Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- I don’t think the post has a lot to offer, but I will share thoughts on how a modern operating system can add _a lot of value_ both for non-agentic users and for agents.
Looking towards macOS and iOS for example, an underrated feature for the “it just runs your web browser” crowd is all of the system-provided databases. These turn rather specialized systems - photo libraries, music streaming, calendaring, contacts management, and mail into shared databases that link applications together.
On iOS it goes much further - because of HealthKit, my Withings scale, my Garmin (when I had one), my MyNetDiary nutrition tracker, my continuous Glucose Monitor, and my third-party analytics apps share the same access to my data. And it is done securely - I can choose exactly which part each sees. This is a tremendous resource for both developers and LLMs.
My calendar, hosted on a personal server but available through Calendar, is transparently available to my to-do list app - so I can see my appointments, my tasks, and even my workouts (thanks to health kit) in one view.
Weather is a system service on iOS. Calendar is a system service. Contacts is a system service. Health data is a system service. Even music - third-party apps can control and consume your music library on iOS - is a system service (if you use it).
My Mac knows how to develop RAW files out of the box, and it can connect with integrations to get them printed, and use my library metadata to sync my favorites to my phone. I have a third party app that syncs my Apple photos library to my NAS.
A good OS, like iOS or macOS, could be the fabric that holds all your vibe-coded and ephemeral LLM apps together. I see it making a whole new category of things possible.
by nxc18 - > That makes a lot of sense in a world where we’re importing all our software from strangers. It makes less sense in the world we’re heading to, where most of the software we’re carving up fiefdoms for has the same provenance.
If I'm running software written by an LLM, even if I was the one who prompted the LLM, I would still want my OS to treat it as if written by a stranger.
I know he's not necessarily talking about relaxing security models here, but he's not being very specific about his vision otherwise.
by chroma_zone - In my opinion, this post misses the forest for the trees. Yes, the model of an OS running various off-the-shelf apps for various tasks will soon become obsolete. But it's not the idea of the OS that's outdated - it's the idea of apps. Why would you ever want to ask an AI to make an app for you to complete some task, if you can instead ask the AI to complete the task directly? The future isn't a bunch of extremely personalized apps - the future is a single AI assistant doing all the work.by Xirdus
- @tptacek I think my question about your post is this — right now people rely on a lot of apps written by professionals for things like mobile banking, messaging, medical providers, mobile payments, and subscription services (music, tv, etc). The interests of the app publishers aren’t aligned with total user freedom — banking app developers don’t want a totally malleable OS where there are no process separations or trust partitions. Rather, they want to rely on OS level guarantees that their code is isolated from other apps, that the binaries are signed, that the platform is somehow trusted.
I love your idea of an OS where users write their own apps as they want. But if they want to connect to third party services or use commercial software, won’t there still have to be a lot of security guarantees? I don’t just mean the depressing “browsers must implement DRM or you can’t watch streaming video.” I mean “how will medical providers trust your device enough to give you your lab results” if you are thinking of majorly changing the mobile device security model.
It’s not a hostile question, I’m genuinely curious about this design tradeoff.
by decasia - I don't know what's wrong with my perspective. I cannot relate to this at all.
I am at a job where I am paid well to build software to do things in a distributed system that processes information across various external systems (some with physical world impacts). The outcome of that computing is revenue for the company. Using agent-wrapped LLMs makes the software authorship component of the job really fast. The rest is virtually the same speed.
For Thomas's mobile computing examples, these are done really well by my phone's operating system's built-in voice command system, mature for a decade by now.
At home, I use software that is well-crafted to do some thing. Or I write a series of scripts to do narrow things that for some reason aren't in Aptitude or Snap, for example "copy all the photos off my iPhone and rename them and sort them by date in my filesystem".
Perhaps I have no imagination, but I don't feel like having an LLM make applications on my phone. I want good software that someone that I trust wrote.
Thomas has been a successful cryptographer, vulnerability analyst, and entrepreneur, so he probably knows what he's doing. I just don't get it.
by linkregister - > "But it booted straight into BASIC. That’s all it did. I was, like, 8 years old. I wasn’t about to learn BASIC. What I learned instead: computers were not as awesome as I’d imagined."
You were the exception. Most kids felt awe. We started learning BASIC and creating "dumb games". That's the origin story of most people my age who ended up in this doomed industry.
- I feel like most articles challenging OSes have no idea what an OS is.
Sure question what the latest tool can offer and how to integrate with it... but that's just being an app, or a window manager, or a package manager, or a distribution, that's not challenging what the OS itself is and does.
If your OS doesn't change how a computer allocates resources, then you are talking about higher level, and that can be super valuable too, arguably more valuable, but ... that's a different thing.
PS: also chroot (half a century, just checked for the fun of it), jails, containers, VMs, etc limiting how the allocation can be done do exist and have been used for decades already.
by utopiah - Hey, all. I really don't know what to do with a post like this.
I'm being sincere when I say (as I've said on two threads here) that this genre of posts --- "I'm leaving this company I've been very publicly associated with, and here's the new thing I'm doing" --- is deeply cursed. There's no way to say anything interesting without it just stinking like an ad for the new thing.
Obviously, anything at all you say about a commercial project you're working on is easily read as promotional. And you're right, this kind of writing almost always is promotional. But there's a way to do it where at least you're trying to be in conversation with your peers, rather than hitting people over the head with how awesome you think the project is.
But I don't know how to do that in a post like this. I think the only way to read it is as, like, an investor memo. Not my goal, but I don't make the rules.
So my strategy here is just to stay kind of vague, and talk about where I think the world is going, rather than the specific thing we're doing. I can talk your ears off about capability systems, datalog, models driving hardware, virtualization, whatever. Those are fun conversations and I'm very psyched to have them; it's what lights me up about the work we're doing now.
But I don't think it can work here. I didn't submit this post and I didn't upvote it. I wrote it because I didn't want the whole thing I'm leaving Fly.io for to be wrapped up in some dumb Twitter thread.
If you're unsatisfied with the post, I don't blame you, but it's less a bid for the front page of HN than it is an update to my "about me" page. I'd literally rather talk about HN meta, and how to write for HN, than I would about operating systems at this moment. I truly appreciate the interest though.
by tptacek