Hacker News

Can't unsee the stark contrast with Anthropic/Bun's recent shenanigans over unreleased rust migrations hehe

At least there's no history of security issues with that.

I think you may have meant to comment on another post :)

I've been working on collie[1] a plugin for herdr that runs a web app on the herdr host which can then be served via tailscale for example. This has made accessing my agents on the go a lot easier, paired with session persistence from herdr it's been quite nice to work with.

[1] https://github.com/AltanS/collie

Since you seem curious about these types of tools, might I interest you in ouijit https://ouijit.com/ https://github.com/ouijit/ouijit

Its intended as a terminal agent orchestrator aimed at power users who want a little more design than what I like to call 'hackerman' UXes.

While not terminal-only, I've been using Orca (https://github.com/stablyai/orca) for the past week and it's been a great experience

The actual header on the page is

> We are building the multiplexer for all work.

imo this would have been a better way to do the HN submission title (with or without [terminal])

> Superlogical: We are building the [terminal] multiplexer for all work

One of the worst I've seen was title editorialized from original "Someone: something [x.com]" to just "Something [x.com]".

Without clicking you’d never know why it’s the title.

NSA and other companies historically had you jumping hoops and writing mini-exploits to be able to see the email you're supposed to send your resume to.

The ssh thing is really fun, but I wish they had another way to view the openings. I'm on a connection with massive latency at the moment and it's pretty horrible UX in this case

In case others were interested, they are not a remote-first company and require you to live in LA, NY or London.

It’s probably a good filter too for applicants

Right, but I'm pretty sure this is not the first time somebody thought about solving it.

https://xkcd.com/927/

They hire for Go.

Considering it's going to consume libghostty, I'm betting it'll be Zig.

The agents have to live somewhere. They address your question in the post.

Scrollbar is fine for me? Just styled

I think IDEs took off plenty well, but several got frozen in time.

Visual Studio as the IDE for building windows client apps, especially WinForms, was and still is a genuinely incredible and productive experience. The ease of which you could click-and-drag a UI, context switch to logic in the code behind the forms, wire up a web service backend, share domain models and have navigation and search tools understand all of it was great. And the TFS integration was quite good too. You really only had to leave VS to do database work or administer some IIS pools.

Unfortunately, it couldn't keep up with the rest of the dev world. "Server Explorer" morphed to include your Azure VMs but froze for 30s each time you opened it. Docker tools came in super late, most Azure services didn't really have a VS tooling experience. The "right click publish" workflow meant for small systems never figured out how to graduate to real-world needs across many teams. Developers had to learn how to use many other tools for modern stuff, and so VS became far less of a center of gravity, often to their annoyance.

I think they definitely have the IDE vision in mind and are trying to move off of terminals, based on two parts of the original article

> so you can close the application, reconnect from another device, and pick up exactly where you left off.

I think this is the remote IDE development dream that no one could quite get right, partly because there wasn't enough reason to run things on a 24/7 device. I think with coding agents that can run in the background this changes.

> Sessions can be accessed through the web and native macOS/iOS applications, and sharing a live session with other people is built in from the start.

I'd really like to point out that "Sharing a live session" is horrible UX for a terminal multiplexer because the terminal has to be the same size (# of rows/cols) on all viewers' screens, and some of these viewers may be on a phone and others on a laptop. Believe it, I tried myself and there's no good way around the problem.

What is interesting is that they added this sentence anyway. This tells me that their long-term endgame is not the terminal. They're most likely trying to build a GUI-based agent IDE and then eventually expand to other work (per their headline of being the "multiplexer for all work") as it all gets more and more agent-based anyways. Well, they're not alone in having that vision.

Somewhat related, I recently heard of fentas/atty from:

https://erock.prose.sh/posix-shell-is-all-you-need

You are missing out a bit, but stay content. :)

Multiplexers have a lot of little papercuts and limitations when you want consistent and native-to-the-OS behavior to flow all the way between the OS and the apps and connections in the multiplexers. For example, gesture scrolling in a tmux/zellij pane is hard because the right events don't flow through. It's an itch to scratch like designing a good terminal lib (like ghostty.)

And then, there's all the unifying and smoothing out of data connections--I imagine they'll be a lot to make sure that, e.g., an ssh session or a Claude code session in a pane can be seamlessly reconnected or never drop when the pane is closed. This takes a lot of hacking to make look seamless in most setups right now.

Like mitchell, I write a lot of software around multiplexers, and I've thought about making something similar, but I don't have quite the same level of personal wealth or angel investor connections. :)

herdr and mosh are a couple of reference points for this, there are others.

I've wondered this myself. I stuck with iTerm2 for a really long time but recently switched to Ghostty. Ghostty definitely seems to renders faster (I say "seems to" because I didn't benchmark or anything).

I don't actually use a multiplexer. I'm not sure I understand the use case when most terminal emulators supports tabs + windows (like iTerm2 does). Or you could just launch multiple windows and use your OS's window manager to "multiplex". Are the terminals connected in some way that I'm missing?

Regardless, I think Ghostty is a reminder that you can win in a space that's crowded. There's probably a lot of widely used software that would get dropped en masse immediately if someone had the guts to make a better alternative.

I imagine something more like herdr, but done at the terminal level, not inside. But that’s cmux…

If you like terminals and tmux, can I suggest you spend a few hours playing with herdr? It clicked for me in a way that a lot of other orchestrators didn't.

If it’s touched by Tobias Lütke, I’m out.

Or you want to hire the kind of seasoned veterans who want to live in those cities. Often later-career, families, looking for exposure to more than just a thriving tech scene.

Another example of the galaxy brain people who like employees to go to an office but then are happy to have offices all over the world, some of which have 8 hours of timezone difference.

Are they the champions of collaboration or the champions of company culture?

Like notable lifestyle businesses Datadog, Squarespace, Etsy, Tinder, and Snap.

I don’t think that’s a parsimonious explanation in a post-COVID world. It seems equally if not more likely that talent is more distributed, and people find fewer reasons to relocate into the Bay Area (in terms of career progression, etc.)

(This is not, in my mind, either a good or a bad thing.)

Its interesting they recognize they've themselves worked remotely for most of their careers but not for our new company. Maybe its a new trend

Funny how just 3-4 years ago on this forum people were openly prognosticating about how the era of fully remote work was here to stay and the centre of gravity of investment no longer had to be SV and YCombinator was running fully remote sessions, etc. etc.

And here we now have people saying stuff like this again.

Also, long live the "lifestyle business" if it means people just making businesses that bring in non-exponential but justifiable and sustainable income ... like, y'know, a business is supposed to. Good grief.

What do you mean by lifestyle business?

At the risk of sounding like a booster, I have noodled over whether this is a place that AI can legitimately provide some improvements. It's harder than just waving an AI at the problem, which is why we don't really have this yet. But one of the major issues with the intermediary language you're talking about is the rigidity of conventionally-applied programming. AI is a bit more flexible and gooey and more able to function as duct tape and bailing wire... but that does come with several tradeoffs of its own that need to be mitigated to some degree.

As a specific example of this, I've been using emacs since about 1997, but it's only this year that I've started down this sort of "integrate everything" road that it promises, because I've never wanted to learn emacs' specific Lisp variant that is used nowhere else. But now with AI I've done more emacs extensions that are providing real value to me than I have in the last... ever. About my previous limit was reifying a keyboard macro into a command, which still isn't really programming emacs.

The ability to indicate "intent" in a way that computers can actually process it as "intent" might change the calculus enough to make this more possible than it used to be in the past.

But those tradeoffs are real. Non-determinism is a problem. Difficulty in knowing what the AI is doing unless you're staring at its output, and even the ability to conceivably monitor an AI's output is only a transient thing we have at the moment, they'll speed up past our ability to monitor them at all soon enough. But it's at least a new area that can be explored and there may be solutions to some problems in there we couldn't reach before.

Looking really cynically, it does seem like it might transition into some kind of cloud-development situation. Your "one access to all development" is one ssh away from our mega 24/7 online compute farm, since we need to keep the sessions persistent. But you can access anything you want!

Hopefully that's not the endgame?

There is value in abstracting the underlying generic systems into more context-relevant primitives for people who just want to get shit done in that context.

Why not meet people where they are at?

If this product isn't valuable, then it will fail. I suspect it will not fail.

This is like the “why not use FTP yourself” response to the YC Dropbox page.

Yes, but good, standardized developer tooling is valuable.

Postman could’ve also been a “just use CURL and dump to JSON” for example.

I actually am bullish on this. Tmux and ssh are somewhat clunky and probably under utilized by most people. I struggle to find commands, navigate sessions, see history, and so on.

Also in an “agent” based world where you might ask an agent to execute things for you, how do you manage that context well beyond just giving it its own Tmux session to drive?

It’s a good idea, and I’m surprised no one’s built it sooner.

The biggest strength to Tmux that a lot of the alternatives lack is you could make a shell script to setup your tmux instance(s) exactly as you would want, so by the time you load into it, its got all the tabs, split terminals, etc where you want them.

The server being their server that you eventually have to a pay a subscription to use? Hmm..

Also, interesting permission problems. Are you going to allow a remote server owned by someone else have shell access to all of your computers? Are they allowed to train an LLM from all the things you type?

Sounds super interesting!

What's your approach?

Can you give me one concrete use case that explains how this is an incredible idea? Thanks

This sounds like given a coherent structure to “I want the prod code here to change the retry policy so pending retries are held in memory in this edge layer instead of the DB writers so load can be shed more cheaply” and it will usher things in from coding, testing, filing tickets, getting prod approvals, using the deploy tools to respond the containers, smoke testing the deploy, looking at Datadog or whatever to ensure the change is having the desired effect. All in one context, the tools and dev/qa/prod systems naturally being orchestrated within the same environment/ session. K8s is just the next to last 100 meters.

I'm curious from the other direction; where do you see overlap between k8s and what's described on the Superlogical website?

So, at least 2 of some dudes.

call me an optimist, but I believe that more and more things are getting more viable to do from the terminal. tools like jj remove the need in Git GUI because mental model and UX gets streamlined, and the AI craze incentivizes practically every company to create their own CLI

The post itself says that the effort is independent of ghostty. Their first work will utilize libghostty, and they'll contribute upstream.

In other words, this probably won't directly involve changes to ghostty.

One of the cofounders is formerly from poolside (poolside.ai), where they produce open-weight coding models. One of their recent models, Laguna, got quite a bit of attention here on HN.

The one they’re currently building. This has to be an agent orchestration and management play.

Would you pay for it to not have data collections and ads?

My take is that it will be tmux as a service. You will have many different sessions which will have their data stored in some central system (read: raft/etcd) and an API built on top of that for create/list/attach to sessions.

Ghostty is the client side aspect of this (the binary that actually runs on your laptop), but there will be a large backend system that the binary talks to through an API. With the API you can then also start automating sessions, for example when a metrics value trips an alert start an agent to triage with a session you can then attach to for further interrogation.

What Im not sure of is if they are going to build on top of kubernetes or not, which already provides a lot of the features necessary (etcd store, API server, workload orchestration, tenant isolation, extendable API).

(This is all speculation)

I have a lot of respect for Mitchellh but taking VC money for a tmux clone is wild.

As long as you are rich all the VCs will throw money at you.. fine.. but if you are rich then you don't have to take the VC money either. Just do it out of passion.

What does tmux alone do way better than the newer tools?

"Any sufficiently VC-funded AI startup will inevitably try to re-invent a log.md file" - Spongebob Squarepants

"With most of them being VC or high-count angel investors, they can combined push whatever product is being built here into 200+ companies independent of whether it is good or not"

Ah, I see you have figured it out the thing that has always been the case.

I think the generous take would be...

- it demonstrates they have funding (pre-empting "no-one will fund something weird like this")

- its transparent and open (pre-empting "which bastards are the ones who really own it?")

> why else would a ~billionaire that could easily fund it himself pull in that many investors that early?

Investors can be good for funding but they can also be good for education. Later, they're also very good for strategy.

> What happened to the good old way of figuring out product market fit first? Is that the only way a product can succeed nowadays

Generally it's very challenging. I have two products I'm building this way and you face a lot of up-hill climb where people with VC backing are getting the door opened for them. Of course, because of those same pressures to deliver and succeed, it's equally possible those same doors will eventually get closed on them.

I think Mitchell also lives in a bit of a bubble now. He went from friendly open source guy to very rich friendly open source guy. That's no knock on him, but I would consider viewing business from the perspective of someone who has been a CEO, CTO, and IC at the same company (his company) as his only job who became insanely wealthy as a result. I have a friend who has largely only worked in big tech, and successful startups, who tells me all the time level and salary don't matter to him. They don't matter to him because he found success early. Again, not a knock but sustained success will shape how you view the world and how you play the game.

> I'm excited to see what Superlogical will build, but the startup scene seems to grow more clique-y by the day.

Insert "It always was" meme

> why else would a ~billionair that could easily fund it himself pull in that many investors that early?

Something something "never spend your own money if you can spend someone else's money." That plus the perception of having those specific investors gives the idea/business fast authority/prestige.

It's not the gangster move, but it's a move.

> why else would a ~billionair that could easily fund it himself pull in that many investors that early

That was my question, too. Does Mitchell even need to take outside funding? I get why you would prefer to use others' poker chips instead of your own, especially if they're happy to give them to you, but usually funding comes with strings attached.

Of course they need VC money! How else are they gonna hire the worlds best vibe coders?

It's really awkward when you consider projects like Bun, which were VC funded, and ultimately bought out by a large company, what would they have even done to pay back the VCs that funded them?

Ghostty and libghostty was product market fit.

Superlogical (allegedly) uses it for the multiplexer.

Mitchell Hashimoto is just the icing on the cake and he has a track record of building defining dev tools used by millions of devs.

So it makes sense why you see 15 angel investors.

More money / investment in the dev tool space is great for everyone!

> Tobias Lutke /../ has most recently advocated for a pay-to-play version of democracy where wealth determines your ability to participate

He’s just looking at america post Citizens United and wishing he could ull the same shenans in Canada. It’s been working very well for the folks and orgs who can afford it

> Tobias Lutke in particular has chronically been on the wrong side of fundamental issues

Just because someone has different ideas about what policies would produce better outcomes for society doesn't mean they are a malicious person.

Clearly there is a lot wrong with society and many suspect it's due to structural problems in how society is organized. Even if he's wrong, it's not surprising at all that a successful well-intentioned person who pays a lot in taxes who is otherwise not experienced in policy would have the opinion that the votes of people who pay more in taxes should have a greater weight.

There is no civilization if it's the norm to assume that someone is acting in bad faith if he has opinions different than yours. E.g. Wikipedia would have fallen apart https://en.wikipedia.org/wiki/Wikipedia:Assume_good_faith

I also recoiled a bit when I say that Lutke was on the investors list.

Does Mitchell even need to take outside funding?

It's really unfortunate that the same traits that make you great at building a business (within capitalism) also often make you have psychopathic tendencies.

People are only allowed to take funding from investors whose politics align with yours?

The vision of having a shared interface for all work seems pretty clear to me.

It’s good to be skeptical, but this guy is the inventor of quite a bit of architecture software supporting modern systems

I think it’ll be fine

It’s a terminal multiplexer. Nothing in this announcement suggests to me that using this multiplexer will require the user to get root or super-root or any additional permissions. It’s what the user already has access to, but more “durable” (which I take a mean terminal session history on by default). And I don’t believe any short-lived secrets like ephemeral keys could have their lifetime extended by a terminal multiplexer. And nothing really supersedes RBAC which any respectable org should be using always.