Anyone GNU/Linux users out there keen to kick the tyres on the #Scuttlebutt / #SSB network, without having to wrestle with #npm? #TIL that #Patchwork, apparently the most noob-friendly SSB app, has been packaged as an #AppImage:
https://appimage.github.io/Patchwork/
#ShowerThoughts I wonder if Scuttlebutt could use a #blockchain to map its gibberish user IDs ("public keys") to human-readable usersnames, like #Jami does?
https://jami.net/why-is-jami-truly-distributed/
But why? Once you start getting into people's distribution zones their "my name is" messages start populate your node and you see their nicknames. You only see public key ids for people who are too far from you, but their id was references in responses etc.
Correct, you can't "reserve" a nickname uniquely purely by means of protocol, but SSB is very much modeled after real life social interactions. When you initially post as `strypey` and you gain some followers, and you then come up with your new SSB node on your phone, you simply post a message on your _original_ account "hey look here's my second account" and you authenticate it through the authority of original account.
@kravietz ... OR, any SSB app that comes across a post with my public key could check that key against the ID namechain (like a decentralized keyserver) and assert with certainty that it was posted by 'strypey'. Just saying ...
If you want to disappear from social networks and delete your account, how do you do this with blockchain?
@kravietz with Jami, the account isn't stored on the blockchain, only a record that associates a certain account ID with a certain username. If you delete your account (or lose access to it), that username effectively points to nothing.
Ok, this indeed makes sense. I guess the blockchain would need to ensure uniqueness of the nicknames stored, which is the key difference with SSB.
@kravietz the blockchain is the means of ensuring uniqueness of usernames without a centralized nameserver. I don't understand why this wouldn't also be useful on SSB (or any social network), especially if it had 1000 times its current user numbers.
No, I don't mean it's not useful - I guess it's just complex from architecture point of view. Uniqueness can be ensured either by scanning the whole blockchain or by some kind of hash-addressed database. SSB on its own is just a quite simple protocol to distribute messages between peers using an append-only log. PatchWork is an application built *on top* of SSB, so this perhaps would require another application to be built.
@kravietz it would require SSB apps to include a node of the namechain (if they want to implement the feature) and check any unfamiliar public keys against it whenever they download new posts. Like the way web browsers include code to reference the #DNS system, so they can map the domain name of a website to the IP address of the webserver that hosts it.
Because all messages are digitally signed, there were discussions to possibly automate it - so if a SSB client sees a signed message "this is my new account" to somehow link them or auto-follow, but because the existing model works well nobody really cared.
@kravietz
> if a SSB client sees a signed message "this is my new account" to somehow link them or auto-follow
... doesn't do anything to allow users to claim a unique username across the network.
> because the existing model works well nobody really cared
This kind of thinking is how #IRC got overtaken by #Slack. IRC "works well" for the existing userbase, so the #UX doesn't get improved, so new users go somewhere else.
@kravietz
> You only see public key ids for people who are too far from you, but their id was references in responses
You kind of answered your own question ;) Also, the Jami namechain allows users to claim a unique human-readable username, without a centralized ID server. AFAIK in SSB as it works now, you'd have to check the public key of a post to be sure it was sent by this "strypey", not another user calling themself "strypey".