Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Superlogical

superlogical.com

Post reply on HN
Print view View original

441–450 of 504 posts

Re: Superlogical

#441

Post by miki123211 »

dregitsky wrote: post #429
brandall10 wrote: post #120

This sounds like a mashup of of several things I've been playing with lately... - pi-web ( https://pi-web.dev/ ): a web frontend for the Pi coding harness which can multiplex pi sessions across multiple machines - herdr ( https://herdr.dev/ ): a fairly polished agentic multiplexer TUI, can easily spawn or destroy windows based on subagent activity - firstmate ( https://github.com/kunchenguid/firstmate ): a meta envir…

These tools all assume you have machines to run the agents on. But for parallel agents I'm pretty convinced you want each agent on its own isolated devbox running your dev environment (not e.g. worktrees on one box) - which isn't trivial to set up and manage. I'm working this with https://boxes.dev - a workspace for launching and managing claude + codex sessions, each running in its own cloud devbox. We launched with…

Running on a Cloudbox has its own set of problems. Some projects need desktop GUI tools (particularly desktop apps themselves, but some may also want these for profilers / debuggers etc). Some projects need USB. Even if you don't need any of those, you need to figure out what to do about build caches, package manager environments, credentials / secrets, VPN config, shell customizations, and those are just a few things that come to mind.

I do like having a permanent, remote workbox, but am a lot less keen on the idea of ephemeral boxes unless you're a company willing to invest into that infra.

Re: Superlogical

#442

Post by zombot »

> And as the work moves to production it lives behind separate systems and controls.

That's not a bug, it's a feature. With exceptions, I think wanting to change that is misguided and won't fly anyway.

Re: Superlogical

#443

Post by theshrike79 »

whilenot-dev wrote: post #431
simonw wrote: post #228

Earlier quoted context omitted.

> Would it really be so bad for Anthropic to admit that a project of this size takes months instead of weeks? I think that's exactly what they've done by not shipping a stable release yet. I don't think they've been pushing that "It only took us 11 days!" message nearly as hard as their critics suggest that they have.

> I don't think they've been pushing that "It only took us 11 days!" message nearly as hard as their critics suggest that they have. I have to disagree here. From their blog post Rewriting Bun in Rust [0]: > This Rust rewrite would've taken a team of engineers with full-context on the codebase a year of work. With 1 engineer using Fable & closely monitoring Claude Code, we went from start to 100% of the test suite pa…

Mentioning it once in a blog isn't "pushing it". It's not like they went on a podcast interview spree touting how they did it in only 11 days.

Re: Superlogical

#444

Post by broodbucket »

pama wrote: post #220

Serious question: if one were willing to give up on curses, isn’t Emacs/elisp providing the best multiplexing system available to humankind? And conceptually, why would agents ever need to depend on curses?

If you can find a way to stop Emacs from being painfully single threaded, yeah.

Re: Superlogical

#445

Post by zazuke »

lemontheme wrote: post #433
backscratches wrote: post #423

Earlier quoted context omitted.

Isnt zellij quite modern? What does herdr do better than zellij?

As a zellij sponsor I’ve been wondering the same. I rebuild herdr from master every few days to gauge the direction it’s moving in. What herdr gets right over zellij: - Default keybinds don’t conflict with helix. Even the leader key mode in zellij keeps alt-bindings around. Herdr keeps things nicely tucked away. - There’s beta support for the kitty image protocol. Three years on, we’re still waiting in zellij. - Noti…

I started to use tmux-claude-session-manager (https://github.com/craftzdog/tmux-claude-session-manager) in tmux, to have a pop-up to see all active sessions. I don't have many in parallel, so usually I don't even need, but if I'm lost or forgot one, it's really helpful.

Re: Superlogical

#446

Post by dolmen »

dregitsky wrote: post #429
brandall10 wrote: post #120

This sounds like a mashup of of several things I've been playing with lately... - pi-web ( https://pi-web.dev/ ): a web frontend for the Pi coding harness which can multiplex pi sessions across multiple machines - herdr ( https://herdr.dev/ ): a fairly polished agentic multiplexer TUI, can easily spawn or destroy windows based on subagent activity - firstmate ( https://github.com/kunchenguid/firstmate ): a meta envir…

These tools all assume you have machines to run the agents on. But for parallel agents I'm pretty convinced you want each agent on its own isolated devbox running your dev environment (not e.g. worktrees on one box) - which isn't trivial to set up and manage. I'm working this with https://boxes.dev - a workspace for launching and managing claude + codex sessions, each running in its own cloud devbox. We launched with…

Also the reason why https://exe.dev exists.

Re: Superlogical

#447

Post by brabel »

dregitsky wrote: post #429
brandall10 wrote: post #120

This sounds like a mashup of of several things I've been playing with lately... - pi-web ( https://pi-web.dev/ ): a web frontend for the Pi coding harness which can multiplex pi sessions across multiple machines - herdr ( https://herdr.dev/ ): a fairly polished agentic multiplexer TUI, can easily spawn or destroy windows based on subagent activity - firstmate ( https://github.com/kunchenguid/firstmate ): a meta envir…

These tools all assume you have machines to run the agents on. But for parallel agents I'm pretty convinced you want each agent on its own isolated devbox running your dev environment (not e.g. worktrees on one box) - which isn't trivial to set up and manage. I'm working this with https://boxes.dev - a workspace for launching and managing claude + codex sessions, each running in its own cloud devbox. We launched with…

I stopped running Claude on the terminal and use only the desktop app now. It had lots of advantages in my opinion. The terminal is really limited unfortunately, don’t know why so many devs still stick to it, and I used it for 30 years.

Re: Superlogical

#448

Post by vultour »

63stack wrote: post #430
jagadaga wrote: post #426

Sorry, the text is too vague - what are they building exactly, another tmux? P.S. That custom disappearing scrollbar doesn't look super logical; it doesn't even look logical, tbh.

Yes. I find it quite funny that 4 people with the flashiest of titles (head of experience?) came up with this amazing plan: 1. it will be incredible 2. it will be composable 3. it will be safe to use in production I would totally invest millions into these guys, they have a bright future ahead.

My first thought was that they have several people who seem like they'll expect very serious compensation but no actual product, then I noticed they already seem to have a whole bunch of VC backers. I'm not gonna lie, I'm a little jealous of people who can make a flashy vaporware pitch deck and ride the VC gravy train.

Re: Superlogical

#449

Post by wild_egg »

hoppp wrote: post #410
wild_egg wrote: post #394

Earlier quoted context omitted.

He says here that they have a direct Tailscale integration of the type that requires using the `tsnet` package, which is Go. Or at least would pull in the Go runtime as part of a shared library. https://x.com/i/status/2082634453474795885

Interesting choice because tailscale is not free. Building in a paid third-party service is a questionable decision. They better acquire it, if their product catches on. Building with wireguard directly instead of tailscale would have been my choice, but of course it requires more plumbing.VC money means they gotta move fast. But as a potential user, I won't pay for tailscale just to use their product.

What's wrong with the Tailscale free tier?

There is also Headscale if you want to self host.

https://headscale.net

Re: Superlogical

#450

Post by myaccountonhn »

bryanlarsen wrote: post #219

Sounds like something that could be done with by pulling in a few dozen Emacs extensions and writing a few thousand lines of elisp. (And one of those extensions would be the libghostty base ghostel.) It's sort of what I'm doing already -- using emacs-server to provide portable persistent sessions on top of a bunch of terminals, agents and editor buffers. It's a 1% effort on my part and super janky and super me-specif…

Sounds like you're using LLMs correctly IMO. The entire appeal of LLMs in dev-tooling is being able to setup hyper-specific, customized workflows and have "extensions" be done by telling the LLM what to fix. For that, the smaller and more minimal the base is, the better.
Post reply on HN