, it is merely a curiosity and academic challenge, thus not a priority.
reply
Given the (current) lack of interoperability, are you concerned that a small pool of users may be spread too thin amongst the various protocols, thus making it even more difficult for any one protocol to achieve network effects?
Given the open source nature of the protocols, what do you see as the best way for protocols to attract users? Court a centralized entity to dedicate itself to a protocol and perform business development functions for that specific protocol?
reply
Specifically interested in which features of the various protocols will be interoperable with each other. Eg/ will the DID’s within Block’s TBD protocol be interoperable with Holepunch and/or Synonym’s Slashtags?
reply
Interoperability in this context is a red herring. Specs and protocols compete, and eventually there is capitulation to the winner that gets the most support and thus most tools, in the most coding languages, for the most OS's, with the most end users, skin in the game, etc.
Interoperability is not easy even when you cooperate. People use different languages to program in on both the base/native level, and then in implementations at app level, and then syncing over protocol/specs is another issue.
Even LN implementations are not fully interop.
Interoperability just means people using the same thing, but these are all different things (Slashtags, Web5, nostr) despite them solving some overlapping use cases. So what is really needed is some combination of migration, resolvers, aliasing, conversion, support for multiple formats in clients, etc, etc.
You probably won't see much of that at this stage. This is not to say that Synonym is closed-minded or stubborn, so much as all of our teams have to manage our priorities. Synonym has actively researched, contacted, and collaborated as often as possible, but this environment, so far is honestly one of the more hostile and collaboration-disinterested I have encountered. Very competitive.
I personally think managing keys across formats, resolving/aliasing, is an interesting problem, but until there are actually two popular formats, it is merely a curiosity and academic challenge, thus not a priority.
reply
That's a good take
Personally, from where I'm sitting, I see it as two horse race between:
Nostr + bitcoin IPFS + altcoins
Not to belittle other efforts, I wish them well. There's various other things out there that may change the game, but more or less those two, have, and will continue to have traction, imho
Smart would be to pivot to one or the other, at least as a side-gig, nostr likely being the non-fiat option
The linked data world has some traction but no payments, in general
IPFS ahead, but nostr with some momentum
reply
Holepunch code is super simple to use. I wish they will port it to some other programming language soon, so one can start building ontop (without node.js)
reply
The "web5" terminology is so cringe.
reply