Now when I see new software projects I just think of how much of them are made with agents and the unreliability/easy abandonment that comes with it. Ten years ago this product would have been of a known quality. Not saying anything about buzz specifically.
Having agents see everything you and your colleagues see is cool.
The challenge comes when you want to make certain things private to certain people. In that world you don’t want multiplayer agents leaking data, so you end up having to write and maintain complex rulesets about what specific resources agents have access to.
Single-player agents, on the other hand, are much more straightforward — they act on a single user’s behalf, and then the challenge is ensuring the single-player agent cannot exfiltrate private data into a shared space unless the user specifically permits it.
I'm from Asana so I'm probably biased too :) and I agree single-player agents are easier, but multi-player agents that understand end to end workflows etc. are very powerful. That's what we've built and yes it took a lot of thought and time to built it in a way that it would be privacy conscious and not leak the wrong thing to the wrong people etc. I haven't player with Buzz yet but signed up and admire them for thinking differently and trying to build something interesting.
Slack exists largely because IRC was insufficient. It didn't natively support channel history, search, etc.
For AI agents to flourish, Slack has to either truly open its network with a protocol or eventually be replaced.
I'd like to see Slack embrace an AT protocol-based chat system, which then apps like Buzz could implement. Then users could log in with domain handles like @yourname.com, and agents could use handles like @agent1.yourname.com, all under their complete control.
No one is going to churn Slack for this just because it is "Agent First". Slack is more than good enough to tag Agents and will fill in the missing pieces slowly
> Buzz is an open-source, self-hosted workspace that combines team chat, AI agents and Git hosting using signed Nostr events, so teams can keep control of their data.
Hey I was prototyping this exact kind of thing, there's something to be said for releasing early.
The project fell to the back burner because I had VM ambitions and just didn't have enough disk space on my macbook to do VMs well. 512 GB and always full
Bots in team chat is not a bad idea. I've been experimenting with that for a few months.
We started with Slack. This kind of works but dialing in the gazillion of permissions needed to end up with something that works is a bit painful. And you have to do this over and over again for each new bot.
So, we started experimenting with self hosted alternatives. First we tried Matrix. It's nice but a bit too strict on end to end encryption which gets in the way if you actually want to share information with bots. So, a few weeks ago we switched to Zulip. That's pretty straightforward to setup as well. It's very easy to create bot users with both and create some automations. We did some with Openclaw and I ended up replacing those with something coded around the haystack framework, which is a bit less of a messy basket case than Openclaw.
And as I discovered after I installed it, the Zulip leadership actually just got hired by Anthropic. It seems Jack Dorsey beat them to market making some announcements but I imagine that Anthropic might have very similar plans.
IMHO team level use cases for agents actually make a lot of sense. Companies are run by groups of people and the larger those groups the more team overhead you get. Perfect for optimizing with AI. And if you have people doing a lot of stuff with AI, that only adds to the need for them to communicate what they are doing and coordinating with others. Doing that out in the open in some shared channel makes a lot of sense. And having some shared team guard rails also makes sense. And if you think about more complex processes with handoffs between people (and possibly some agents), a team chat is a good place to do those as well.
Interesting idea. I think the "multiplayer mode for AI development" bit resonates somewhat. Biggest problem is that I do think most companies have a ton of key contributors that find git/code intimidating, so part of me thinks the real answer here will be stuff like Surfboard, Town, etc. that can work with git AND the work tools in a more user friendly, browser-first way.
I've recently built some project that also supports Nostr (https://getbirdfeeder.com, but more for the fun of playing with Nostr) but it feels very forced to use it for something like that. It feels like it's solving an issue that no team chat ever has.
> Buzz is built around a self-hostable Nostr relay. Every message, reaction, workflow step, code event and approval is stored as a cryptographically signed event. Human employees and agents receive the same basic identity structure, including their own key pairs, channel memberships and audit trails.
The name is probably not the best fit but if we can get a good solution that's not horrible that's easy to self host I and open-source I am fine with it.
Though I fear this might not be well maintained or have a good foundation being so heavily agent focused but I will out of sheer ambition of a more open stack support anything that help make it a possibility.
That screenshot is some Lynchian horror. "#engineering. New direction, team – we're moving the prototype to Flutter" and then the humans and agent bots engage in flirty emoji-filled chats with cutesy names "I nailed the physics, UI shell @Honeybot?"
I'm trying to imagine a world in which this makes sense as a way to organize software development work, but coming up short. It's on a blockchain or something though, so there ya go.
I have a premonition of future programming despair, and I wonder if it'll be like what Ian McKellen felt: A veteran of Shakespearean plays, doing all those [correction, not LoTR] The Hobbit green-screen shots interacting with no real person...
__________
> McKellen said: "I felt pretty miserable … and thought perhaps, has the time come for me to stop acting altogether if I can't cope with these difficulties?" [...]
> "It was so distressing and off-putting and difficult that I thought 'I don't want to make this film if this is what I'm going to have to do'," McKellen added. "It's not what I do for a living. I act with other people, I don't act on my own."
P.S.: I want to make it clear that this isn't just about dealing with other humans. The dread I contemplate should affect even introverted shut-ins.
I want to reduce accidental complexity, discover and understand real problems, solve them well, and leave the world somehow better for it. Not just some infinitely process ping-pong of a hallucinated digital dollhouse fifteen levels removed from objective reality.
Maybe it's just me, but when I was younger one reason that got me into AI research was the idea that one day we would have these seemingly intelligent entities that we could talk to and interact with just like we do with humans.
Now that we have that, everybody hates it.
Except me. I still like this. There might still be room for improvement, yes, but I think this is a cool attempt to making AI agents part of the team.
I think what could help make this better is investing into giving agents actual personality, instead of them being just sycophantic clones of each other. AI agents as team members should have clear aspects of distinction both in their abilities and behaviors.
The screenshot is totally scripted. It’s from a video capture, which involves a few moving parts that are difficult to choreograph live with multiple humans and agents. You can see the PR for the scripted demo in the repo.
I wouldn't underestimate whimsy as a way to make fundamentally strange things feel intuitive. Bots are managed by people or orgs and can have different capabilities and authorization, so names and images are convenient shorthand for humans to distinguish these instances from one another. Whether or not you make them cutesy is entirely up to you.
Nobody notices that all the messages are labeled "5:42 pm"? From the proposition of the idea to the the review of the PR. This of course is impossible. I mean, even AI bots can understand and implement the features that fast, it is impossible for your human colleagues to notice your message and respond in the same minute. So, this image is also generated by AI?
115 comments
[ 0.25 ms ] story [ 32.1 ms ] threadWebsite link is here: https://buzz.xyz
The buzz project itself is also open source: https://github.com/block/buzz
Having agents see everything you and your colleagues see is cool.
The challenge comes when you want to make certain things private to certain people. In that world you don’t want multiplayer agents leaking data, so you end up having to write and maintain complex rulesets about what specific resources agents have access to.
Single-player agents, on the other hand, are much more straightforward — they act on a single user’s behalf, and then the challenge is ensuring the single-player agent cannot exfiltrate private data into a shared space unless the user specifically permits it.
For AI agents to flourish, Slack has to either truly open its network with a protocol or eventually be replaced.
I'd like to see Slack embrace an AT protocol-based chat system, which then apps like Buzz could implement. Then users could log in with domain handles like @yourname.com, and agents could use handles like @agent1.yourname.com, all under their complete control.
1. https://en.wikipedia.org/wiki/Google_Buzz
> Buzz is an open-source, self-hosted workspace that combines team chat, AI agents and Git hosting using signed Nostr events, so teams can keep control of their data.
Very different expectations of 5-year-olds.
Anyone know what it actually does?
The project fell to the back burner because I had VM ambitions and just didn't have enough disk space on my macbook to do VMs well. 512 GB and always full
We started with Slack. This kind of works but dialing in the gazillion of permissions needed to end up with something that works is a bit painful. And you have to do this over and over again for each new bot.
So, we started experimenting with self hosted alternatives. First we tried Matrix. It's nice but a bit too strict on end to end encryption which gets in the way if you actually want to share information with bots. So, a few weeks ago we switched to Zulip. That's pretty straightforward to setup as well. It's very easy to create bot users with both and create some automations. We did some with Openclaw and I ended up replacing those with something coded around the haystack framework, which is a bit less of a messy basket case than Openclaw.
And as I discovered after I installed it, the Zulip leadership actually just got hired by Anthropic. It seems Jack Dorsey beat them to market making some announcements but I imagine that Anthropic might have very similar plans.
IMHO team level use cases for agents actually make a lot of sense. Companies are run by groups of people and the larger those groups the more team overhead you get. Perfect for optimizing with AI. And if you have people doing a lot of stuff with AI, that only adds to the need for them to communicate what they are doing and coordinating with others. Doing that out in the open in some shared channel makes a lot of sense. And having some shared team guard rails also makes sense. And if you think about more complex processes with handoffs between people (and possibly some agents), a team chat is a good place to do those as well.
Indifferent to "groupchat for agents".
Like "maxing-out-8-cores"-crazy?
> Buzz is built around a self-hostable Nostr relay. Every message, reaction, workflow step, code event and approval is stored as a cryptographically signed event. Human employees and agents receive the same basic identity structure, including their own key pairs, channel memberships and audit trails.
Though I fear this might not be well maintained or have a good foundation being so heavily agent focused but I will out of sheer ambition of a more open stack support anything that help make it a possibility.
Also the git hosting stuff seems a bit sus tbh.
I'm trying to imagine a world in which this makes sense as a way to organize software development work, but coming up short. It's on a blockchain or something though, so there ya go.
https://github.com/block/buzz/blob/main/docs/assets/screensh...
__________
> McKellen said: "I felt pretty miserable … and thought perhaps, has the time come for me to stop acting altogether if I can't cope with these difficulties?" [...]
> "It was so distressing and off-putting and difficult that I thought 'I don't want to make this film if this is what I'm going to have to do'," McKellen added. "It's not what I do for a living. I act with other people, I don't act on my own."
-- https://www.theguardian.com/film/2013/nov/20/the-hobbit-gand...
I want to reduce accidental complexity, discover and understand real problems, solve them well, and leave the world somehow better for it. Not just some infinitely process ping-pong of a hallucinated digital dollhouse fifteen levels removed from objective reality.
I don't see anything about 'nailing' the physics either. It's just a design instruction.
Now that we have that, everybody hates it.
Except me. I still like this. There might still be room for improvement, yes, but I think this is a cool attempt to making AI agents part of the team.
I think what could help make this better is investing into giving agents actual personality, instead of them being just sycophantic clones of each other. AI agents as team members should have clear aspects of distinction both in their abilities and behaviors.
To me, it seems like a loud online minority, and it's mostly entangled with politics or job security.
Most good programmers I know are using it effectively and like it.
I wouldn't underestimate whimsy as a way to make fundamentally strange things feel intuitive. Bots are managed by people or orgs and can have different capabilities and authorization, so names and images are convenient shorthand for humans to distinguish these instances from one another. Whether or not you make them cutesy is entirely up to you.