I really like the bit where he transferred ownership of Ghostty to a non-profit, and is now building this new company on that as an open source dependency:
> We will build on libghostty exactly as it was designed to be used: as a public building block for terminal applications. Superlogical will consume the same MIT-licensed components available to everyone else, and we will continue to upstream shared terminal work so every libghostty consumer can benefit.
Bun and the Rust pre-release have continued to be open source; the pre-release isn’t exclusive to Claude Code but available to anyone to download and use. There hasn’t been any public motion to indicate they may close source anytime soon. This FUD is so easily fact-checkable that I’m genuinely confused at how it caught on
Probably because they said AI rewrote it in 11 days, yet haven't cut a release almost 2 months later. If their AI finished it, where's the release?
They ported enough of it to run a single app (Claude Code). Good job. But if I ported Linux to Rust and it was only capable of running a single app, should I be able say it was a complete port? We'd all agree that was misleading. It's the same with Bun.
Producing a stable release for their community of code that had been written by AI in 11 days and had not yet been extensively tested in the real world would be grossly irresponsible.
Instead, they made the AI rewrite available as a canary release and shipped it for a month as part of Claude Code to millions of their own users.
So... not finished, then? There have been lots of commits between then and now.
If I pay someone to build a house and they left me a voicemail saying "Yep the house is finished, it only took us 11 days!" And then later, "No, you can't move in yet, still fixing some last-minute issues (there's no plumbing yet)." We would classify this as: a lie. And if you told it to your investors: fraud.
Would it really be so bad for Anthropic to admit that a project of this size takes months instead of weeks?
I'm not saying AI is not capable of this. I'm saying that their dishonest marketing has given ammo to their haters.
Hmm, you are right. The 11 days number came from the Bun blog, not the Anthropic blog. Anthropic has a history of making dishonest claims like this (for example their C compiler) so when I saw the news coverage I assumed this was just another one of those. My mistake.
Even then, can you point out where Jarred said this on the Bun blog? Because AFAICT, it seems pretty clear that he said that it took 11 days to get to all integration tests passing, but not to a full release candidate. In his blog post from a few weeks ago, he specifically calls outs:
>Bun v1.4 makes Bun faster, smaller, use less memory and gives the team incredibly powerful tools for systematically improving stability going forward: Rust's borrow checker, Miri (which runs for a growing chunk of code in CI), LeakSanitizer, and 24/7 coverage-guided fuzzing for parsers. There's still more to refactor, but things are off to a great start.
Well i was mostly talking about the "handing over to a non-profit" control parts. Not the release.
but since you said so, could you link me to the release / tag on github? I don't think they've put it out yet?
Maybe they won't change licenses, but Bun now feels more like SQLite as in "source code available" kinda open source, than the "come build with us, community first" kind that i think used to be classified on at least for me mentally.
No horse in this race, just pattern matching as we meatbags love to do.
Idc about the downvoting i can take a bit of heat on my fake internet points
(I'm not a fan of how they reuse that single tag rather than putting out version-numbered alphas, but they do at least update the description to include the commit hash - "canary" is basically a pointer to the most recent commit hash that they have judged worthy of distributing to their canary users.)
This reminds me somewhat of OLE, COM, DCOM, ActiveX, ... [1]. You can paste an Excel chart into a Word document and if you update the data in Excel, the chart in Word will update. You can edit the chart in Word using the Excel UI because you are essentially running an Excel instance inside of Word. You can interact with Excel through an API, no need for a human. But it is painful, you can probably imagine the kind of API you need in order to be able to do all the things you can do with the Excel UI.
The idea is amazing, making this work consistently requires effort, and a lot of it. And as only a few applications invested this effort, it only works with a few applications and never really took off outside of Microsoft. I think this used to work with WordPad and maybe even Notepad but it no longer does, you just get an image instead of the chart object in WordPad and nothing at all in Notepad. It is a dying technology.
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 environment for coding harnesses - more naturally handles spawning and communicating with subagents, managing worktrees, and other SDLC functions, and has a fairly large number of constraints thru bash scripts to ensure better results with less intelligent models
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.
Meta, won't comment on the site, won't click. I'm fed up with this category of click baity titles that are designed to be enigmatic rather than informative.
The guideline do not mention this, apart maybe from:
If the title includes the name of the site, please take it out, because the site name will be displayed after the link.
I'd prefer HN if we made it a habit to editorialize such titles consisting solely of a Domainname or a single word.
that's kinda the point -- titles on HN usually indicate the sort of content that's behind clicking through, and whether it might be something interesting to the prospective reader. this one doesn't because it's the name of the company.
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
They're basically building in one layer what a lot of people are using multiple different apps for currently. Exactly as another commenter said, it's sort of a mix between herdr and proper remote access.
I use pi as my primary harness and then Claude code and Codex (but with the desktop app for some very specific things)
The last two have remote capability with their own app. With pi, you have extensions that can do remote access, but you still end up with three different places to look to get remote access properly from your phone.
People also use tmux and Zellij and any other number of combinations of different apps combined with some kind of remote access, but nothing that's unified and does everything all at once in one place. I feel like that's a good problem to solve if you can do it better.
It doesn't solve the fundamental problem however of inverting the flow from one human to many agents to many agents to one agent to one human.
I feel like the main benefit of such a thing would be centralization. Part of the issue with the terminal ecosystem is precisely that everything's so fragmented and everybody needs to support arcane terminal protocols and wrap a billion wrappers around each other. The easiest way to solve this is to be a definer yourself rather than an integrator, which I think is what this is trying to be. If we could actually have standardized terminal extensions in one place (you'd be surprised how many extension codes there are in terminal land) that has a different semantics not from the 80s, but still captures the terminal experience, that would be great.
The 'cleanest' experience I've got for what I want is mosh + tmux and even that has issues (no block cursor for example).
If superlogical's ultimate goal is to revamp the Linux developer ecosystem, all for it. Am potentially worried long term about these kinds of strategies from a political point of view but as long as things stay open source and stay as local dev tools, its fine.
Invisible until moused over, or scrolled by other means. Barely visible even then. In the wrong place. It may be "just styled", but it's certainly not fine.
Agent orchestrators are the IDE dream repackaged. IDEs never really took off, excepting lock-in/monolopolization (Visual Studio, Xcode). They never lived up to the grand vision of the likes of Alan Kay. IDEs of today are still basically just text editors with a file explorer. Agent orchestrators seem similar. The vision is grand but the implementation is just tabbed terminals.
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.
Do you think the future is more sophisticated agent management, or will tooling increasingly fall away to a single chat session with an executive agent of sorts?
I've spend the last ~20 years inside tmux/vim — and the last month almost exclusively in the Codex desktop app, and increasingly am just doing remote control from the ChatGPT app on my phone.
Ghostty rocks and it brings me no pleasure to say this, but it seems unclear how much longer this world will exist?
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.
Typically, the benefit of the mux over just more windows or tabs is the ability to detach sessions from a term and have them continue to run. But yea, the “split a term window into multiple terms” doesn’t have quite as much value in a world of lots of tabs. That said, you can do it under program control on the remote host, so it does save remote login overhead, etc. But for local, less value.
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.
To make it more relevant: right now I mostly rely on tmux. Before that I’ve been a long time user of screen (irssi over ssh people know this play). Looking forward to this.
These comments are absurd about their investors. I’m sure if their investor list included socialist nobody on HN would bat an eye. Ahh that’s right, socialist are “takers” not “makers”. Age old standing on the pedestal of moral superiority.
why are you bringing your socialist boogeymen into this. what's absurd is that this is somehow louder for you than the real, life-changing actions of the genocidal maniac.
in this case mitchell is an exited founder with a net worth of over $1B so he can be choosier than almost anyone in this regard. Deeply disappointing he is taking money from Guillermo
things look different from Billionaire Island than they do to us plebs down here. it's crazy to expect such people to do the "right" thing, they're too far gone.
https://en.wikipedia.org/wiki/Tax_returns_of_Donald_Trump
"Trump paid no net federal income taxes in 11 of 18 years of the past two decades. After the refund, Trump had an average tax bill of $1.4 million per year over the 18 years.[52][176] In 2016, Trump paid only $750 in federal income tax, and in 2017, he paid another $750 in federal income tax."
If there was an unequal vote, I am pretty sure rich people would want votes as heavy as possible.
If you read it again you will notice that technically he thinks that, to satisfy those who believe the wealthy don't pay tax, Canada could use a system to encourage the wealthy to claim income, where that is incentivized by being able to vote. Often the wealthy have no need for income as they already have so much wealth and can easily borrow against that wealth as cash needs present themselves. It is believed (at least as far as the thread went) that the wealthy currently don't pay tax, and thus would be unable to vote, or their vote would count for little, under the proposed since they don't have sufficient income to qualify.
That would be true if he advocated for votes according to effective taxation rate or something similar. He caps his example at 800k reported income, if I remember, which is a rounding error to his income class.
Instead, he makes sure to include the idea of stripping low-income people of all voting rights.
Not at all. The idea is that his class claim very little income, if any income. His income class is ~$0 (or believed to be, at least). You can certainly make an argument that the claim isn't true, that the ultra-wealthy are actually already paying huge sums in income tax, but then you cannot take the thread out of context. It was clearly directed at those who believe the wealthy pay no income tax. It literally says so. Did you skip step one and not go back to read it again?
> He caps his example at 800k reported income, if I remember
You are correct, approximately, but there would have been no need to remember given step one, so that was clearly ignored. Why? What is the purpose of replying without even reading the comment you are replying to?
Lutke's own post was about pension = no votes, which I identify as the thing he decided to actually spent time on. His other tweet is a Musk-like "Good idea" reply to someone else's proposal regarding tax brackets, no?
I don't think memory fails me on the main points. If you think ranking the same read facts differently from you means faulty reading, I don't really owe you the labor of explaining where that fails.
Why would they care? The ultra-wealthy impact election through donations, which is many order of magnitudes more impactful than their personal ballot, even if you were to give them 10x weight.
Elon Musk impact on the U.S. election wasn't because he drove to the booth.
Care about what? The thread was clearly making fun of those who think the wealthy don't pay any income tax. It literally states that the plan is proposed for those who complain the wealthy don't pay tax, not a plan to be set in motion. Tobi, who fits the wealthy profile of who supposedly doesn't pay tax better than almost anyone, will know well that he already does pay income tax, hence why he humorously threw salt in the wounds of those who don't have a solid grasp of reality.
Being wealthy doesn't mean you can't be a troll. You'd think after seeing Musk in action that would have already been obvious to all, but I guess not.
About what you yourself said? "Canada could use a system to encourage the wealthy to claim income, where that is incentivized by being able to vote".
The ultra-wealthy can already influence elections in very significant ways. Getting 5 votes instead of 1, which makes no actual difference to any election, is not something that will suddenly make the ultra wealthy wanting to pay more income tax.
What does make a difference though, is if you can prevent a lot of lower income voters from casting their ballots.
Also, no, I'm not buying your unfounded claim that Tobi is just joking bro. His doubling down on it in other comments + his past make it clear to me that is seriously thinks like this.
> I'm not buying your unfounded claim that Tobi is just joking bro.
Fair enough. Let me be a little more precise here: Eric Thor was clearly poking the bear of those who believe that the wealthy don't pay tax. I mean, his voting suggestion concludes with exactly that. Maybe Tobi is so far out to lunch that he didn't notice and took it as an actual plan, but, then again, if you are suggesting that he is that far out to lunch what are we to do with that?
> His doubling down on it in other comments
The fact that he was writing comments about it suggests that he doesn't take it too seriously. Nobody of sound mind wastes time writing stupid comments on the internet if they have an actual motive. They go out into the real world and make it happen.
If he is completely out to lunch like you suggest then anything is possible, but the more likely explanation still seems to be making fun of those who don't believe the wealthy pay tax. If people were making shit up about you, claiming that you didn't pay any tax, you'd likely end up making fun of them too.
> Nobody of sound mind wastes time writing stupid comments on the internet if they have an actual motive.
I mean Musk? You even used it as an example of "just trolling" in your previous message. He invested north of $300M+ in the last presidential election, also injected massive amounts in various local races, as well as foreign elections.
What more do you need to take them seriously? Even if somehow Musk was doing all of this to troll people because to him $500M is nothing, the impact on those elections is real, so it doesn't even matter what is internal reasons are.
One who is not normally considered to be of sound mind, but you do you.
> What more do you need to take them seriously?
A good idea? The voting thought, while possibly an interesting thought experiment, is not suitable for execution. Why would I take it seriously?
> the impact on those elections is real
Maybe you aren't asking why I would take them seriously, when there was nothing presented that is worth taking seriously, but why do others take their hobo shouting on the street ramblings seriously, presumably only because they have wealth? The other hobos on the street seem to be happily ignored.
Good question, but one I do not have the answer to. I'm still trying to figure out how these particular hobos shouting nonsense on the street corner became people I am aware of. They're just people, same as all the others that manage to remain anonymous. Definitely some kind of malfunction in my brain going on there.
I found this just when I googled his name plus controversy
> Shopify CEO Tobias Lütke faced fierce online backlash in July 2026 after endorsing a social media proposal for a tax-tiered voting system that would strip voting rights from non-taxpayers and give multiple votes to high-income earners.
but if we go as far as to say
> If it’s touched by Tobias Lütke, I’m out.
does it mean control or sponsored, for example ladybird web browser...
That vague posting is a bit frustrating to be honest.
Are you talking about the fact they do not accept pull requests? If yes, they had good reasons for doing so, LLMs make it really time consuming and exhausting to manage an open source community, I completely understand why they decided to stop accepting PRs for now
Along with a myriad of different controversies. Just searching his name will bring up many articles about his chronically online Elon Musk wannabe takes.
He came to Canada as an immigrant, made some buck and now unashamedly calls 'idiots' the very people that built the communities where he lives. The guy is a complete POS.
Some people believe that anyone who does not maintain 100% ideological purity and goodness, which can be measured by if they align 100% with the persons own and therefore morally good political ideas, which also coincidentally matches the status quo of Brooklyn based journalists, that the heretic must be ostracized from all of society and it is impossible to work with them. Since Tobi Lutke made a mortal sin of disagreeing on suffrage requirements, they should be cast into fire and ostracized.
I don't think "citizens have a right to vote" is anywhere near the realm of ideological purity or "Brooklyn based journalist" thinking. It's literally just democracy.
Freedom of speech seems like something Lutke would pretend to like, and him getting torched for his anti-human opinions is a great demonstration of that principle.
> Lutke made a mortal sin of disagreeing on suffrage requirements
I think this is a very tame and apologetic and dishonest way to put it. Suffrage and voting is how people have a voice in the future of their democracy. He advocates for a policy that materially removes that right, that say in the future, from those who already have little power. I don’t know how else to emphasize that it’s more than “they disagree a bit with him and now he’s an impure sinner”, it’s more akin to “this rich dude wants to rewrite the rules to give himself more power and us less”, which is a sober assessment of his policy. I think people actively trying to avoid giving him any more money or power is a very understandable reaction to that
I think ultimately some parts of ideology are bound to and by the reality we live in, and acknowledging that reality gives a good answer here for when _it seems like_ people freaking out at an “impure idea” - is there a documented harm that we can look at?
Many people boycott Elon and his products. Do you raise this same “moral purity” complaint about them? Does it ever go in the other direction?
I think technology and how we build it is deeply tied to us as social, communal beings. Tech can be a reflection of our capabilities, our values. Tech shapes our reality in turn. It’s worth thinking about, being critical of, and believing people when they say what they believe
> Location Though the team has spent many years scaling remote-first companies and is currently remote, we are hiring specifically in Los Angeles, London, and New York, with an expectation for a flexible in-office schedule as we build our presence in each region.
I always get a "lifestyle business" vibe when tech companies are based in LA or NY. When you're a seasoned veteran you'd prefer to live in a larger, more fun city, even though tech is concentrated in the Bay Area
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?
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.
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.)
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.
That's not the original meaning of a "lifestyle business" anyways, which is more like a business that makes enough (to its owners, nothing to do with the employees) for them to maintain their desired lifestyle, like maybe a few million in revenue or whatever, but can't / won't be scaled beyond that because their lifestyle is already well supported by the current business growth / volume / revenue so they have no interest on blitzscaling, taking VC money, and running the perpetual growth treadmill or because the business is already topped-up and won't support further scaling (its not always a choice)
Yes, indeed, although i think in this case it was just signalling how not being "SF focused" might be seen as a lack of ambition or whatever in such VC contexts or something and thus the exaggerated comparison to it being a lifestyle business merely due to the location requirements on ny/la (nicer cities to live in, maybe, than SF, or whatever)
i think just michael is based on LA, so he prefers having the offices near his home (has kids, etc), and has enough FU money to not need to let the VC's dictate where the HQ's are located at.
I guess also they dont love the idea of managing people remotely for a small startup/first hires, although im sure for the right person they wouldn't really care (or usually this is how it goes on most small companies/startups)
Really exciting IMO. Many people are working on new work interfaces (e.g. Codex/Claude/Cursor, multiplayer agent spaces, unified Slack/Notion/Zoom, etc.), and I think the Superlogical vision makes sense as a competitor in the space, at least to try.
There’s clearly room to improve our existing tools.
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-specific -- if somebody put a small fraction of the effort into it that others are putting into superlogical and the other alternatives listed here, it could be awesome....
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.
Devs don't use Dropbox, they use git. Git is probably the poster-child example of an unfriendly interface, and yet it's the standard way for devs to share files.
Emacs is downright friendly compared to git & tmux IMO.
I love how ghostty does windows/tabs/splits, but ultimately, I don't find myself using them. I find that I only have so much muscle memory in my hands, and I need to stick with one solution that works everywhere. A multiplexer let's me do this on my own machine, or if I'm SSH'd in elsewhere. Personally I use Zellij, because tmux feels like learning another vim. But I do use tmux when using tmate, because it's the simplest way I've found to share terminals.
Anyway, I'm very happy to see someone with Michell's talent entering this space. We can use more options with sane defaults and great performance.
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.
Anybody else using Superset as your multiplexer rn? I'm finding it quite useful but wondering if there are other existing products that I should consider -- Mac Silicon: https://superset.sh/ Curious to see what Superlogical cooks up.
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?
Maybe I just missed the boat on hashicorp and similar solutioning, but I find a lot of the solutions for working with multiple systems using an intermediary shared language, such as terraform, end up obfuscating the problems they intend to solve and result in people who deal with and support those systems on a daily basis having very little understanding of how to interact and troubleshoot the core infra that is being built up with those tools.
Yes, building those solutions requires a lot of expertise, but I've never found the tools to be worth engaging with over just using what the standard platform tools they abstract away were replacing. Then you add in the economic rug-pull scenarios that have happened when utilizing software like this and I wonder why everyone hops on the bandwagon so easily.
What is the company they're building trying to solve that system maintainers can't already do? Isn't this just rebuilding OS windowing?
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!
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.
Funny. I started working on a similar product myself just a few days ago. I have multiple machines for multiple different purposes. Each one runs different agents (some local, some cloud), for different tasks. Different ones store different data (a NAS on one, my VMs on another, etc.). I wanted a durable layer over all of them to be able to start work on any one of them and resume on any other, and for all the disparate agents to share memory/configurations where possible, and for each node in the network to make all the others aware of its purpose (so when the NAS is online, the agents know how to connect to it and what it contains). Sadly no VC funding for me, building it on my lonesome :)
Still exploring different options but at the moment I run a daemon on each node that connects via iroh to link the nodes (and NAT traversal). Once the link is established the nodes share data via a Raft-based datastore. I can launch a TUI client on any machine. This presents a chat interface like a typical LLM harness. From here I can configure which model I want to use, how the model is executed and on which node (for example I can use Ollama on machine A to serve a model that’s running on machine B, but executing commands via this system on machine C). Each daemon then acts like a typical LLM harness, but headless. So the commands and tool calls are streamed into the shared store and finally end up in my TUI. Since all the state is in the distributed store I can pick up the progress wherever I want, as long as there’s a node running there.
Still a lot of big problems like latency so I’m not too happy with the current design. It’s a LOT of state to replicate, and obviously things get problematic when an agent was working on a project on machine A but then ends up on machine B which doesn’t have that project directory around at all. So the agent needs to re-orient itself to figure out what happened. Still pretty fun to play with.
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?
This is an incredible idea. I would usually be skeptical that such an engineering-focused tool would "make it," but Mitchell Hashimoto is one of the very few people I think can pull it off.
This is going to change so many things if it starts getting adopted.
That explains why he built sessions into libghostty instead of Ghostty itself. Looking forward to it.
I spend most of my time in the terminal and have been using tmux to manage my sessions and workspaces, mainly via sesh.
But recently, I decided to move to Kitty and use its built-in sessions instead. I built something around it as well. It’s still early, but I’m really liking this workflow so far.
I love Ghostty. But I miss out on lots of its features because I use zellij exclusively to work with agents. This sounds exciting and I hope they'll choose some cliff oss license like gitbutler does
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.
Doesn't K8s already practically provide a layer of abstraction across all infrastructure? You might not think of it as being a session, but to the degree that it's responsible for managing scaling and failover, it kind of is.
This isn't a statement of value, I'm genuinely curious myself. In fact, I bet this new thing will probably be better.
> For us, that means interactive work, automatic work, and production work would share one well-crafted underlying system instead of living in separate tools.
Unless this is only for the narrow band of purely local work, I would be surprised if k8s didn't play a big role in the backend solution. It is the standard "well-crafted underlying system" for automatic production work. But, it isn't strong in the interactive-work/frontend aspect or other aspects, so it certainly wouldn't be the main focus of the final product.
Random example: you want to multiplex some work across multiple (self-hosted) agents, using niche fine-tuned models. It would make good sense if those agents were automatically provisioned and deprovisioned (and isolated to your subtenant), and k8s is a great underlying system for that sort of thing.
This sounds awesome enough to see if they're hiring, and they are, but unfortunately only in certain areas with an expectation to be in-office with some regularity. Bummer. Exciting to see this team together, though. This is a tool that I want and think about daily.
It’s interesting because in an ideal world he is practically building a NixOS config.
I’m pretty new to NixOS (a few months) but i could envision using Nix to declarativey configure tmux windows and panes in a dev shell. That said, Nix feels like it might still be too clunky at this point in time to quickly re-declare window and panel arrangements. But I never tried and it could be possible.
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
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.
> We will build a high-quality terminal multiplexer that remains excellent at that job, even as it grows to support the second and third parts of the plan.
why this feels the next step is data collection 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).
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.
I'm sorry to take it out on this specific announcement, as it just seems to be one of many "offenders", but why are the 15 investors listed on an announcement blog post of a semi-stealth startup?
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. What happened to the good old way of figuring out product market fit first? Is that the only way a product can succeed nowadays, or why else would a ~billionair that could easily fund it himself pull in that many investors that early?
---
I'm excited to see what Superlogical will build, but the startup scene seems to grow more clique-y by the day.
"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.
> 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.
> 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.
> 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.
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?
I'm not doubting that Mitchell Hashimoto is capable of building great products with great product market fit. Quite the opposite, as the few strands of idolization that exist within me rank him as one of the best to ever do it in the dev tooling space.
I'm also not doubting that there would be tons of investors throwing themselves at him for a new venture like this.
But why, apart from forced market access would it be in his interest to take that money, given that he can nicely fund it for a few years with ~1% of his net worth if not for forced market access?
It said they are funded, but it doesn't mean they lost control or have to be VC hyper growth. Apart from Notable Capital and Amplify Partners all others are individuals. There are lots of benefits having these group of company founders and CEO as investors. They are likely interested in the problem space as well.
Had it been hyper growth startup you would be looking at a16z or other bigger names.
1. Accountability - even very talented people can lose motivation or get distracted. Having investors to report to helps with this.
2. Recruiting - many people would be wary of working on a rich guy's pet project. Taking investment makes it more of a real company that has to deliver something and is not subject solely to the whims of one person.
I really wish mitchellh wouldn't take funding from some of these people. Tobias Lutke in particular has chronically been on the wrong side of fundamental issues, and has most recently advocated for a pay-to-play version of democracy where wealth determines your ability to participate (and in particular your "stake" in the democracy).
> 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.
I'm sure the people who would have their right to vote stripped from them would consider that malicious, and they would be correct. That's not democracy.
> I'm sure the people who would have their right to vote stripped from them would consider that malicious, and they would be correct.
I don't agree with him but you're incorrect here. To play devil's advocate, if his policy idea resulted in better outcomes for them, e.g. better material conditions, then they may end up preferring not having a right to vote and consider him a good person.
> That's not democracy.
Democracy isn't necessarily the same thing as direct democracy, or "everyone votes." The Ancient Greeks didn't practice democracy this way. In any case, democracy isn't a virtue in itself. If one system is democratic according to your standards but the vast majority of its citizens live in squalor while another system is not democratic according to your standards but the vast majority of its citizens live in prosperity, which system is better?
Historical precedent (both recent and not) shows that concentrating wealth and then giving those wealthy few an inordinate amount of political power does not end well for those disenfranchised by this system.
It is not reasonable to give the benefit of the doubt on the basis of Lutke and his ilk maybe being benevolent dictators if given largely unfettered access to the levers of power.
> Historical precedent (both recent and not) shows that concentrating wealth and then giving those wealthy few an inordinate amount of political power does not end well for those disenfranchised by this system.
Historical precedent isn't a rational basis for argument. You don't make stock picks based on precedent. You need to actually examine the causes that result in outcomes to evaluate whether the same outcome would happen again. That said, I have seen good arguments for why weighting votes based on tax contribution would have bad outcomes so I don't agree with him.
> It is not reasonable to give the benefit of the doubt on the basis of Lutke and his ilk maybe being benevolent dictators if given largely unfettered access to the levers of power.
I have no reason to assume Lutke is a bad actor but maybe you know something I don't. Has he done anything to harm anybody or any group of people? Him being rich doesn't make him automatically a bad or suspicious person. There are many rich people who do good things with their wealth and free time.
I think most people will agree that empowering the edgelord Shopify CEO is not going to improve most people's outcomes. All you need to do is read his tweets to see that he should have less power (much less), not more.
It's not a logical fallacy. You made the unsubstantiated claim that people would want their right to vote stripped from them. I do not think people want guys like this in power: https://xcancel.com/tobi/status/2081155797640171641
I'd start with some German guy calling native Canadians who are actively civically engaged "useful idiots" and his suggestion that the unemployed and retried shouldn't have a voice in their community.
> his suggestion that the unemployed and retried shouldn't have a voice in their community.
I don't see him making that suggestion. It's pithy so it's hard to really pin down a meaning to the tweet but I think he's saying that these town sessions aren't a meaningful expression of democracy because only a specific section of society is in attendance and one that may not understand key issues relevant to working or young people.
In this context, I would call it pithy sure. It's concise and direct. He's not explaining further.
He clearly isn't calling the other categories he listed (the "
Retired, Unemployed, Unemployable") useful idiots, that's why he listed "useful idiots" separately. I don't know to whom he is referring as useful idiots but I don't think it's immoral in itself to use the term "useful idiot."
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.
Easy to say this when the original commenter incorrectly described Tobias' comments. He suggested that those who pay more in taxes should have a greater say in how the government is spending the money. This is not a crazy idea on its face; you could say that taken to the limit it would not be a good system to live in (the highest taxpayer dictating what gov't does).
But it does make sense that those who are footing the bill for everything might be consulted a little on the plans to spend.
reading "those who pay more in taxes should have a greater say in how the government is spending the money" and NOT taking that to mean what the OP is suggesting it to mean is naive. Everyone knows what Tobias meant when he said that.
"I think this guy is dumb and wrong. Oh no, this dumb wrong guy is going to have outsized influence on a project I'm interested in. Oh no, this dumb wrong guy may profit from a project I'm interested in. Oh no, this dumb wrong guy is using his wealth to do dumb wrong things."
It sounds like they're building a super-root which can destroy your whole infrastructure at once ???
From their website:
"Building and operating software today spans local machines, remote hosts, sandboxes, services, and production systems. It has many modes of operation: interactively with a human developer, automatically through CI and background processes, and increasingly through agents working in parallel.
[...]
We believe the missing layer is a durable session around the work itself: one that can span applications and environments, provide relevant context by default, expose structured data and actions, preserve history, and be driven by software while remaining visible and controllable by people.
[...]
This is our plan to build a multiplexer for all work:
1 Build an incredible multiplexer.
2 Make everything in it composable.
3 Make it safe and operable in production"
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.
I really like the bit where he transferred ownership of Ghostty to a non-profit, and is now building this new company on that as an open source dependency:
> We will build on libghostty exactly as it was designed to be used: as a public building block for terminal applications. Superlogical will consume the same MIT-licensed components available to everyone else, and we will continue to upstream shared terminal work so every libghostty consumer can benefit.
Can't unsee the stark contrast with Anthropic/Bun's recent shenanigans over unreleased rust migrations hehe
Bun and the Rust pre-release have continued to be open source; the pre-release isn’t exclusive to Claude Code but available to anyone to download and use. There hasn’t been any public motion to indicate they may close source anytime soon. This FUD is so easily fact-checkable that I’m genuinely confused at how it caught on
HN discussions are filled with negative-sentiment AI comments that quickly get upvoted regardless of their connection to reality.
> I’m genuinely confused at how it caught on
Probably because they said AI rewrote it in 11 days, yet haven't cut a release almost 2 months later. If their AI finished it, where's the release?
They ported enough of it to run a single app (Claude Code). Good job. But if I ported Linux to Rust and it was only capable of running a single app, should I be able say it was a complete port? We'd all agree that was misleading. It's the same with Bun.
The marketing lies simply backfired.
> If their AI finished it, where's the release?
`bun upgrade --canary`. You can use it today.
Keeping a large release, like a total rewrite, on canary for a while to iron out bugs is a very reasonable software engineering practice.
> If their AI finished it, where's the release?
Producing a stable release for their community of code that had been written by AI in 11 days and had not yet been extensively tested in the real world would be grossly irresponsible.
Instead, they made the AI rewrite available as a canary release and shipped it for a month as part of Claude Code to millions of their own users.
So... not finished, then? There have been lots of commits between then and now.
If I pay someone to build a house and they left me a voicemail saying "Yep the house is finished, it only took us 11 days!" And then later, "No, you can't move in yet, still fixing some last-minute issues (there's no plumbing yet)." We would classify this as: a lie. And if you told it to your investors: fraud.
Would it really be so bad for Anthropic to admit that a project of this size takes months instead of weeks?
I'm not saying AI is not capable of this. I'm saying that their dishonest marketing has given ammo to their haters.
> 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.
Would you release it when it's not complete?
> Would you release it when it's not complete?
Nope. I also wouldn't lie and say it's complete when it isn't.
Can you provide a source where they said it was complete? I can't seem to find any evidence of this.
Hmm, you are right. The 11 days number came from the Bun blog, not the Anthropic blog. Anthropic has a history of making dishonest claims like this (for example their C compiler) so when I saw the news coverage I assumed this was just another one of those. My mistake.
Even then, can you point out where Jarred said this on the Bun blog? Because AFAICT, it seems pretty clear that he said that it took 11 days to get to all integration tests passing, but not to a full release candidate. In his blog post from a few weeks ago, he specifically calls outs:
>Bun v1.4 makes Bun faster, smaller, use less memory and gives the team incredibly powerful tools for systematically improving stability going forward: Rust's borrow checker, Miri (which runs for a growing chunk of code in CI), LeakSanitizer, and 24/7 coverage-guided fuzzing for parsers. There's still more to refactor, but things are off to a great start.
Perhaps you never perused beyond the HN headline?
Well i was mostly talking about the "handing over to a non-profit" control parts. Not the release.
but since you said so, could you link me to the release / tag on github? I don't think they've put it out yet?
Maybe they won't change licenses, but Bun now feels more like SQLite as in "source code available" kinda open source, than the "come build with us, community first" kind that i think used to be classified on at least for me mentally.
No horse in this race, just pattern matching as we meatbags love to do.
Idc about the downvoting i can take a bit of heat on my fake internet points
> could you link me to the release / tag on github?
https://github.com/oven-sh/bun/releases/tag/canary
(I'm not a fan of how they reuse that single tag rather than putting out version-numbered alphas, but they do at least update the description to include the commit hash - "canary" is basically a pointer to the most recent commit hash that they have judged worthy of distributing to their canary users.)
They inflamed an online programming language fandom. It's kinda like if they dissed BTS
This reminds me somewhat of OLE, COM, DCOM, ActiveX, ... [1]. You can paste an Excel chart into a Word document and if you update the data in Excel, the chart in Word will update. You can edit the chart in Word using the Excel UI because you are essentially running an Excel instance inside of Word. You can interact with Excel through an API, no need for a human. But it is painful, you can probably imagine the kind of API you need in order to be able to do all the things you can do with the Excel UI.
The idea is amazing, making this work consistently requires effort, and a lot of it. And as only a few applications invested this effort, it only works with a few applications and never really took off outside of Microsoft. I think this used to work with WordPad and maybe even Notepad but it no longer does, you just get an image instead of the chart object in WordPad and nothing at all in Notepad. It is a dying technology.
[1] https://en.wikipedia.org/wiki/Component_Object_Model
I think you may have meant to comment on another post :)
No, correct post. But maybe I am misunderstanding the idea behind the post, that is certainly possible.
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 environment for coding harnesses - more naturally handles spawning and communicating with subagents, managing worktrees, and other SDLC functions, and has a fairly large number of constraints thru bash scripts to ensure better results with less intelligent models
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
Meta, won't comment on the site, won't click. I'm fed up with this category of click baity titles that are designed to be enigmatic rather than informative.
The guideline do not mention this, apart maybe from:
I'd prefer HN if we made it a habit to editorialize such titles consisting solely of a Domainname or a single word.One of the worst I've seen was title editorialized from original "Someone: something [x.com]" to just "Something [x.com]".
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
Without clicking you’d never know why it’s the title.
that's kinda the point -- titles on HN usually indicate the sort of content that's behind clicking through, and whether it might be something interesting to the prospective reader. this one doesn't because it's the name of the company.
Wow, that "We're hiring/Career Page" link is geeksmaxxing.
First time seeing `ssh superlogical.jobs` would be a way to show job descriptions in terminal!
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.
"I completed the Synacor challenge" is still a good way to get experienced CTOs' attention.
I actually heard of that.
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
That’s probably not the type of connection they’re looking for employees to have.
I'm on holiday.
But could also be a good software tester if given any kind of remote connection support feature.
It’s probably a good filter too for applicants
Yes, prefilter for sure.
They're basically building in one layer what a lot of people are using multiple different apps for currently. Exactly as another commenter said, it's sort of a mix between herdr and proper remote access.
I use pi as my primary harness and then Claude code and Codex (but with the desktop app for some very specific things)
The last two have remote capability with their own app. With pi, you have extensions that can do remote access, but you still end up with three different places to look to get remote access properly from your phone.
People also use tmux and Zellij and any other number of combinations of different apps combined with some kind of remote access, but nothing that's unified and does everything all at once in one place. I feel like that's a good problem to solve if you can do it better.
It doesn't solve the fundamental problem however of inverting the flow from one human to many agents to many agents to one agent to one human.
I feel like the main benefit of such a thing would be centralization. Part of the issue with the terminal ecosystem is precisely that everything's so fragmented and everybody needs to support arcane terminal protocols and wrap a billion wrappers around each other. The easiest way to solve this is to be a definer yourself rather than an integrator, which I think is what this is trying to be. If we could actually have standardized terminal extensions in one place (you'd be surprised how many extension codes there are in terminal land) that has a different semantics not from the 80s, but still captures the terminal experience, that would be great.
The 'cleanest' experience I've got for what I want is mosh + tmux and even that has issues (no block cursor for example).
If superlogical's ultimate goal is to revamp the Linux developer ecosystem, all for it. Am potentially worried long term about these kinds of strategies from a political point of view but as long as things stay open source and stay as local dev tools, its fine.
Right, but I'm pretty sure this is not the first time somebody thought about solving it.
https://xkcd.com/927/
Flashback to when Mitchell first introduce Vagrant [1] to HN, which of course was the how everything from HashiCorp started.
I thought it was't that long ago, may be 8, or at best 10-12 years.
Turns out 16 years has passed.
Curious if it would be another project using Zig.
[1] https://news.ycombinator.com/item?id=1175901
They hire for Go.
Considering it's going to consume libghostty, I'm betting it'll be Zig.
The application layer is in Swift at least.
libghostty exports a C ABI. For an application from Mitch at this layer I would guess Go.
Mitch said he doesn't like go since they added generics.
I checked the available jobs and they are hiring Typescript for web, Go for infrastructure,systems, Objective-C, swift for Apple
I dig the ssh jobs page, UX is very good.
> We're also addressing the most common papercuts of existing tools, such as making scrollback, selection, and scrolling all work natively.
Goes ahead and butchers the browser scrollbar.
Scrollbar is fine for me? Just styled
Invisible until moused over, or scrolled by other means. Barely visible even then. In the wrong place. It may be "just styled", but it's certainly not fine.
On iOS the whole page jumps up and down when you try to grab it.
Agent orchestrators are the IDE dream repackaged. IDEs never really took off, excepting lock-in/monolopolization (Visual Studio, Xcode). They never lived up to the grand vision of the likes of Alan Kay. IDEs of today are still basically just text editors with a file explorer. Agent orchestrators seem similar. The vision is grand but the implementation is just tabbed terminals.
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.
Do you think the future is more sophisticated agent management, or will tooling increasingly fall away to a single chat session with an executive agent of sorts?
I expect the product from this company to resemble a combination of Orca + zmx
[0]: https://github.com/stablyai/orca [1]: https://github.com/neurosnap/zmx
Somewhat related, I recently heard of fentas/atty from:
https://erock.prose.sh/posix-shell-is-all-you-need
I've spend the last ~20 years inside tmux/vim — and the last month almost exclusively in the Codex desktop app, and increasingly am just doing remote control from the ChatGPT app on my phone.
Ghostty rocks and it brings me no pleasure to say this, but it seems unclear how much longer this world will exist?
This doesn't seem like a niche that isn't filled. Am I missing out on something by still just using iTerm2 and tmux?
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.
Typically, the benefit of the mux over just more windows or tabs is the ability to detach sessions from a term and have them continue to run. But yea, the “split a term window into multiple terms” doesn’t have quite as much value in a world of lots of tabs. That said, you can do it under program control on the remote host, so it does save remote login overhead, etc. But for local, less value.
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.
Multibuffer program like Emacs but not Emacs. All for it.
If its got Mitchell, I’m in.
To make it more relevant: right now I mostly rely on tmux. Before that I’ve been a long time user of screen (irssi over ssh people know this play). Looking forward to this.
If it’s touched by Tobias Lütke, I’m out.
Yep, a lot of toxic investors in this. Not touching it with a ten foot pole and lost a lot of respect for Mitchell. A couple of the other investors:
* https://x.com/rauchg/status/1972669025525158031
* https://x.com/search?q=from:lennypruss%20israel&src=typed_qu...
Just linking to a login nag.
Yuck.
These comments are absurd about their investors. I’m sure if their investor list included socialist nobody on HN would bat an eye. Ahh that’s right, socialist are “takers” not “makers”. Age old standing on the pedestal of moral superiority.
You may be surprised to hear that the famously socialist paulg might also have a problem with this.
why are you bringing your socialist boogeymen into this. what's absurd is that this is somehow louder for you than the real, life-changing actions of the genocidal maniac.
VCs tend to be a toxic bunch. Hard to avoid that in our business
in this case mitchell is an exited founder with a net worth of over $1B so he can be choosier than almost anyone in this regard. Deeply disappointing he is taking money from Guillermo
things look different from Billionaire Island than they do to us plebs down here. it's crazy to expect such people to do the "right" thing, they're too far gone.
more info please!
He thinks ability to vote (and the number of votes you get) should be based on income.
https://vancouverisland.ctvnews.ca/business/article/shopify-...
Taxes paid was actually what was said.
I may be going astray, but I can see a way to degenerate "taxes paid" to "taxes before deductions" to "income since it relates to taxes anyways"
It doesn't have to be that complicated. Just absolute terms could have the desired effect.
- 1% tax of 10 million = 100k
- 20% tax of 100k = 20k
Hence, the rich pays more tax.
https://en.wikipedia.org/wiki/Tax_returns_of_Donald_Trump "Trump paid no net federal income taxes in 11 of 18 years of the past two decades. After the refund, Trump had an average tax bill of $1.4 million per year over the 18 years.[52][176] In 2016, Trump paid only $750 in federal income tax, and in 2017, he paid another $750 in federal income tax."
If there was an unequal vote, I am pretty sure rich people would want votes as heavy as possible.
Maybe worth mentioning that the founding fathers in the US believed only property owners should vote.
Let's make everyone a property owner by default at birth
Not the most important(IMO) of our founding fathers, Thomas Paine.
If you read it again you will notice that technically he thinks that, to satisfy those who believe the wealthy don't pay tax, Canada could use a system to encourage the wealthy to claim income, where that is incentivized by being able to vote. Often the wealthy have no need for income as they already have so much wealth and can easily borrow against that wealth as cash needs present themselves. It is believed (at least as far as the thread went) that the wealthy currently don't pay tax, and thus would be unable to vote, or their vote would count for little, under the proposed since they don't have sufficient income to qualify.
That would be true if he advocated for votes according to effective taxation rate or something similar. He caps his example at 800k reported income, if I remember, which is a rounding error to his income class.
Instead, he makes sure to include the idea of stripping low-income people of all voting rights.
> which is a rounding error to his income class.
Not at all. The idea is that his class claim very little income, if any income. His income class is ~$0 (or believed to be, at least). You can certainly make an argument that the claim isn't true, that the ultra-wealthy are actually already paying huge sums in income tax, but then you cannot take the thread out of context. It was clearly directed at those who believe the wealthy pay no income tax. It literally says so. Did you skip step one and not go back to read it again?
> He caps his example at 800k reported income, if I remember
You are correct, approximately, but there would have been no need to remember given step one, so that was clearly ignored. Why? What is the purpose of replying without even reading the comment you are replying to?
Lutke's own post was about pension = no votes, which I identify as the thing he decided to actually spent time on. His other tweet is a Musk-like "Good idea" reply to someone else's proposal regarding tax brackets, no?
I don't think memory fails me on the main points. If you think ranking the same read facts differently from you means faulty reading, I don't really owe you the labor of explaining where that fails.
Why would they care? The ultra-wealthy impact election through donations, which is many order of magnitudes more impactful than their personal ballot, even if you were to give them 10x weight. Elon Musk impact on the U.S. election wasn't because he drove to the booth.
Care about what? The thread was clearly making fun of those who think the wealthy don't pay any income tax. It literally states that the plan is proposed for those who complain the wealthy don't pay tax, not a plan to be set in motion. Tobi, who fits the wealthy profile of who supposedly doesn't pay tax better than almost anyone, will know well that he already does pay income tax, hence why he humorously threw salt in the wounds of those who don't have a solid grasp of reality.
Being wealthy doesn't mean you can't be a troll. You'd think after seeing Musk in action that would have already been obvious to all, but I guess not.
> Care about what?
About what you yourself said? "Canada could use a system to encourage the wealthy to claim income, where that is incentivized by being able to vote".
The ultra-wealthy can already influence elections in very significant ways. Getting 5 votes instead of 1, which makes no actual difference to any election, is not something that will suddenly make the ultra wealthy wanting to pay more income tax.
What does make a difference though, is if you can prevent a lot of lower income voters from casting their ballots.
Also, no, I'm not buying your unfounded claim that Tobi is just joking bro. His doubling down on it in other comments + his past make it clear to me that is seriously thinks like this.
> I'm not buying your unfounded claim that Tobi is just joking bro.
Fair enough. Let me be a little more precise here: Eric Thor was clearly poking the bear of those who believe that the wealthy don't pay tax. I mean, his voting suggestion concludes with exactly that. Maybe Tobi is so far out to lunch that he didn't notice and took it as an actual plan, but, then again, if you are suggesting that he is that far out to lunch what are we to do with that?
> His doubling down on it in other comments
The fact that he was writing comments about it suggests that he doesn't take it too seriously. Nobody of sound mind wastes time writing stupid comments on the internet if they have an actual motive. They go out into the real world and make it happen.
If he is completely out to lunch like you suggest then anything is possible, but the more likely explanation still seems to be making fun of those who don't believe the wealthy pay tax. If people were making shit up about you, claiming that you didn't pay any tax, you'd likely end up making fun of them too.
> Nobody of sound mind wastes time writing stupid comments on the internet if they have an actual motive.
I mean Musk? You even used it as an example of "just trolling" in your previous message. He invested north of $300M+ in the last presidential election, also injected massive amounts in various local races, as well as foreign elections.
What more do you need to take them seriously? Even if somehow Musk was doing all of this to troll people because to him $500M is nothing, the impact on those elections is real, so it doesn't even matter what is internal reasons are.
> I mean Musk?
One who is not normally considered to be of sound mind, but you do you.
> What more do you need to take them seriously?
A good idea? The voting thought, while possibly an interesting thought experiment, is not suitable for execution. Why would I take it seriously?
> the impact on those elections is real
Maybe you aren't asking why I would take them seriously, when there was nothing presented that is worth taking seriously, but why do others take their hobo shouting on the street ramblings seriously, presumably only because they have wealth? The other hobos on the street seem to be happily ignored.
Good question, but one I do not have the answer to. I'm still trying to figure out how these particular hobos shouting nonsense on the street corner became people I am aware of. They're just people, same as all the others that manage to remain anonymous. Definitely some kind of malfunction in my brain going on there.
I found this just when I googled his name plus controversy
> Shopify CEO Tobias Lütke faced fierce online backlash in July 2026 after endorsing a social media proposal for a tax-tiered voting system that would strip voting rights from non-taxpayers and give multiple votes to high-income earners.
but if we go as far as to say
> If it’s touched by Tobias Lütke, I’m out.
does it mean control or sponsored, for example ladybird web browser...
Ladybird contributors have their own controversies that make the project hard to stomach
It is a good thing you can use Google chrome instead which has zero contributors with stances you disagree with... I guess.
That vague posting is a bit frustrating to be honest.
Are you talking about the fact they do not accept pull requests? If yes, they had good reasons for doing so, LLMs make it really time consuming and exhausting to manage an open source community, I completely understand why they decided to stop accepting PRs for now
Ok, what is the new drama this time around?
https://fortune.com/2026/07/27/shopify-ceo-voting-rights-str...
Along with a myriad of different controversies. Just searching his name will bring up many articles about his chronically online Elon Musk wannabe takes.
He came to Canada as an immigrant, made some buck and now unashamedly calls 'idiots' the very people that built the communities where he lives. The guy is a complete POS.
https://xcancel.com/tobi/status/2081155797640171641
Some people believe that anyone who does not maintain 100% ideological purity and goodness, which can be measured by if they align 100% with the persons own and therefore morally good political ideas, which also coincidentally matches the status quo of Brooklyn based journalists, that the heretic must be ostracized from all of society and it is impossible to work with them. Since Tobi Lutke made a mortal sin of disagreeing on suffrage requirements, they should be cast into fire and ostracized.
I don't think "citizens have a right to vote" is anywhere near the realm of ideological purity or "Brooklyn based journalist" thinking. It's literally just democracy.
Freedom of speech seems like something Lutke would pretend to like, and him getting torched for his anti-human opinions is a great demonstration of that principle.
> Lutke made a mortal sin of disagreeing on suffrage requirements
I think this is a very tame and apologetic and dishonest way to put it. Suffrage and voting is how people have a voice in the future of their democracy. He advocates for a policy that materially removes that right, that say in the future, from those who already have little power. I don’t know how else to emphasize that it’s more than “they disagree a bit with him and now he’s an impure sinner”, it’s more akin to “this rich dude wants to rewrite the rules to give himself more power and us less”, which is a sober assessment of his policy. I think people actively trying to avoid giving him any more money or power is a very understandable reaction to that
I think ultimately some parts of ideology are bound to and by the reality we live in, and acknowledging that reality gives a good answer here for when _it seems like_ people freaking out at an “impure idea” - is there a documented harm that we can look at?
Many people boycott Elon and his products. Do you raise this same “moral purity” complaint about them? Does it ever go in the other direction?
I think technology and how we build it is deeply tied to us as social, communal beings. Tech can be a reflection of our capabilities, our values. Tech shapes our reality in turn. It’s worth thinking about, being critical of, and believing people when they say what they believe
This is literally what capitalism tells us to do though, isn't it? If you don't like it, don't buy it.
No, this is "omnicause" / "intersectionality" type thinking.
I think we might be talking about different things. My understanding is that voting with your wallet is orthogonal to why you want to do it.
> Location Though the team has spent many years scaling remote-first companies and is currently remote, we are hiring specifically in Los Angeles, London, and New York, with an expectation for a flexible in-office schedule as we build our presence in each region.
I always get a "lifestyle business" vibe when tech companies are based in LA or NY. When you're a seasoned veteran you'd prefer to live in a larger, more fun city, even though tech is concentrated in the Bay Area
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?
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.
And salaries are lower than SF.
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?
A business that lets employees have a healthy work-life balance.
Thanks, it’s a bit concerning that’s not considered the default
That's not the original meaning of a "lifestyle business" anyways, which is more like a business that makes enough (to its owners, nothing to do with the employees) for them to maintain their desired lifestyle, like maybe a few million in revenue or whatever, but can't / won't be scaled beyond that because their lifestyle is already well supported by the current business growth / volume / revenue so they have no interest on blitzscaling, taking VC money, and running the perpetual growth treadmill or because the business is already topped-up and won't support further scaling (its not always a choice)
Ah ok, in that case Superlogical doesn’t sound like a lifestyle business given the number of VCs involved
Yes, indeed, although i think in this case it was just signalling how not being "SF focused" might be seen as a lack of ambition or whatever in such VC contexts or something and thus the exaggerated comparison to it being a lifestyle business merely due to the location requirements on ny/la (nicer cities to live in, maybe, than SF, or whatever)
i think just michael is based on LA, so he prefers having the offices near his home (has kids, etc), and has enough FU money to not need to let the VC's dictate where the HQ's are located at.
I guess also they dont love the idea of managing people remotely for a small startup/first hires, although im sure for the right person they wouldn't really care (or usually this is how it goes on most small companies/startups)
Generally I think lifestyle business means a business that is going to make the owners a comfortable living but not likely to make anyone filthy-rich.
In silicon valley/tech it's sometimes treated as a derogatory term.
Really exciting IMO. Many people are working on new work interfaces (e.g. Codex/Claude/Cursor, multiplayer agent spaces, unified Slack/Notion/Zoom, etc.), and I think the Superlogical vision makes sense as a competitor in the space, at least to try.
There’s clearly room to improve our existing tools.
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-specific -- if somebody put a small fraction of the effort into it that others are putting into superlogical and the other alternatives listed here, it could be awesome....
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.
Devs don't use Dropbox, they use git. Git is probably the poster-child example of an unfriendly interface, and yet it's the standard way for devs to share files.
Emacs is downright friendly compared to git & tmux IMO.
I love how ghostty does windows/tabs/splits, but ultimately, I don't find myself using them. I find that I only have so much muscle memory in my hands, and I need to stick with one solution that works everywhere. A multiplexer let's me do this on my own machine, or if I'm SSH'd in elsewhere. Personally I use Zellij, because tmux feels like learning another vim. But I do use tmux when using tmate, because it's the simplest way I've found to share terminals.
Anyway, I'm very happy to see someone with Michell's talent entering this space. We can use more options with sane defaults and great performance.
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.
Im assuming that they've noticed tools like CMUX and Herdr (cmux is using libghostty too and i use herdr in ghostty) and this is gonna be similar?
Anybody else using Superset as your multiplexer rn? I'm finding it quite useful but wondering if there are other existing products that I should consider -- Mac Silicon: https://superset.sh/ Curious to see what Superlogical cooks up.
My understanding: it’s like tmux but the session lives on a server instead of one machine and you can open it on your computer/phone/etc
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?
Maybe I just missed the boat on hashicorp and similar solutioning, but I find a lot of the solutions for working with multiple systems using an intermediary shared language, such as terraform, end up obfuscating the problems they intend to solve and result in people who deal with and support those systems on a daily basis having very little understanding of how to interact and troubleshoot the core infra that is being built up with those tools.
Yes, building those solutions requires a lot of expertise, but I've never found the tools to be worth engaging with over just using what the standard platform tools they abstract away were replacing. Then you add in the economic rug-pull scenarios that have happened when utilizing software like this and I wonder why everyone hops on the bandwagon so easily.
What is the company they're building trying to solve that system maintainers can't already do? Isn't this just rebuilding OS windowing?
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.
OT, but the superlogical browser tab icon gives me flashbacks (pun intended) to the flash player icon:
https://upload.wikimedia.org/wikipedia/commons/3/31/Adobe_Fl...
Funny. I started working on a similar product myself just a few days ago. I have multiple machines for multiple different purposes. Each one runs different agents (some local, some cloud), for different tasks. Different ones store different data (a NAS on one, my VMs on another, etc.). I wanted a durable layer over all of them to be able to start work on any one of them and resume on any other, and for all the disparate agents to share memory/configurations where possible, and for each node in the network to make all the others aware of its purpose (so when the NAS is online, the agents know how to connect to it and what it contains). Sadly no VC funding for me, building it on my lonesome :)
Sounds super interesting!
What's your approach?
Still exploring different options but at the moment I run a daemon on each node that connects via iroh to link the nodes (and NAT traversal). Once the link is established the nodes share data via a Raft-based datastore. I can launch a TUI client on any machine. This presents a chat interface like a typical LLM harness. From here I can configure which model I want to use, how the model is executed and on which node (for example I can use Ollama on machine A to serve a model that’s running on machine B, but executing commands via this system on machine C). Each daemon then acts like a typical LLM harness, but headless. So the commands and tool calls are streamed into the shared store and finally end up in my TUI. Since all the state is in the distributed store I can pick up the progress wherever I want, as long as there’s a node running there.
Still a lot of big problems like latency so I’m not too happy with the current design. It’s a LOT of state to replicate, and obviously things get problematic when an agent was working on a project on machine A but then ends up on machine B which doesn’t have that project directory around at all. So the agent needs to re-orient itself to figure out what happened. Still pretty fun to play with.
My biggest token sink so far has been something very much like this, but I'm still not sure whether it's a good idea. But it has been a lot of fun.
might be a related sneak peek here: https://x.com/mitchellh/status/2079327969416482859
can't wait for more details!
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?
why you want to build new terminal multiplexer? Is there issues with current? tmux or zellij?
Still very interested. Especially if you will go further than terminal - multiplexer for window environment, not just in terminal
This is an incredible idea. I would usually be skeptical that such an engineering-focused tool would "make it," but Mitchell Hashimoto is one of the very few people I think can pull it off.
This is going to change so many things if it starts getting adopted.
Can you give me one concrete use case that explains how this is an incredible idea? Thanks
That explains why he built sessions into libghostty instead of Ghostty itself. Looking forward to it.
I spend most of my time in the terminal and have been using tmux to manage my sessions and workspaces, mainly via sesh.
But recently, I decided to move to Kitty and use its built-in sessions instead. I built something around it as well. It’s still early, but I’m really liking this workflow so far.
https://github.com/alienxp03/kesh
I love Ghostty. But I miss out on lots of its features because I use zellij exclusively to work with agents. This sounds exciting and I hope they'll choose some cliff oss license like gitbutler does
They're not hiring in EU remote? Barely managed to somehow open the links on my phone since I don't have my laptop around.
`ssh superlogical.jobs` gives me same vibe as telnet bbs when I was in college :)
I dig that the jobs page is via ssh!
> We believe the missing layer is a durable session around the work itself
Curious as to how this will be different than K8s?
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?
Doesn't K8s already practically provide a layer of abstraction across all infrastructure? You might not think of it as being a session, but to the degree that it's responsible for managing scaling and failover, it kind of is.
This isn't a statement of value, I'm genuinely curious myself. In fact, I bet this new thing will probably be better.
> For us, that means interactive work, automatic work, and production work would share one well-crafted underlying system instead of living in separate tools.
Unless this is only for the narrow band of purely local work, I would be surprised if k8s didn't play a big role in the backend solution. It is the standard "well-crafted underlying system" for automatic production work. But, it isn't strong in the interactive-work/frontend aspect or other aspects, so it certainly wouldn't be the main focus of the final product.
Random example: you want to multiplex some work across multiple (self-hosted) agents, using niche fine-tuned models. It would make good sense if those agents were automatically provisioned and deprovisioned (and isolated to your subtenant), and k8s is a great underlying system for that sort of thing.
Commercializing and integrating (openclawing ?) cmux to take all the jerbs?
https://github.com/manaflow-ai/cmux
Dam it, he did it again. Amazing guy.
Bloomberg terminal dead soon
Extremely excited for this as a tmux and Ghostty user. Given the team, I know a great deal of care will be taken in building this product.
This sounds awesome enough to see if they're hiring, and they are, but unfortunately only in certain areas with an expectation to be in-office with some regularity. Bummer. Exciting to see this team together, though. This is a tool that I want and think about daily.
I have no idea what this is but I appreciate the roster.
This but software: https://en.wikipedia.org/wiki/Supergroup_(music)
is this the same thing as/related to https://superb.systems/ ?
https://go.superlogical.com --> https://superb.systems/
edit: aand someone fixed it :) I guess that must've been their older branding (it was a very beautiful website)
Very cool to see a man execute on his vision. Ghostty -> libghostty -> terminal multiplexing -> ???
It’s interesting because in an ideal world he is practically building a NixOS config.
I’m pretty new to NixOS (a few months) but i could envision using Nix to declarativey configure tmux windows and panes in a dev shell. That said, Nix feels like it might still be too clunky at this point in time to quickly re-declare window and panel arrangements. But I never tried and it could be possible.
Congrats. Not entirely clear on the vision but most things he touches are great so am sure whatever they come out with will be good stuff.
Funded by Mario Zechner (pi.dev) among others.
So, at least 2 of some dudes.
Kind of funny that he lists himself as "Creator of Ghostty" first and not Hashicorp.
"Why is terminal usage more widespread than ever?" - I guess because of AI agents?
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
How can I get points in the company?
So excited to see what this dream team cooks!
ssh superlogical.jobs
This actually works!
I am guessing Ghosty will become closer to Warp.
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.
domain could be very good for some relevant AI startup and not multiplexer..
Literally everything Mitchell does is top notch. Looking forward to seeing how this turns out!
the careers page is just `ssh superlogical.jobs` which is kinda cool
We're a team that has spent our entire careers building some of the most widely used developer tooling, infrastructure software, and AI systems.
Not being snarky. What AI systems have they built?
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.
> We will build a high-quality terminal multiplexer that remains excellent at that job, even as it grows to support the second and third parts of the plan.
why this feels the next step is data collection and Ads?
Would you pay for it to not have data collections and ads?
Taking VC money to being a better (probably worse) tmux is the sort of clownworld stuff only people in the Thiel circle can come up with
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?
Surely the opposite question makes more sense: what's the value add of newer tools?
I will admit that I've used https://herdr.dev/ before because unlike tmux, it displays the status of running LLM agents.
Can someone explain as pi as a harness vs Claude Code or OpenCode?
People keep saying "yeah, it's customizable!", but I can't be bothered doing that.
i just have log.md that keeps a running log of things as i work though a project. thats all my claude code needs to navigate 'session' across systems.
"Any sufficiently VC-funded AI startup will inevitably try to re-invent a log.md file" - Spongebob Squarepants
another IDE for AI harnesess?
https://xkcd.com/1406/
I'm sorry to take it out on this specific announcement, as it just seems to be one of many "offenders", but why are the 15 investors listed on an announcement blog post of a semi-stealth startup?
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. What happened to the good old way of figuring out product market fit first? Is that the only way a product can succeed nowadays, or why else would a ~billionair that could easily fund it himself pull in that many investors that early?
---
I'm excited to see what Superlogical will build, but the startup scene seems to grow more clique-y by the day.
"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 don't doubt that, but that phenomenon seems to be reaching egregious levels.
You are commenting that on a website where this is part of the value proposition since the early days, so I wouldn't say that's a new phenomenon.
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!
I'm not doubting that Mitchell Hashimoto is capable of building great products with great product market fit. Quite the opposite, as the few strands of idolization that exist within me rank him as one of the best to ever do it in the dev tooling space.
I'm also not doubting that there would be tons of investors throwing themselves at him for a new venture like this.
But why, apart from forced market access would it be in his interest to take that money, given that he can nicely fund it for a few years with ~1% of his net worth if not for forced market access?
I seriously doubt it is for the money.
It said they are funded, but it doesn't mean they lost control or have to be VC hyper growth. Apart from Notable Capital and Amplify Partners all others are individuals. There are lots of benefits having these group of company founders and CEO as investors. They are likely interested in the problem space as well.
Had it been hyper growth startup you would be looking at a16z or other bigger names.
1. Accountability - even very talented people can lose motivation or get distracted. Having investors to report to helps with this.
2. Recruiting - many people would be wary of working on a rich guy's pet project. Taking investment makes it more of a real company that has to deliver something and is not subject solely to the whims of one person.
yes, dev tooling investment has always been disappointing. Mainly due to stingy buyers.
I really wish mitchellh wouldn't take funding from some of these people. Tobias Lutke in particular has chronically been on the wrong side of fundamental issues, and has most recently advocated for a pay-to-play version of democracy where wealth determines your ability to participate (and in particular your "stake" in the democracy).
> 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'm sure the people who would have their right to vote stripped from them would consider that malicious, and they would be correct. That's not democracy.
> I'm sure the people who would have their right to vote stripped from them would consider that malicious, and they would be correct.
I don't agree with him but you're incorrect here. To play devil's advocate, if his policy idea resulted in better outcomes for them, e.g. better material conditions, then they may end up preferring not having a right to vote and consider him a good person.
> That's not democracy.
Democracy isn't necessarily the same thing as direct democracy, or "everyone votes." The Ancient Greeks didn't practice democracy this way. In any case, democracy isn't a virtue in itself. If one system is democratic according to your standards but the vast majority of its citizens live in squalor while another system is not democratic according to your standards but the vast majority of its citizens live in prosperity, which system is better?
Historical precedent (both recent and not) shows that concentrating wealth and then giving those wealthy few an inordinate amount of political power does not end well for those disenfranchised by this system.
It is not reasonable to give the benefit of the doubt on the basis of Lutke and his ilk maybe being benevolent dictators if given largely unfettered access to the levers of power.
> Historical precedent (both recent and not) shows that concentrating wealth and then giving those wealthy few an inordinate amount of political power does not end well for those disenfranchised by this system.
Historical precedent isn't a rational basis for argument. You don't make stock picks based on precedent. You need to actually examine the causes that result in outcomes to evaluate whether the same outcome would happen again. That said, I have seen good arguments for why weighting votes based on tax contribution would have bad outcomes so I don't agree with him.
> It is not reasonable to give the benefit of the doubt on the basis of Lutke and his ilk maybe being benevolent dictators if given largely unfettered access to the levers of power.
I have no reason to assume Lutke is a bad actor but maybe you know something I don't. Has he done anything to harm anybody or any group of people? Him being rich doesn't make him automatically a bad or suspicious person. There are many rich people who do good things with their wealth and free time.
I think most people will agree that empowering the edgelord Shopify CEO is not going to improve most people's outcomes. All you need to do is read his tweets to see that he should have less power (much less), not more.
> I think most people will agree that empowering the edgelord Shopify CEO is not going to improve most people's outcomes.
This is another logical fallacy https://en.wikipedia.org/wiki/Argumentum_ad_populum
> All you need to do is read his tweets to see that he should have less power (much less), not more
I'd like you to make a real argument. What specific tweet of his implies he should have less power and why?
It's not a logical fallacy. You made the unsubstantiated claim that people would want their right to vote stripped from them. I do not think people want guys like this in power: https://xcancel.com/tobi/status/2081155797640171641
> You made the unsubstantiated claim that people would want their right to vote stripped from them.
That wasn't a claim, it was a hypothetical.
> I do not think people want guys like this in power: https://xcancel.com/tobi/status/2081155797640171641
What is it specifically about this tweet that means him being in power would have bad outcomes?
I'd start with some German guy calling native Canadians who are actively civically engaged "useful idiots" and his suggestion that the unemployed and retried shouldn't have a voice in their community.
> his suggestion that the unemployed and retried shouldn't have a voice in their community.
I don't see him making that suggestion. It's pithy so it's hard to really pin down a meaning to the tweet but I think he's saying that these town sessions aren't a meaningful expression of democracy because only a specific section of society is in attendance and one that may not understand key issues relevant to working or young people.
Calling people "useful idiots" is pithy?
In this context, I would call it pithy sure. It's concise and direct. He's not explaining further.
He clearly isn't calling the other categories he listed (the " Retired, Unemployed, Unemployable") useful idiots, that's why he listed "useful idiots" separately. I don't know to whom he is referring as useful idiots but I don't think it's immoral in itself to use the term "useful idiot."
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.
Easy to say this when the original commenter incorrectly described Tobias' comments. He suggested that those who pay more in taxes should have a greater say in how the government is spending the money. This is not a crazy idea on its face; you could say that taken to the limit it would not be a good system to live in (the highest taxpayer dictating what gov't does).
But it does make sense that those who are footing the bill for everything might be consulted a little on the plans to spend.
reading "those who pay more in taxes should have a greater say in how the government is spending the money" and NOT taking that to mean what the OP is suggesting it to mean is naive. Everyone knows what Tobias meant when he said that.
Kind of says something about the system we've chosen to run our societies on, huh?
Now we’re ruled by an oligarchy of psychopaths and everyone’s wondering how this happened
People are only allowed to take funding from investors whose politics align with yours?
"I think this guy is dumb and wrong. Oh no, this dumb wrong guy is going to have outsized influence on a project I'm interested in. Oh no, this dumb wrong guy may profit from a project I'm interested in. Oh no, this dumb wrong guy is using his wealth to do dumb wrong things."
This is a normal and reasonable thought process.
They stated their preference, not that there should be a rule to enforce their preference.
more info: https://mitchellh.com/writing/superlogical
I’m sorry. I tried to understand what is this about, but I couldn’t. The title didn’t explain it. The description neither.
I remember I heard a saying somewhere that if you can explain what you are doing in a few simple sentences, then is is this really something valuable?
The vision of having a shared interface for all work seems pretty clear to me.
Blog post: https://mitchellh.com/writing/superlogical
> Guillermo Rauch
This guy posted selfies with the war criminal Benjamin Netanyahu, did you really need his money?
https://x.com/rauchg/status/1972669025525158031
It sounds like they're building a super-root which can destroy your whole infrastructure at once ???
From their website:
"Building and operating software today spans local machines, remote hosts, sandboxes, services, and production systems. It has many modes of operation: interactively with a human developer, automatically through CI and background processes, and increasingly through agents working in parallel.
[...]
We believe the missing layer is a durable session around the work itself: one that can span applications and environments, provide relevant context by default, expose structured data and actions, preserve history, and be driven by software while remaining visible and controllable by people.
[...]
This is our plan to build a multiplexer for all work: 1 Build an incredible multiplexer. 2 Make everything in it composable. 3 Make it safe and operable in production"
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.