So what happened? I was away for a while, spent some time at a psychiatric hospital. Now I'm feeling better and feeling more motivation again. So I reconnected my scuttlebutt identity. Because scuttlebutt provides a better protocol than the #fediverse. Key based identities, replication based on social distance make a foundation for trust and credibility.

But the feeling isn't there anymore.

Three years ago I was fascinated by this community. It felt like the place that could revolutionize the way computer help us communicate and gain knowledge.

Now it feels more like a community sworn in around an old code base and some rather esoteric design patterns.

I really tried hard to make it work. Learned pull-streams and tried to make it easier to create new apps on top of the ssb stack. Webdevelopers should be able to build apps on top of scuttlebutt, so I developed #patchboot

Eventually I gave up with node and pull-streams and decided to implement the scuttlebutt protocol from scratch using modern JS. Implementing the protocol I discover some of its flaws, like inconsistent encoding and assumptions about the JSON serialization not warranted by the spec.

I'm being told that the feed format will change anyway and something new and better is about to come.

Then the spec is changed to include "Room 2.0". It seems overly complex to me. I try to express my concerns, I'm being told that I'm too late for these proposals. That I should have participated in earlier discussions.

I believe that the remaining parts of the spec are enough to build great app, so I go on coding with #scuttlesaurus and #tricerascuttler.

Then other stuff happens in my life. And now I wanted to come back.

First I learn that Manyverse no longer supports the protocol documented in the protocol guide. Apparently oasis too now uses ebt.

Once connected I want to see what my community is saying about #nostr. Not a lot. @Dominic wrote that it wouldn't scale. @andrestaltz📱 describes it as a smaller derivative of scuttlebutt. @Rabble mentions the idea of moving #planetary away from the scuttlebutt protocol.

When I look at nostr I see a community around a well documented and modular potocol that has the most relevant features of the scuttlebutt protocol. The scuttlebutt protocol provides a multi-layered stack with Handshake, box-connections, RPC, and Feeds. The nostr protocol by contrast, bases on websockets to exchange events. The protocol is described in NIPs (something like RFCs), implementing a few of them is enough to partcipate in the network.

So yes, I think that time spent on the #nostr protocol us better invested than trying to catch up with the undocumented changes of the scuttlebutt protocol. Maybe many ideas from scuttlebutt can be implemented on #nostr in a better way.

On #nostr I'm 9b9a4c51ed78effff213baec91e8f24ec0ee7399b9e3705928ce2d8e7d296c6c , do you have a key I could follow?

I see where you’re coming from. For me nostr has more standard tech, similar to how p2panda also made less bespoke choices like websockets and postgres vs custom rpc and a custom db. The writing out of specs is refreshing too.

The big difference for me is that nostr supports delete, multi device using a single key, and relay servers so we don’t rely on punching through NATs and running everything in device. Running everything locally makes ssb damned hard.

I’m experimenting with what a version of planetary running nostr protocol as a fork of the app. There is a landing page at nos.social.

Andre wrote up a critique of nostr on twitter. I think you can’t claim you care about user safety and not prioritize delete. Addressing spam is important, but I’m not sure nostr’s subscribe is that different from ssb’s contact pull streams.

My nostr id is: npub1wmr34t36fy03m8hvgl96zl3znndyzyaqhwmwdtshwmtkg03fetaqhjg240

I've only just started looking at the apps. For the twitter like apps such as https://nostr.rocks/ I haven't yet found the users to follow. I want to have a closer look at nvote.

This echos a number of things I felt working on Tilde Friends (0.0.2, aww yeah). I'm still feeling good about isolating the protocol specifics and building apps that only interact with a database using a few key functions, though. I will have to read up on nostr for sure.

Sad that you're leaving, but if you made your mind, then that's it. I have tried to give or do some of the things you complained about, but doesn't ever feel enough.

About our recent interactions, yes EBT is not documented on the protocol guide, but the protocol guide isn't the authoritative specification of the whole of Scuttlebutt, there is no authoritative specification "yet", and I have already told you this. And EBT is by no means "new", it was presented by Dominic some 5 years ago, implemented in JS 5 years ago based on an academic paper, and deployed in Patchwork, Oasis, Manyverse, etc.

Andre wrote up a critique of nostr on twitter. I think you can’t claim you care about user safety and not prioritize delete.

About Nostr, yes I've said all that I want to say on Twitter already. I don't like it, @Rabble, how you're baiting me into a discussion here. I am working on deletes, this is what database compaction is about. You can't just "implement deletes and ship it", it comes with a ton of breaking changes to Scuttlebutt and (bits lip) crashes that I have to debug painfully on my weekends. It has to be done super carefully.

I think you can't claim you care about user safety and host open pubs, that's a worse problem. In a P2P system, the difference between spammers and pedophiles is just the meaning of the bytes transmitted. When I came to SSB, what really surprised me about it was how clean the network was in terms of content, while not having any official moderators of any kind. When I open Nostr apps, it's similar to Lens/Lenster in my experience: a wall of "gm", "hi", "test", "test" showing very few guarantees regarding content safety.

Running everything locally makes ssb damned hard.

I think many features of SSB can be implemented on nostr, and that this would lead to a cleaner and more accessible protocol architecture. For instance there's nothing preventing a relay from acting as client and synchronize the messages of contacts of contacts with peers it discovers over UDP Broadcast.

moid we’re just exploring what might be involved with making an app using nostr. There are parts of nostr which are better than ssb. There are things ssb has figured out. There are things which p2panda and earthstar are able to do that ssb doesn’t. The question is how to move forward. It’s been really hard to get ssb as a community to solve problems which are important.

So it’s something we’re exploring. Planetary is open source and won’t be going always even if we do create a fork of it to run on nostr.

farewellutopia I’m glad you’re doing better and I hope you were able to get help. Mental health is both important and invisible.

my nostr is npub1wmr34t36fy03m8hvgl96zl3znndyzyaqhwmwdtshwmtkg03fetaqhjg240

I think you can’t claim you care about user safety and host open pubs, that’s a worse problem.

@andrestaltz and anyone could host open pubs for sbb to though right? Does user safety == adding more barriers to entry? I can buy that, less users you have the more safe they are forsure.

One thing about nostr is that is not offline friendly. While you could probably build a client that caches all the events you are interested in, how you exchange those events with another peer would be outside the protocol.