Oddbean new post about | logout
 Im curious which bits of bluesky spec rabble thinks are done better.

Nostr is far from perfect, but there's a path to clean up everything and add functionality.

Atproto, imho, cant be fixed. 
 They have for sure thought about the spec way more but it does not mean that it’s better at the end As Rabble mentioned in his speech today: Betamax did not win even though it was superior.

The speed of Nostr is amazing. The trial-and-error messiness is going to give results that are hard to beat. The multithreading by so many users, apps, builders and ideas at work at the same time. 

https://media.tenor.com/ch2RGZjVzHIAAAAC/vhsmts-vhs.gif
 
 In 15 years experience of standards, I honestly dont think they thought it through at all.  It's just a mix of different ideas.  Credit to them they made it work, they made a great app, and they invested in infrastructure.  Certainly in those aspects they are ahead and nostr can learn.  But the atproto spec I do not think will stand the test or time, or do they fully understand it.  Im glad they exist, all credit to jack for trying different approaches.  That man's intuition is pheonomenal.  Hopefully we'll all interop one day. 
 There seems to be a lot of momentum for people joining Bluesky now since they don’t know anything better and it has the feel of familiarity from X.

Yet, it feels less developed than nostr in many aspects. There are no possibility of viewing videos, the lists are barely existing, the API gives random errors etc.

I have come across arguments where people justify it because it’s open-source and you can export your data etc. Lot of the things are still intentions without practical actionable substance but they are part of the reasoning. Like Telegram managed to create a reputation for things that it really never was but people still repeat those points.