This all feels like a race where the model companies try to solve doing work locally in a way that doesn't suck, before the major operating systems companies figure out AI integration into their OS that doesn't suck. It also makes me wonder why Google which has both Gemini and Android can't figure this out, and if there are lessons to draw from that.
Google is historically terrible as a product company (and has succeeded in spite of that) As their technical innovations become less of a moat (we're already there) they won't be able to win on engineering alone (they are no longer winning on engineering alone)
But Kubernetes, being a former Google project whilst donated to CNCF, is still one of the mainstream container orchestrator if not the biggest one (I believe we need to go over to supercomputers for that). Nomad is kind of dead and Apache Mesos is basically dead. Leaving Kubernetes as a natural moat, plus Google employees are still seemingly actively maintaining it.
And also Go. While I'm not a Go guy (speaking as a C# and Rust guy, I did wrote a good amount of Go before) it has a huge dominance in Cloud-Native application. For one, Zitadel, an alternative to Keycloak, is written in Go and only takes a fraction of what Keycloak needs.
Flutter/Dart is catching up, but the ecosystem is still relatively weak.
Kubernetes and Go are not moats of any sort; Google doesn't make any money off of them, and they doesn't get them any useful user data like Gmail, Docs, etc. do.
Kubernetes makes me want to go running to WebSphere 5 XML configuration files (version 5 is relevant here).
Whereas Go only exists because a bunch of renowed Oberon and UNIX heads did not want to keep using C++ for their work, and got twice lucky, first with their line manager, secondly with Docker and Kubernetes getting adoption after their rewrites into Go.
> But Kubernetes, being a former Google project (...)
You are confusing Kubernetes, the software project with Kubernetes, the container orchestration system.
There are many Kubernetes distributions out there which are not maintained in any way by Google. Even Canonical provides its own Kubernetes implementation.
Nowadays Kubernetes is a keyword much like Unix or Unix-like.
Please, Kubernetes is the Chlamidya of infrastructure software. Ever hear of a large K8s migration going well? No, because it needs to be completely re-implemented under the hood for anything halfway serious. Mesos was pain in the ass to use, but was mature when K8s arrived. Nomad was cool though.
Dart is also complete shit: Speed of javascript with verbosity of Java. Who the hell though that was a good idea?
I've seen lots of successful large k8s installs. This same argument from the nomad crowd comes up constantly. There's a reason why nomad lost, and it's not because it's better.
...well, AI infra stacks are more or less k8s/cloud native, especially on the inference side. Nvidia GPU operator plus KubeFlow makes deploying models easy, and I did that manually as well using just the ollama, llama.cpp containers and especially vllm operator. I'm not sure about what's the measuring notation of "large" is, but OpenAI had a blog post about how to manage 7500+ nodes Kubernetes (https://openai.com/index/scaling-kubernetes-to-7500-nodes/). That's 5 years ago and I speculate it will only be even more.
> completely re-implemented under the hood for anything halfway serious
...an example please? I'm building a platform to deploy hundreds of open source apps on Kubernetes, you know, the boring thing you can do with a VPS and maybe use Docker Compose to start -- with an ergonomic twist. I relied on so many features of what Kubernetes and its ecosystem provided, especially with persistent volume, volume snapshot, CRDs and cron job, and all of it is just composed from open source and cloud native software...I'm not sure if that sounds "half way serious" to you.
I know those boring monolithic apps can be done with a VPS, as I exactly came from that background, and I know what the shitty points of having just a VPS are. You don't have a clean control plane, you don't have HA, you don't have distributed storage, and you have to be super aware of the apps that you are deploying with Docker Compose.
What you would think "completely re-implemented under the hood for anything halfway serious", perhaps it means that you don't want to get in the fuzz to manage the complexity and prefer to trade for a far simpler but more primitive solution, and that's totally fine.
Pro tip: Codex seems to love generating great Helm charts so much for some reason unknown. I tried GPT-5.4 high in codex and it easily beats Opus 4.7 max on my own internal helm chart generation benchmark evaluation, which measures HA, deployment and app-specific probes. I've given source code to both Codex and Opus and research about the app config structure, and Codex did so well to just generate secret key-value and convert it to JSON and environ, while Opus insisted on using a dynamic generator at runtime.
> Nomad was cool though
Ahem. I used to deploy Nomad too, but I found it too underwhelming, especially with the networking side of thing.
Consul is perhaps one of the most hated thing in my entire career that I would call it bullshit. I know that later Nomad has its own Raft mode, but the Consul brain rot to me is already very sickening such that I don't want to touch Nomad no more. Vault is fine though, but I prefer OpenBao now.
> Dart is also complete shit: Speed of javascript with verbosity of Java
I don't get where the "speed of javascript" is from. Dart/Flutter is JIT on dev mode, and AOT compiled on production, is that you didn't get the compiler options right?
As a person who gets paid to make Chrome (CEF really) do its bidding, I would say Chrome is really as close to an OS as it can get, as in I've found API or service typically an OS or an external tool would provide, that wasn't built into Chrome.
On a semi-related note, I bought a Pixel phone about a month ago, and I'm shocked by how unpolished it is. I've had so many little annoyances pop up, issues I never had on other android phones. Keyboard hiding/appearing when it's not suppose to, bluetooth dropping, WiFi dropping, network switching taking forever, screen becoming unresponsive... It's mostly all small things, but they really start to add up after a while.
lmao this is a terrible take. Google is the only company with AI deployed in every market where AI can exist, and _all_ of their products in those segments gush money.
They are all pretty par for the course. Google used to be outstanding... but I'm not sure of a single product they have that is outstanding (def: significantly better than the competition) anymore. On the other hand I rarely use any google products these days, so maybe I'm not the one to be judging.
I'm not a fan of Google, and also not attached to Apple or Microsoft, so this isn't me trying to stan for Google, but I'd like to request that you give examples of what competing products are categorically better (and, by what metric(s) you're judging - code quality? stability? robust set of features?) -- for Gmail, Docs/Drive/etc, Google Calendar, Maps, Classroom, YouTube.
As far as I can tell, if judged by the marketplace (and breaking ties with which product I like better), Google has run away with the ball on all of those, and Gemini seems to at least be competitive.
The only major product I'd say they've sunk below acceptability on is Search, which is demonstrably dogshit now...though I suspect it's more that they have changed their definition of what Search is for, from "helping users efficiently find other websites that are useful to them" to "A convenient on-ramp to, many times per day, capture the current user intent and steer them toward something that earns Google some ad revenue."
The major differentiating factor that Google has had in every product category is that their products are free and you have to deal with ads (and they monitor your behavior for profiling you and your interests).
GMail and Google Maps were revolutionary when they came out, sure, but the vast majority of Google's products now are... fine? at best? And a lot of their "big products" were acquisitions that they absorbed in order to further the core goal of the business - to organize all the world's information and use it to serve ads to people.
Meanwhile, Google has a litany of products they've started internally, launched, ran for a while, and then let stagnate or canned entirely; anecdotally I've heard that this is because your bonuses at Google hinge on your ability to launch a product and not your ability to support a product, so it's beneficial to get something launched and then immediately leave to go launch another project rather than polish the one you just launched into something to be proud of.
I'm not sure if that's true, but it would certainly explain a lot; if Google launches something and it's bad or it doesn't click, they just give up on it. Google Wave, a half-dozen chat apps that I can think of, Stadia, and dozens of others. Things that Google launched, which had problems or didn't hit mass adoption instantly, and then just petered out and were retired with all of the time and energy and money put into them arguably wasted - products that people wanted, and wanted to succeed, but which weren't revolutionary successes at launch so they weren't worth further investment.
Meanwhile, they (and most of the industry) are pushing AI for some reason despite the fact that almost no one actually wants AI to be the only way that people interact with information.
This all reinforces what I've been saying about Google for decades: they're not creating things that users want to use, they're creating things that they want users to use. Sometimes those things align, but when they don't then it's not worth further investment (except, apparently, AI).
I just don’t think your opinion is shared by most people.
Gmail is the most popular email service in the world, people are always telling me how they prefer Google Docs over everything else and their only competition is Microsoft.
Yes it’s free but there is no other service that I rather switch to, and I actually pay for additional storage.
> Gmail is the most popular email service in the world
That's because it's been around for quite a while, and for a long time it was the best webmail service. It's also free, unlike most alternatives. And switching to a new provider means a new email address, unless you're using a custom domain with Google Workspace (or whatever they call it these days), which is a small minority of personal accounts.
(I gave up on Gmail a few years ago and switched to Fastmail, and like it much more than Gmail. But I'm the rare person who is willing to pay for email, and had been using a custom domain with Gmail, so my non-monetary switching costs were minimal.)
You're absolutely right that Gmail was the best webmail service when it launched, and for some time afterwards. How many people now commenting on HN remember when Gmail launched? Remember how revolutionary it was at the time? Every other email webapp, when you clicked on an email, would refresh the page. Gmail, when you clicked on an email, did not cause a browser navigation. It simply replaced the page contents with the contents of the email.
We're so used to setting webapps do this that we take this for granted, but Gmail was the first email webapp to do this. It's possible it was the first webapp, period, to do this; I feel like Gmail's use of XmlHttpRequest was innovative at the time.
Fast forward twenty years, and what about Gmail is innovative today? Nothing that I can think of. It's mediocre (there are lots of filtering improvements they could make that they aren't making, for example), and everything that made it good has been copied by other webmail clients. There's no particular reason except momentum to stay on Gmail.
Gmail spam filtering also used to be revolutionary and an unsung hero. I haven’t put effort into finding out if other options have caught up with that (because of aforementioned tedium of changing email addressed)
I have had a pobox.com email address (just a forwarding one) longer than I've had a Gmail one, and their spam filtering was pretty amazing too. Even before I set my pobox.com address to forward to Gmail, I never saw very much spam.
Now that Pobox is owned by Fastmail, I rather suspect that Fastmail is going to have the same good spam filtering. Can't speak from experience, though, as I haven't actually used my new Fastmail account yet (it still forwards to Gmail, and so far I haven't switched. Momentum, again).
You can’t beat free. The Fastmail web interface is snappier than gmail. And you can’t beat dedicated mail clients like thunderbolt in terms of workflow.
Google doc is wordpad level with very good collaboration (but that’s mostly what people need). People were fine with typewriters, so they are fine with a word processor like google doc. But it’s not at the level of even Libreoffice or Apple’s page in terms of features.
By any definition of good usability, Gmail is not good and Google Docs are not far behind. It’s not that they are functionally bad, just really poor UX.
That's hyperbole. They have flaws, but at the very least, when they were launched, they were arguably best in class. I'm not sure how much me sticking with them is due to familiarity and muscle memory but I know they won we over purely on merit in the beginning.
As someone who was an original invitee to Gmail it was the clarity of function that was the differentiator. They were “grown up” and acknowledged user agency vs their competitors.
But as others have mentioned, they operating model of Google as a company incentivises creating products but does not incentivise refining it. Gmail has gotten far richer in functionality but at the same time the interface has gotten far less consistent. Their competitors (mainly Microsoft but not only them) also got richer functionality, but they also paid attention to UX. While none are perfect, there are definitely some better than others. Familiarity definitely breeds inertia though, I’ll grant that.
Oh, yes, typo aside (which is the lowest form of critique) it was.
It's not about why this or that person would want to install the MS Office on their Mac (millions do after all). So the pointed flew above their head.
It's about the point that many people having to use Google Docs are doing so because corporate wants them to. (In fact most people using MS Office do it because it's a work requirement too).
> GMail and Google Maps were revolutionary when they came out, sure, but the vast majority of Google's products now are... fine? at best?
Is that... good? I mean take maps -- what more can possibly be done to that product that wouldn't just make it worse? It's done. The fact that's the default choice for mapping and just works is fantastic really. There aren't any competitors doing anything revolutionary either because there isn't anything revolutionary to be done.
Maps is far from done. At the very least it's still riddled with usability issues. One bug-bear I have in particular: when I zoom into a very specifically chosen area, search for a Chinese restaurant, and it zooms out to half the state. Maddening. And it's rife with problems like this.
I think it’s a major feature gap that Gmail (paid or free) cannot create filters on headers.
I also can’t do wildcard filters on “to” or “from”. For example, in my GApps I have it set up to route all emails not associated with a specific user to my primary user. So that it’s easier to make throwaway emails. I want to filter all to:`X.X@domain.tld` to a certain folder. No can do.
For many years I’ve been creating filters on free Gmail for to:, from:, subject:, etc.
I set them up on desktop web.
Perhaps there is something more specific you’re trying to do?
> a major feature gap that Gmail (paid or free) cannot create filters on headers.
You can create filters on header fields like from:, to:, and subject:, so I am guessing you mean something different than “cannot create filters on headers”?
Mail headers also include to:, from:, subject:, etc, as well as more obscure items too, which is why I think OP commenter meant something very different than “cannot filter on headers”.
Also, more items that might help OP (as I can’t edit parent comment) - they mentioned wanting to use wildcards on to: field. Those header fields do allow specifying just part of the header, like just the domain, or one part of the to address. (But those match at word boundaries and I’m not aware of being able to match sub parts of words or more complex items.)
Regardless, I don’t think I’d call this a “major” feature gap - maybe minor or more of a niche feature.
Funny. Search is the only thing that is outstanding as it is the big revenue arm, that and youtube.
The last time i tried using gmaps i got ads and the thing could figure out where i was on the roads. It was comical as i always remembered google maps being better than apple. Today tho, apple beats them hands down.
Googles products that do not get cancelled are pretty mediocre in todays market. They can build useful things but if it doesnt have ads in it, it gets axed
> but I'd like to request that you give examples of what competing products are categorically better
Personally I much prefer Fastmail to Gmail. The site is way faster and more cohesively designed. Fastmail supports jmap, and way more imap extensions (including push support on Apple mail). They have helpful humans handling support requests. And they do all of that with what seems like 1/10th or less the number of employees.
The only thing I like more about Gmail is their native mobile apps. Fastmail’s official mobile app is a web view.
Maps & Gmail & Search all have plenty of accumulating flaws... but they also completely defined their product category and today are among the most popular software products ever made.
More than plenty of people (including me) would argue that Windows is an awful product, but a quite significant number of people (maybe even including me) would argue that its an outstanding product.
Wow I wonder why maps and gmail are extremely popular by the company that controls the largest browser, search, android, and advertising. It's based solely on their merits and not abusing their monopolistic position to thwart competition right?
Garmin was and is better than Google Maps and Mapquest was better than Google Maps when you needed to print directions. If Google didn’t have Android would maps matter as much?
Apple Maps and Waze is better for directions. Apple has better CarPlay integration and HUD. Google Maps is way better at searching for things like restaurants or local businesses but not as much the nav part.
Apple Maps is - or soon will - show ads. A trillion dollar company that charges thousands for the device. So that’s out.
OsmAnd’s UI really doesn’t suit me, and there are a few others that I personally liked even less.
But Organic Maps ticks enough boxes to come close. I’ve been using Organic Maps on iOS for driving, and it’s tolerable. And for offline mapping it’s a godsend.
… sure, but until someone creates a competitor that is good enough to overcome the switching costs, people are going to stay with Google. Most of Google's then-competitors in the categories listed are not just inferior products, they're dead. ISPs no longer do email, Alta Vista & Ask Jeeves are gone, MapQuest is a thing that makes us that used it sound old.
Not only does an upstart have to overcome the switching costs, they have to actually survive, and not just get hoovered up by acquisition and then Our-Incredible-Journeyed.
Search has degraded for sure, but still better than anything else? Maps - I guess you mean Apple ones are better? Can't tell, I am not on Apple, but if you don't use Apple products, there are not many alternatives to Google maps
At the enterprise level, if you know of something better than Bigquery, please let me know.
Similarly, Kubernetes and Kubeflow are both outstanding - and Licenses Kubernetes has no meaningful competition for what it does - but Google did everyone a solid by making them open source, so you can get them from other sources than Google. But the Google managed versions are certainly extremely good.
As for the idea that Gmail, search, and Maps aren’t outstanding, an easy way to refute that is to ask what the outstanding alternatives are. I doubt there’s a single list that many people would agree on.
All of these are outstanding! In so far they are not singular or new anymore, well... If for the past ~20 years nobody has come up with something clearly better, then I would say that speaks to how outstanding the product that are being copied are to this day.
It's … okay … but it still falls down in a fair few areas? It's crap at finding restrooms. Finding a stop on the road is also difficult, as it seems like it just defaults to a basic radial search, when as a driver you want things down-route, not radially out. All the AI in the world can't seem to figure out when I'm looking for gas or food that closed businesses are not results I want to see. It eats enough CPU to melt phones, such that Android now has built-in support for this?!¹. Attempting to report things often goes in vain². Some of the notifications need work ("object in road ahead" … I'd kill for what lane! this one is just anxiety in a notification), and it'd be nice to see the lane designations ahead of time (it only shows them once you're like <1mi out). I've never gotten the AI-home detection to work. Attempting to navigate to the house of anyone with an Irish name gets me a bar, and the forced-voice-navigation when in a car means I have to be able to pronounce the destination. Google does not seem to grok that sometimes … there's a person in the car who is designated navigator. They can type, it's fine. Some turn directions could be better if you incorporate more precise language into them³. Some directions could be abbreviated "Navigate to I-4 North": I live here, I don't need step-by-step hand-holding to the interstate, but I'd like to plug in the destination before the car is rolling.
¹literally, phones can now demand you put them in A/C b/c they're dying
²I reported once that a jetway was 3D modeled as being like 8 stories high. Google couldn't confirm that, and closed the request. I reported a business as not being present, while my GPS showed me as being at the alleged address, that also couldn't be confirmed. My GPS trace would have seen me walk the whole block, twice!
³as designated navigator in my relationship, I can tell her "leftish" or "rightish", and she understands what I mean. Where I live a lot of the intersections' designs appear as if a civil engineer was given artistic license, and so sometimes the direction is "5-way intersection, left-ish". "Left" is a bad direction when there are two lefts. Of course … me & her have developed a fairly extensive lexicon over years of long road trips, too.
I'm no google fanboy, but gmail is a very good web-based email, and google maps is a very good web-based map program. I would say "outstanding" with no reservations.
The Pixel series outside of security (to which their own flavor of Android doesn't even take advantage of like we see with GrapheneOS) doesn't have any particular outliers that would make it any more or less enticing than another company's phone.
Their ChromeOS hardware was nice but had lackluster software and by the time it was EoL'd, never got the love of ChromeOS-present.
Google TV generally gets outpaced by onn (Walmart's brand) on cost and value proposition.
And also the fact they have shown time and time again that they just kill products over and over again.
A big part of why Stadia was cancelled is because it didn't get traction, and a big part of why it didn't get traction was because of how many people assumed it was just going to get cancelled.
Other than search, in its heyday of the early 00’s, every google product success was either a 20% time project (e.g. Gmail) or an acquisition (YouTube) or a direct clone of someone else’s working product (android).
I think both can be true. Google has a history of annoying churn while still being good enough (or just … being large enough) that switching to competitors is still too high a cost for most.
For example, their "chat" app has churned 3? 4? times now? Their assistant app has churned from whatever the OG assistant was to now Gemini. Wave churned to "+" in the social category, and that's dead now.
The default placement in Android probably helps a lot, or other things, like forced signups into adjacent products (e.g., like + was doing for a while).
Do they have any supported way to export a user's account (e-mail, calendars, etc) for offline archiving yet? I used to have to reset their password, disable their 2FA, log in as their user, initiate a 'Takeout' request to export their account data into an archive, wait until the request was done (between minutes and days depending on the account), download it manually (often in chunks if it was large enough), store it somewhere, and then delete it and delete the account.
I can't imagine that no other 'Google Workspace' organizations want to actually save their employee data rather than irrevocably delete it forever.
Most of their best products are VERY old. Gmail, Search, Docs/Drive, Maps, Street View, Chrome, all solid but old consumer apps. A lot of early entrants/market defining products, too! Most of them are still very good. Calendar is... fine. Gmail/Calendar as corporate products in some ways feel like downgrades compared to the functionality of the older Office desktop suite, but they integrate nicely into a lot of things.
The bad rep Google gets now is because while they've polished their nice money machine very well, they haven't done much to pioneer new segments and the old stuff is pretty stagnant. When's the last time Gmail gained a must-have feature? Maps?
Ok but maybe those old products are just mature and there's not much room for product innovation? Even if we assume that's the case (though there are third party email and calendar apps that I use instead of Google's first-party ones on my phone), let's look at some newer stuff:
Android was a copy of Blackberry that turned into a copy of Apple before being launched. It's done well numbers-wise, particularly in lower-cost markets, but hasn't got a lot of novel product stuff going on in the last decade or more. Less luck as a tablet platform. A lot of the nifty novel stuff in mobile space (wireless earbuds, "find my" stuff, mobile payments, magnetic wireless charging) was done first by Apple.
Meet vs anything else - well, Meet is cheap if you're already on Gsuite, at least.
Voice/Hangouts/Chat/Duo/Whatever - whole lotta abandonware or dead attempts.
Google Plus - famously mis-targeted product.
Google Assistant - Siri copycat then with Alexa copycat hardware. Didn't do anything novel.
Chromecast - nice novel product, caught on decently well, lost the market to a combo of set-top boxes (Roku, Apple) and native-TV integrations. Tried doing Google TV as that native platform with a copycat pivot vs the original Chromecast, no particular success.
Stadia - abandoned quickly, never set itself apart from the things it was copying.
Smart home hardware (like alarms and smoke alarms) - dramatically scaled back and now pretty unambitious.
Gemini - copycat of ChatGPT which is particular damning since so much of the original research came from Google in the first place!
Arguably, "exceptional" products are not ones that can vanish on a whim, like a great, great many of Google products have. Or they actually compete with other products in the same space, like a great, great many of Google products have not. Also, one would argue a good product is not one that is bought out and then deliberately destroyed to prevent its expansion into or development of a market for itself. Google is an advertising company with tremendous reach because of a handful of very aggressive and very fortunate business decisions that successfully exploded. It now uses its massive influence to exert market pressure, but the market does not always bend to its whim because sometimes it does things wrong, some of those products it pushes fail, and I can only assume some products are slaughtered because of projections on their performance regardless of their quality or utility.
I really like Fastmail. It’s just - clean and lightweight. It loads much faster. Has all the features you actually need. No mystery meat. And the UI doesn’t churn every few years for no reason.
You're probably not going to like my answer — I swapped to Hey shortly after it came out, and I'm very happy with it. I check my choice every now and then; recently taking a look at Tuta and Proton. Tuta just isn't there yet, and Proton is just Gmail but in Switzerland. Meh.
Search for one is absolutely horrendous. It used to be great, but not so in the last years. Nowadays it’s filled with spam sites that they don’t seem to be able to filter out. And don’t get me started on the crappy AI overview that hijacks all queries.
Just today I tried a query for water filters and 1/3 of the results were ads. The other third were product pictures, or businesses in close proximity based on my ip. Then there was a box with related products/services, which was completely irrelevant to my needs, a second box with places, yet more product images and so on and so forth. Practically 70% of the real estate of the page was occupied by things I didn’t ask for. All I want is a list of relevant sites to go there and judge for myself. I don’t want Google to spoon feed me.
A lot of the products they killed were promising, it's just that Google just has no stomach for investing in anything for the long haul if it's not going to either capture the entire market or prevent someone else from capturing the entire market.
It's looking like a slightly updated reskin of chromeOS with gemini features built in.
Definitely not a developer machine based on how they presented it in google IO. So if you write software, it's not looking like it'll be relevant whatsoever. I hope to be proven wrong.
And one interesting aspect is the number of children getting these types of neutered machines as their first learning tool. I read another thread comment saying people that started with react actually feel that using straight html is more complicated. My professors say that the best textbook is the one you've read. The next generation is being indoctrinated into this way of thinking
as AI native developer I need VS code forks for AI to be pre-installed. Also every single command or work need to be vetted by AI by default. I am going hardcore now.
The only lesson I'm taking away is that we are still very early in the AI era. AI workflows look entirely different today than they did 18 months ago and I wouldn't bet on them looking the same in 18 months from now.
Scary that your first shot is Google with Gemini instead of copilot and Microsoft who are certainly in a better position to make this happen.
I guess nobody escorts msft to lead anything anymore.
I put Microsoft in the camp of operating system maker, as they are only able to run others' models. The "Give them three tries" meme can play out well for them yet!
I absolutely want AI in my OS. I just want it to be one I can trust to serve my interests, and not the company's. I'm literally in the middle of baking one in as I type this.
You can’t have one without the other. Sorry to be the one to break it to you, but any AI baked into the OS will be so in favour of the company, not you.
Microsoft implementation of Copilot is bad. It’s like they’re trying too hard and the reason they’re probably trying too hard is that they have no real presence in mobile computing, like Intel they have a hard time competing when the playing field is level.
They were too late to the party on music players (which people wanted), mobile devices being any good (which people wanted), and cloud services (which people wanted), so now they're going to be damned if they're going to be late to the party on AI (which nobody wants).
Re: AI OS integration: I recently retired so most of my LLM use is just implementing and fixing fairly mundane OS and networking things along with light scripting for OS automation (AHK) and Home Assistant. So far, I just use web chat and cut-paste to the OS which is fine for little things but it starts to suck after the 15th round back and forth. For example, debugging intermittent Windows crash logs on my wife's laptop by doing multi-line PowerShell incantations from browser chat window, paste into PowerShell window. Cut multi-line error messages back to browser. Rinse / Repeat.
I'm leery about just giving an LLM free run of my laptop, but with reasonable restrictions on which app(s) it can access and how many steps it can do before checking in, and maybe even a throttle on how fast it works, I'd be fine (I'm not in a hurry and I can learn by watching it work at double-speed). It doesn't have to be mil-spec locked down, it's not like I have production code accessible or millions in crypto keys, the biggest downside would be a few hours hosing out and restoring the laptop, which would be annoying but not the end of the world.
I get those that say, "just spin up a VM and run it there", but I 'spin up a VM' rarely enough that the versions have changed and UXs drifted enough that it's exactly the kind of thing I'd actually want the LLMs help to do without me being a cut-paste bot. I'm mostly Windows at the moment and I don't understand why MSFT insists on spamming LLM features everywhere except the one place I'd not only use it, but pay for it. The usage model could be as simple and intuitive as a Zoom remote desktop share with a collaborator. That's already constrained and users have a mental model for the interaction pattern.
I asked Gemini earlier today to search recent user reviews of the latest 'drive my Windows desktop for me' and it reported that the capability is still slow, expensive, and prone to getting lost navigating the interface or interpreting window boundaries etc.
Anyone have any suggestions for my lightweight, casual use case?
Yeah unironically just let an agent harness rip with full admin access without monitoring anything it does or using a VM. It’ll be fine, probably. “How I Learned To Stop Worrying And Love the AgentDOS and Only Exfiltrate Secrets Occasionally”
would "before the major operating systems companies figure out AI integration into their OS that doesn't suck" including running a process with its own address space, like every other thing?
Folks that are interested in a way of doing work locally that doesn't suck, but which integrates LLMs, may be interested in [Barnum](https://barnum-circus.github.io/). The TLDR is that it's a programming language whose frontend is a DSL in TypeScript that is well suited for managing async and parallel work, focused on control flow, from which it is easy to invoke LLMs, and which is easy for LLMs to write. I use it to autonomously ship a very large number of PRs.
I didn’t get a screenshot of this, but I just found a really pointed example of Anthropics lack of craft / rush to build. If you open Claude on Windows, and click Dispatch (under cowork) to start that up, it will tell you that you need permissions windows doesn’t have. When you click the buttons for those permissions, it has broken links to macOS system preferences. I really encourage someone to try it and post the images as a reply as I am writing this from my phone.
The issue here isnt that an AI wrote the code, its that a human didn't spend a lot of time thinking about it. LLM's allow you to rush designs out the door with unprecedented speed.
The VM itself is for Claude Cowork which does all work within the VM sandbox. That doesn't help answer why they spin it up immediately and don't have a way to disable it though. Just the "why it exists" question.
If you're not going to give Claude access to anything on your machine, why are you using Desktop instead of web chat? (Real question, I don't use these much!)
At least in a corporate environment, Claude Desktop is a pretty decent compromise. Preconfigured internally deployed MCP servers and third-party connectors make many of the necessary integrations relatively easy to control.
I use Claude Code CLI myself (inside a VM, to isolate it from the host) for >90% of my needs. For the remaining fraction - email scours, cloud drive searches, other third-party connections - the desktop application is surprisingly decent. I don't even have more than half a dozen connectors enabled. In the VM I have separate, personally managed access tokens available for various third-party services. Wouldn't really try to maintain more than 5-6, otherwise it gets too confusing. [ß]
The desktop application mostly Just Works[tm] with SSO. At least when M365 doesn't suffer from their 4-times-a-day auth outage.
ß: A lot of APIs and authentication systems were designed in the stone age. You either need a 1:1 permissioned access token that can do horrendous damage, or you deal with ultra-granular, confusing and ill-designed scoping jungle where nothing makes sense. Atlassian, I'm looking at you especially. At least an MCP server, provisioned with a reasonably done service account, doesn't have all of your powers to get things wrong with.
The cloud instance certainly runs a full mitm proxy. There are complications for that when dealing with logging, but maybe on the desktop it doesn't log, just running a permissions engine.
I do use Claude Cowork and hence the VM is important, but I also leave the desktop app running all the time since I have many scheduled tasks at different times. The thing is that the VM could shutdown after being idle for some amount of time and then fire back up when you are ready to use it.
Shouldn't it get swapped out of RAM if it's not being used for a long time? My understanding was that modern memory management is very good at this, there's no need to be shutting idle things down and starting them up all the time. I could be wrong.
This whole thing seems kind of silly to me I must admit. It seems obvious that Claude Desktop needs a VM for security for the majority of it's actual real world use. VM's take up memory, yeah. Them's the breaks. If other competitors have managed to provide as good (or ideally better) security scenario with less RAM, that would be interesting, but just complaining about it seems weird and uninformed to me.
There's such a spectrum between "give it everything" and "give it nothing". Imagine you just want to use it to code and want to make sure any commands it runs doesn't mess up your actual machine.
I mean, that's kind of a stretch given how popular and well-regarded Claude Code is at this point. They're not perfect but they seem to be the best out there at this point.
> I mean, that's kind of a stretch given how popular and well-regarded Claude Code is at this point.
Popularity here is irrelevant.
Just because the software is "popular", it does not always mean that the quality of the engineering and their choices is the best. Objectively, it's the model that everyone regards as the best rather than their desktop apps or the harness that drives it.
That is why people want to use the model and their subscription in other harnesses.
In this case, the desktop app is evidently poor with very embarrassing bugs and glitches like this and I expect well funded startups with the best engineers releasing very high quality code.
I can tell it's still early days - it's obvious. What they have today actually kind of sucks compared to what it could be. And I use it every day, and get frustrated every day because it could be so much better.
It's popular because they have the best models and they are burning obscene amounts of money handing out tokens via subscriptions (for now) at a huge discount compared to what the API costs are.
Claude Code itself is incredibly buggy and as we have seen the codebase is a complete mess of slop.
Anthropic has pretty consistently been shitty about how they roll out their software. Extreme lack of engineering rigor and thoughtfulness.
The answer is probably as simple as "no one thought not to do that."
---
I know different people work on these things so I can't do more than guess about how engineering culture cuts across teams, but given the sheer amount of carelessness and sloppiness in Anthropic's software I have to imagine they're burning investor money in training and inference because the code to do it is as bad as the rest of their software.
Rule 1 with making number go up is you eliminate friction at all costs. The user's hard drive is free to you, so there's no reason to gate a feature you want them to use based on that. 98% of them will have no idea you're foisting garbage on them.
RIP, every base model mac from the past 10 years with the <= 256GB SSD. Including the new Neo. When you consider how much of that is eaten up by the system, swap space, caches, reserved space to download OS updates, and apps (2GB a piece is far from uncommon) -- having less than 15GB free is completely unsurprising on that size disk.
I've found that the easiest way to 'remove' the bundle is to delete its contents, then change the permission on the folder, so Claude can't write to it.
Back in the day, personalization / customization was all the rage, as it lets the user feel the control, power and freedom. Now it's the opposite. It's about not letting user to have any control at all. I can't delete some junk apps from my phone and mac, because they are "system" apps. As a non-geek, I can't deal with complexity of the browser and account settings to stop it from what is doing. We are at the mercy of the machines.
It hurts to have all this control stripped away. Once upon a time, you bought iLife (suite of iPhoto, iMovie, etc) on a CD or DVD and installed it. Today, you physically cannot delete the Photos app no matter what.
On my work computer, where I never manage any photos, have no iCloud account and never will, I have to keep this app installed and anytime I so much as AirDrop a png to my computer I am prompted to "Add to Photos" with it. No thank you.
The .app is actually only 41MB, so obviously they've moved the majority of it to some mystery-meat libraries or frameworks installed elsewhere anyway.
Highly recommend trying a linux computer with kde. I have been having similar annoyances with my macbook and switching to kde felt really good. Even though the hardware of the linux pc isn’t nearly as good.
I’ve stopped using Claude on the desktop, just because of how slow the app is to start up and interact with. It’s an absolute clunker; I’m mystified why they can’t ship something that works well given their rhetoric about ai.
They vibecoded it, and admitted as much. Once it was able to self-vibecode, that's all they did. That's why it's written in React and uses gigabytes of RAM as a chat client.
Not only did they decide to write a terminal application in React, but it's 500K lines of code. It's strange because I'm sure Claude is capable of writing a decent TUI in C. It says a lot about the engineering culture at Anthropic, at least on the software side.
Oh, a nice subthread place to vent. Their CLI is so f tragic that it is ridiculous. It keeps scrambling the terminal, scroll and basic shortcuts keep breaking, I've used so many tuis and terminal apps and many of them are a single man operation and a side project and I have never seen anything so bad.
If I didn't know from experience that directed properly claude can be powerful, knowing that they used it to create that CLI would be instant runaway based on very reasonable heuristics - if they are not able to use their product to create a decent piece of software that is not even sophisticated then it seems futile for me to try.
I just do not understand. I feel like most HN could vibe code better claude CLI in claude than the CLI (and certainly just write one) than what we have to deal with to use subscription.
I could not agree more that Claude itself is a janky, hacky, crappy piece of software.
When management at $DAYJOB brought the hammer down and said, "Everyone has to use genAI all the time, OR ELSE," I expected to be blown away by the tool I was avoiding due to ethical concerns, aesthetic objections, humanism, and long-term thinking.
I was born away, but not in a good way.
The CLI is _bad_. I've seen it randomly fail to render anything at all on the terminal multiple times. It has a vim-mode, but it's painfully buggy, and I can literally outrun it - if I try to type too quickly after hitting Esc for normal mode, it just doesn't return to normal mode. It's I was keeping track of the bugs in the Claude TUI, but gave up because it was taking _too much of my time_ to do so.
If nothing else, I'd say Claude shows convincingly that success is not the default for vibecoding.
Yes, it technically does the job, and no, I don't think I've ever used a worse TUI.
Had to make a decision for a TUI I'm working on and opted for curses rather than something like textual. If I wasn't using an LLM to do some of the plumbing I'd probably have used textual to avoid the inconvenience.
There's a lot of opportunity to leverage LLMs to make codebases less bloated and less reliant on complex but human user friendly dependencies that not many people seem to be taking advantage of.
I thought they were all in on agentic coding? They are probably just building at a surface level with only an eye towards shipping, without considering the impact of all the changes. I've seen less and less coordination between engineers as well under that model. If that's the case (Claude Code is this way). it is sort of what you get, no matter the rhetoric about "make sure to review all your changes!" It's always trade offs.
It is surprising that the Claude web app lags pretty easily when using either chromium or firefox on ubuntu linux. Chats that delay my laptop work without issues on my ipad or iphone using the app.
The web app is definitely a bit of a problem. IF there is a native app on desktop or if claude cli is much faster, i haven't tried them.
All of the LLM web interfaces have serious lag when typing after a few turns, at least on iOS safari. I’m talking seconds to start rendering input after typing or when it needs to line wrap the input.
The Codex app slows to a crawl after a few back and forths. In fact it’s so bad that I had to completely switch to CLI, where at least it’s the optimized terminal that takes care of the rendering rather than whatever monstrous electron scaffolding they use
I'm with you. I have the Claude web app pinned as a PWA for quick queries, and then use the CLI for everything project-based.
I did consider experimenting with the Routines feature on the desktop app, but I'm leaning towards whipping together something with cron. I saw another poster here who has a daily PR summary routine that I think would be handy, as I have quite a few repos where I'm a sporadic contributor but would like to keep tabs.
> Claude Desktop spins up a VM without no way of stopping it
I frequently make this error when I talk. My brain thinks of different ways to phrase what I want to say, but when I speak it starts with one and finishes with another. The result is almost always wrong in the way the title is, ie some variant of a double negation.
Sometimes it happens when I type, though I try to read it multiple times so often catch it.
When you realize that in some languages, for instance, in Spanish, double-negatives are not just tolerated, but correct, it helps you to let go of this particular type of pedantry when it accidentally appears in an English sentence.
This question is answered by the post? There is reportedly actually no way of stopping it happen. Perhaps the poster had a brain fart while typing it. Maybe they speak a different dialect of English from you.
Many kinds of double negative are acceptable in many English dialects, and are interpreted as emphasis. The negatives add, rather than multiply. (Though I admit I myself don't speak such a dialect, hence the equivocation.)
I run Claude Desktop inside a Hyper-V VM. My VM doesn't have the "Virtual Machine Platform" feature installed at all. The app accepts this and simply disables the Cowork tab. I wonder if there's some other way to block the creation of the VM to force Claude Desktop onto this code path without having to uninstall Hyper-V.
That said, Claude (both Desktop and CLI) ships on Windows without any sandboxing support for Code. They only have sandboxing for Linux and macOS. If you need to run it on Windows, I really recommend running it in an isolated VM, which then allows you to omit the "Virtual Machine Platform" feature in the VM and solves this issue. The "Windows Sandbox" OS feature provides such a VM without needing another Windows license.
The VM causes observability problems, which corporate platform stakeholders are sensitive to. They will eventually figure out a way to provide an ESR/logging interface to the sandboxed code that is distinct from the user level telemetry, then Defender/Crowdstrike etc will just support that.
I wonder. I spin up tons of projects, but I do it differently.
I'm also on windows. But I'm using a WSL linux vm to do the work. Then I have claude ./sync-to-windows, so the src is copied to a windows share via /mnt/drive/work and I just run a windows built.bat.
I have claude take care of the sync/build.bat so it can do things like clear cache, add options, look at logs, etc. This is also how I build multi platform, build on linux, src works on linux/osx/win
They're some of the only new UIs to be made in the last decade. Almost everyone else stays in the browser (or something close like electron- claude code is actually mostly written in React, they couldn't get far from web dev). The problem is they need to interact with the local filesystem, and not many people have built apps for such a wide range of devices in a long time, and of that small talent pool I bet most are corpo coders- moving too slow and to focused on "the right way" to actually ship more than detailed Jira tickets. They also don't have time for stable releases because competition is so fierce.
But I almost always think of things from a talent-pool-first perspective. Perhaps there are actual technical issues like what Boris was referring to.
Luckily for them, every OS has (at least one) native way of building applications, and with the power of AI they could easily make 3 different desktop UIs, while reusing the same core logic.
If only there was an easy high-level language that's taught to first year students that allows them to write once, run anywhere.
If they're too lazy to learn java, haxe has hxwidgets[0]. Haxe is pretty damn close to js. If a dev can't handle that, they should turn in their keyboard and get a job that doesn't require a brain, like being a senator or federal judge.
There are lots of good answers in this thread but I think it's because they are AI companies and not UI companies. When you look at tools like AnythingLLM, OpenCode, pi, etc. you see all kinds of different interfaces, and while they might make disagreeable choices at least they do it with intentionality and direction.
Because 90% of those UIs is written by these new AI models. That is how they are able to churn so much new user facing stuff all the time. The fact that it works at all is proof that these new AI models are actually pretty decent.
Many people will say it’s because of the slop. I think it’s because they have no product vision. The roadmap is pretty much a random walk, which combined with the velocity of agentic coding is like digging a moat with atomic bombs.
They are moving at breakneck speed deploying on scales most of us can't even imagine. They are working in a space that's completely unexplored where getting information as quickly as possible is preferred above iterating on some feature until it's "done" while your competitor has released fifteen other features, all sucky, but one of which turns out to be a killer and makes a billion bucks overnight.
> They are moving at breakneck speed deploying on scales most of us can't even imagine. They are working in a space that's completely unexplored where getting information as quickly as possible is preferred above iterating on some feature until it's "done" while your competitor has released fifteen other features, all sucky, but one of which turns out to be a killer and makes a billion bucks overnight.
Not on the client, they're not. It's a chatbot as glue between different backend systems.
Understand that it is annoying to spin up a heavyweight VM whenever running Claude Desktop, but I actually think per-app + agent virtualization is the future. Next version of an app is:
1. micro VM
2. agent on the VM
3. software bundled into the VM
Then the agent is totally sandboxed at the hardware virtualization level. It can use the software tools on the VM or write its own. VM can control which software is "frozen" and which is open to agent modification. And VM can also control which services are exposed outside the VM through sockets, HTTP server, X window system, whatever
It's self-modifying apps that are sealed off from touching parts of the computer they shouldn't.
Yeah, Docker Sandbox is in the right direction. But there's a lot of parts that are still not ironed out yet.
How do you package a Docker Sandbox up into an app that can expose UI widgets, with an agent hiding behind them? What widgets is the agent allowed to modify? How do you run a workflow like "give agent all these files, modify the files, and do changeset management on the modifications?"
I'm not 100% sure which part of these will be baked into the application standard format, and which are orthogonal. But current way of packaging up and running these agents doesn't feel right.
I think about this a lot because my startup is building cloud VMs for agents to do code-gen and auto-validate changes, so we have a workflow like:
1. git repo, skills, CLI tools, biz context goes in
2. agent iterates against running dev environment
3. changes go out into git PRs and CI
I think this type of app/agent workflow will expand outside coding use-cases.
Yep, I've been using a local vm-centric agent setup for about 3 months, and it works great. I think there is also value in the fact that with a local VM, you can have the same public IP address, so you're not relying on an EC2 EIP that may be blacklisted somewhere.
...and if a title is incorrect and says the opposite of what's intended, by way of a language misunderstanding or otherwise, it's helpful to note that and get it corrected.
There will be a point in time where they will have to rewrite the whole thing from scratch, they cannot afford to do it now since the race is still going but we will get there, you can never maintain such a product long term especially with the way they are spitting out features like never before.
The weird thing is that this is probably a performance optimization for quick responses when a user asks a question.
My agent harness spins up a VM too, but it spins up on demand, cools down in 10 minutes and warms up when I focus back on the app.
The files it works on actually lives in a mount.
People take more time to type a prompt than the VM takes to spin up on a fast machine and on a slow machine, the cooldown naturally frees RAM back to the machine.
That sounds like a lot, but your music player probably costs that much in RAM anyway.
Seriously, with the current RAM prices we as an industry have to figure out a way to use less of it. Laptops with 4GB of RAM are still common and are going to remain so for the rest of the decade. Spotify using more than 1GB of RAM is obscene.
I also discovered this while
noticing my Mac was low on storage, I only clicked on cowork once and after deleting it from the folder i’m scared to
open the cowork tab coz ik it’ll just fill up the space
Apart from you have no idea what's going on in the VM. It's not as it has a virtual terminal. I'll play the skeptic archetype: What's not to say they're transmitting all prompts back HQ?
Don't be naive and don't think they don't already do this.
Why not ask itself and see what it says about it. "Claude, why are you running in a virtual machine and what are you doing?".
Claude transmits all prompts back to HQ as a part of its basic functionality.
If you are using an AI system to read your codebase from your local folder and make changes, whether or not you have a VM running or not is inconsequential. The Claude extension and/or CLI doesn’t need a VM to send code back to the mothership, you’re already running an executable program and granting it directory access.
Whether you trust a company as a vendor is typically based on their privacy policy, EULA, and your contract with them (if applicable). Those are the bits that have legal enforceability.
I have two friends that are using coding agents on Windows, which was surprising to learn.
Edit: yes, with WSL2 I believe in both cases.
I would have assumed almost everyone would get a Mac/Linux computer to use coding agents because Unix is their "native" platform. It's Bash tool calls all the way down.
Does anyone know a source for reliable data on what coding agent apps devs are using? How many are using Code Claude CLI vs Claude Desktop, etc?
I mean I’m using coding agents on windows, because I’m not just going to learn a whole new operating system just to make robots write code for me.
I want tools that meet me where I’m at, not tools that demand I change up my entire UX to interact with them.
The assumption is not “what’s wrong with Windows that it doesn’t work with <technology>,” more “what’s wrong with <technology> that it doesn’t work with Windows”
Why wouldn’t you want your thing to be cross platform
Is it just me, but this feels like Claude gets to have a nigh-impenetrable black box right on your machine and you have no idea what is going on inside it.
After all, the last time I encountered Hyper-V it was in the context of copy protection that prevented crackers from observing or interfering with video game protection
And if they didn't users would instead be whining that it ran rm -rf on their root directory. Sometimes it seems like the people here just want to act like insufferable neckbeards for no good reason.
and on my Mac any time I accidentally click Cowork which I don't use whatsoever, it re-makes the same VM, without asking me. It's one of the dumbest things ever. You're about to hijack nearly 20GB of my storage (which gets eaten up as it is) and you don't think to ask me if I even want the VM before you shove one into my system?
I can’t reconcile the super intelligence koolaid with the objectively bad performance of Claude code. Surely they could have used their “too powerful to release” model to vibe better performance? with a few skills thrown in for good measure. Heck, even throw in an auto research loop
A lot of the commenters here say they’re running Claude or other harnesses inside a VM or with various permissions and levels of access.
Am I weird or missing something using pi as my regular harness with gpt models or kimi in essentially yolo mode with mostly all system access? I haven’t experienced negative consequences of this.. yet.. and I don’t know I’d if I will? I don’t think I’m ever letting an agent run For more than 5 minutes before it’s done with the current small task.
I run totally unprotected with gpt5.4/5. I've been through thousands of dollars worth of API tokens through both copilot and custom harnesses that have local admin and arbitrary powershell access. I've never seen anything that could even remotely be construed as malicious.
I see a lot of people making a really big deal about safety and sandboxing while I'm busy getting shit done. If you can't handle your current source code checkout getting screwed up by a bad prompt, that's on you 1000%. Source control is the answer for anything information over time.
Unless you intentionally try to make a scene, these models aren't going to go fuck with your system shell or do anything you couldn't recover from in a few minutes. Connecting chatgpt to the enterprise sql server as sysadmin is not what I'm advocating for. This is another example of "on you, not the AI". There's a tiny amount of nuance you can apply at the edges that makes it easy to allow broad access with negligible risk.
This all feels like a race where the model companies try to solve doing work locally in a way that doesn't suck, before the major operating systems companies figure out AI integration into their OS that doesn't suck. It also makes me wonder why Google which has both Gemini and Android can't figure this out, and if there are lessons to draw from that.
Google is historically terrible as a product company (and has succeeded in spite of that) As their technical innovations become less of a moat (we're already there) they won't be able to win on engineering alone (they are no longer winning on engineering alone)
But Kubernetes, being a former Google project whilst donated to CNCF, is still one of the mainstream container orchestrator if not the biggest one (I believe we need to go over to supercomputers for that). Nomad is kind of dead and Apache Mesos is basically dead. Leaving Kubernetes as a natural moat, plus Google employees are still seemingly actively maintaining it.
And also Go. While I'm not a Go guy (speaking as a C# and Rust guy, I did wrote a good amount of Go before) it has a huge dominance in Cloud-Native application. For one, Zitadel, an alternative to Keycloak, is written in Go and only takes a fraction of what Keycloak needs.
Flutter/Dart is catching up, but the ecosystem is still relatively weak.
Kubernetes and Go are not moats of any sort; Google doesn't make any money off of them, and they doesn't get them any useful user data like Gmail, Docs, etc. do.
>Leaving Kubernetes as a natural moat, plus Google employees are still seemingly actively maintaining it.
Neither Go nor Kubernetes are even remotely moats.
Kubernetes makes me want to go running to WebSphere 5 XML configuration files (version 5 is relevant here).
Whereas Go only exists because a bunch of renowed Oberon and UNIX heads did not want to keep using C++ for their work, and got twice lucky, first with their line manager, secondly with Docker and Kubernetes getting adoption after their rewrites into Go.
> But Kubernetes, being a former Google project (...)
You are confusing Kubernetes, the software project with Kubernetes, the container orchestration system.
There are many Kubernetes distributions out there which are not maintained in any way by Google. Even Canonical provides its own Kubernetes implementation.
Nowadays Kubernetes is a keyword much like Unix or Unix-like.
Honestly, Kubernetes has no moat. Give me a better, more simple alternative any day.
Please, Kubernetes is the Chlamidya of infrastructure software. Ever hear of a large K8s migration going well? No, because it needs to be completely re-implemented under the hood for anything halfway serious. Mesos was pain in the ass to use, but was mature when K8s arrived. Nomad was cool though.
Dart is also complete shit: Speed of javascript with verbosity of Java. Who the hell though that was a good idea?
I've seen lots of successful large k8s installs. This same argument from the nomad crowd comes up constantly. There's a reason why nomad lost, and it's not because it's better.
> Ever hear of a large K8s migration going well
...well, AI infra stacks are more or less k8s/cloud native, especially on the inference side. Nvidia GPU operator plus KubeFlow makes deploying models easy, and I did that manually as well using just the ollama, llama.cpp containers and especially vllm operator. I'm not sure about what's the measuring notation of "large" is, but OpenAI had a blog post about how to manage 7500+ nodes Kubernetes (https://openai.com/index/scaling-kubernetes-to-7500-nodes/). That's 5 years ago and I speculate it will only be even more.
> completely re-implemented under the hood for anything halfway serious
...an example please? I'm building a platform to deploy hundreds of open source apps on Kubernetes, you know, the boring thing you can do with a VPS and maybe use Docker Compose to start -- with an ergonomic twist. I relied on so many features of what Kubernetes and its ecosystem provided, especially with persistent volume, volume snapshot, CRDs and cron job, and all of it is just composed from open source and cloud native software...I'm not sure if that sounds "half way serious" to you.
I know those boring monolithic apps can be done with a VPS, as I exactly came from that background, and I know what the shitty points of having just a VPS are. You don't have a clean control plane, you don't have HA, you don't have distributed storage, and you have to be super aware of the apps that you are deploying with Docker Compose.
What you would think "completely re-implemented under the hood for anything halfway serious", perhaps it means that you don't want to get in the fuzz to manage the complexity and prefer to trade for a far simpler but more primitive solution, and that's totally fine.
Pro tip: Codex seems to love generating great Helm charts so much for some reason unknown. I tried GPT-5.4 high in codex and it easily beats Opus 4.7 max on my own internal helm chart generation benchmark evaluation, which measures HA, deployment and app-specific probes. I've given source code to both Codex and Opus and research about the app config structure, and Codex did so well to just generate secret key-value and convert it to JSON and environ, while Opus insisted on using a dynamic generator at runtime.
> Nomad was cool though
Ahem. I used to deploy Nomad too, but I found it too underwhelming, especially with the networking side of thing.
Consul is perhaps one of the most hated thing in my entire career that I would call it bullshit. I know that later Nomad has its own Raft mode, but the Consul brain rot to me is already very sickening such that I don't want to touch Nomad no more. Vault is fine though, but I prefer OpenBao now.
> Dart is also complete shit: Speed of javascript with verbosity of Java
I don't get where the "speed of javascript" is from. Dart/Flutter is JIT on dev mode, and AOT compiled on production, is that you didn't get the compiler options right?
As a person who gets paid to make Chrome (CEF really) do its bidding, I would say Chrome is really as close to an OS as it can get, as in I've found API or service typically an OS or an external tool would provide, that wasn't built into Chrome.
The Chromium source code is used to build ChromeOS so you're spot on.
They did build chrome, gmail, docs and search afaik so that is a pretty impressive list imo.
I don’t specifically like these but they are much better than microsoft for an example
Yes, two decades ago.
Docs was acquired (writely)
And throughout those two decades they also canned a ton of projects, including maybe some good ones.
I’d say they are batting .250 like everyone else.
Could they do it today? Or have they squandered the trust they created?
“Don’t be evil”
On a semi-related note, I bought a Pixel phone about a month ago, and I'm shocked by how unpolished it is. I've had so many little annoyances pop up, issues I never had on other android phones. Keyboard hiding/appearing when it's not suppose to, bluetooth dropping, WiFi dropping, network switching taking forever, screen becoming unresponsive... It's mostly all small things, but they really start to add up after a while.
Mine's been mostly fine. Second hand Pixel 10 (previous 2nd hand 8a) and I am extremely happy with an excellent phone. No problems like you've had.
Interesting. My pixel experience has been the opposite; very polished and pleasant. 10A. Wonder why the variance?
Same. I was excited for a Pixel (assuming a much better Android experience) but had same experience as you
People say this but it is wishful thinking. What are some actual metrics here we can use. Let’s compare google’s products with others.
> Let’s compare google’s products with others.
Gmail: Reliably excellent spam filtering (for most users).
Outlook.com: Reliably excellent at losing validation emails.
that said, Gmail finds lots of other ways to inflict misery (Gemini beat-downs) and I'd rather work in Excel online that Sheets.
lmao this is a terrible take. Google is the only company with AI deployed in every market where AI can exist, and _all_ of their products in those segments gush money.
?
How are Google products anything but outstanding in their categories? What are you comparing to?
Which products are outstanding in their categories?
Gmail isn't outstanding, search isn't outstanding, maps isn't outstanding.
They are all pretty par for the course. Google used to be outstanding... but I'm not sure of a single product they have that is outstanding (def: significantly better than the competition) anymore. On the other hand I rarely use any google products these days, so maybe I'm not the one to be judging.
I'm not a fan of Google, and also not attached to Apple or Microsoft, so this isn't me trying to stan for Google, but I'd like to request that you give examples of what competing products are categorically better (and, by what metric(s) you're judging - code quality? stability? robust set of features?) -- for Gmail, Docs/Drive/etc, Google Calendar, Maps, Classroom, YouTube.
As far as I can tell, if judged by the marketplace (and breaking ties with which product I like better), Google has run away with the ball on all of those, and Gemini seems to at least be competitive.
The only major product I'd say they've sunk below acceptability on is Search, which is demonstrably dogshit now...though I suspect it's more that they have changed their definition of what Search is for, from "helping users efficiently find other websites that are useful to them" to "A convenient on-ramp to, many times per day, capture the current user intent and steer them toward something that earns Google some ad revenue."
The major differentiating factor that Google has had in every product category is that their products are free and you have to deal with ads (and they monitor your behavior for profiling you and your interests).
GMail and Google Maps were revolutionary when they came out, sure, but the vast majority of Google's products now are... fine? at best? And a lot of their "big products" were acquisitions that they absorbed in order to further the core goal of the business - to organize all the world's information and use it to serve ads to people.
Meanwhile, Google has a litany of products they've started internally, launched, ran for a while, and then let stagnate or canned entirely; anecdotally I've heard that this is because your bonuses at Google hinge on your ability to launch a product and not your ability to support a product, so it's beneficial to get something launched and then immediately leave to go launch another project rather than polish the one you just launched into something to be proud of.
I'm not sure if that's true, but it would certainly explain a lot; if Google launches something and it's bad or it doesn't click, they just give up on it. Google Wave, a half-dozen chat apps that I can think of, Stadia, and dozens of others. Things that Google launched, which had problems or didn't hit mass adoption instantly, and then just petered out and were retired with all of the time and energy and money put into them arguably wasted - products that people wanted, and wanted to succeed, but which weren't revolutionary successes at launch so they weren't worth further investment.
Meanwhile, they (and most of the industry) are pushing AI for some reason despite the fact that almost no one actually wants AI to be the only way that people interact with information.
This all reinforces what I've been saying about Google for decades: they're not creating things that users want to use, they're creating things that they want users to use. Sometimes those things align, but when they don't then it's not worth further investment (except, apparently, AI).
I just don’t think your opinion is shared by most people.
Gmail is the most popular email service in the world, people are always telling me how they prefer Google Docs over everything else and their only competition is Microsoft.
Yes it’s free but there is no other service that I rather switch to, and I actually pay for additional storage.
> Gmail is the most popular email service in the world
That's because it's been around for quite a while, and for a long time it was the best webmail service. It's also free, unlike most alternatives. And switching to a new provider means a new email address, unless you're using a custom domain with Google Workspace (or whatever they call it these days), which is a small minority of personal accounts.
(I gave up on Gmail a few years ago and switched to Fastmail, and like it much more than Gmail. But I'm the rare person who is willing to pay for email, and had been using a custom domain with Gmail, so my non-monetary switching costs were minimal.)
You're absolutely right that Gmail was the best webmail service when it launched, and for some time afterwards. How many people now commenting on HN remember when Gmail launched? Remember how revolutionary it was at the time? Every other email webapp, when you clicked on an email, would refresh the page. Gmail, when you clicked on an email, did not cause a browser navigation. It simply replaced the page contents with the contents of the email.
We're so used to setting webapps do this that we take this for granted, but Gmail was the first email webapp to do this. It's possible it was the first webapp, period, to do this; I feel like Gmail's use of XmlHttpRequest was innovative at the time.
Fast forward twenty years, and what about Gmail is innovative today? Nothing that I can think of. It's mediocre (there are lots of filtering improvements they could make that they aren't making, for example), and everything that made it good has been copied by other webmail clients. There's no particular reason except momentum to stay on Gmail.
Gmail spam filtering also used to be revolutionary and an unsung hero. I haven’t put effort into finding out if other options have caught up with that (because of aforementioned tedium of changing email addressed)
I have had a pobox.com email address (just a forwarding one) longer than I've had a Gmail one, and their spam filtering was pretty amazing too. Even before I set my pobox.com address to forward to Gmail, I never saw very much spam.
Now that Pobox is owned by Fastmail, I rather suspect that Fastmail is going to have the same good spam filtering. Can't speak from experience, though, as I haven't actually used my new Fastmail account yet (it still forwards to Gmail, and so far I haven't switched. Momentum, again).
You can’t beat free. The Fastmail web interface is snappier than gmail. And you can’t beat dedicated mail clients like thunderbolt in terms of workflow.
Google doc is wordpad level with very good collaboration (but that’s mostly what people need). People were fine with typewriters, so they are fine with a word processor like google doc. But it’s not at the level of even Libreoffice or Apple’s page in terms of features.
By any definition of good usability, Gmail is not good and Google Docs are not far behind. It’s not that they are functionally bad, just really poor UX.
That's hyperbole. They have flaws, but at the very least, when they were launched, they were arguably best in class. I'm not sure how much me sticking with them is due to familiarity and muscle memory but I know they won we over purely on merit in the beginning.
As someone who was an original invitee to Gmail it was the clarity of function that was the differentiator. They were “grown up” and acknowledged user agency vs their competitors.
But as others have mentioned, they operating model of Google as a company incentivises creating products but does not incentivise refining it. Gmail has gotten far richer in functionality but at the same time the interface has gotten far less consistent. Their competitors (mainly Microsoft but not only them) also got richer functionality, but they also paid attention to UX. While none are perfect, there are definitely some better than others. Familiarity definitely breeds inertia though, I’ll grant that.
We probably talk to different sets of people, as I don't even know anybody who ever used Google Docs.
I've been at multiple companies where Google Drive and Docs/Sheets are the only thing people use.
That's because the execs force them.
LMAO Why the fuck would I wanna install office on my mac.
Woosh
No. That wasn't a whoosh. Or even a woosh.
Oh, yes, typo aside (which is the lowest form of critique) it was.
It's not about why this or that person would want to install the MS Office on their Mac (millions do after all). So the pointed flew above their head.
It's about the point that many people having to use Google Docs are doing so because corporate wants them to. (In fact most people using MS Office do it because it's a work requirement too).
That's really surprising to me.
You should checkout startups. They would blow your mind.
> GMail and Google Maps were revolutionary when they came out, sure, but the vast majority of Google's products now are... fine? at best?
Is that... good? I mean take maps -- what more can possibly be done to that product that wouldn't just make it worse? It's done. The fact that's the default choice for mapping and just works is fantastic really. There aren't any competitors doing anything revolutionary either because there isn't anything revolutionary to be done.
Maps is far from done. At the very least it's still riddled with usability issues. One bug-bear I have in particular: when I zoom into a very specifically chosen area, search for a Chinese restaurant, and it zooms out to half the state. Maddening. And it's rife with problems like this.
I’ve never experienced that issue - which platform/what location? I just tested and I was unable to replicate that issue.
I worked at google for 3 years and can confirm what you've heard. Obviously, every org is somewhat different.
I think it’s a major feature gap that Gmail (paid or free) cannot create filters on headers.
I also can’t do wildcard filters on “to” or “from”. For example, in my GApps I have it set up to route all emails not associated with a specific user to my primary user. So that it’s easier to make throwaway emails. I want to filter all to:`X.X@domain.tld` to a certain folder. No can do.
It just feels restricted.
> I want to filter all to:`X.X@domain.tld`
You can filter based on the to: field, yes.
For many years I’ve been creating filters on free Gmail for to:, from:, subject:, etc. I set them up on desktop web.
Perhaps there is something more specific you’re trying to do?
> a major feature gap that Gmail (paid or free) cannot create filters on headers.
You can create filters on header fields like from:, to:, and subject:, so I am guessing you mean something different than “cannot create filters on headers”?
I think he refers to mail headers. Those are normally hidden away from you, at least on a PC browser you can see them:
1. Open the specific email.
2. Click the three vertical dots (More options icon) next to the "Reply" button.
3. Select "Show original" from the dropdown menu.
Mail headers also include to:, from:, subject:, etc, as well as more obscure items too, which is why I think OP commenter meant something very different than “cannot filter on headers”.
Also, more items that might help OP (as I can’t edit parent comment) - they mentioned wanting to use wildcards on to: field. Those header fields do allow specifying just part of the header, like just the domain, or one part of the to address. (But those match at word boundaries and I’m not aware of being able to match sub parts of words or more complex items.)
Regardless, I don’t think I’d call this a “major” feature gap - maybe minor or more of a niche feature.
This so much this. Gmail overall is pretty meh and then it has these stupid footguns that make it awful.
Funny. Search is the only thing that is outstanding as it is the big revenue arm, that and youtube.
The last time i tried using gmaps i got ads and the thing could figure out where i was on the roads. It was comical as i always remembered google maps being better than apple. Today tho, apple beats them hands down.
Googles products that do not get cancelled are pretty mediocre in todays market. They can build useful things but if it doesnt have ads in it, it gets axed
> the thing could figure out where i was on the roads
Is this missing a "not"?
Yes.
> but I'd like to request that you give examples of what competing products are categorically better
Personally I much prefer Fastmail to Gmail. The site is way faster and more cohesively designed. Fastmail supports jmap, and way more imap extensions (including push support on Apple mail). They have helpful humans handling support requests. And they do all of that with what seems like 1/10th or less the number of employees.
The only thing I like more about Gmail is their native mobile apps. Fastmail’s official mobile app is a web view.
> Gmail, Docs/Drive/etc, Google Calendar
MS is the overwhelming favorite in each of these markets if you only consider paying users.
Has hacker news lost it?
Maps & Gmail & Search all have plenty of accumulating flaws... but they also completely defined their product category and today are among the most popular software products ever made.
"defined their product category" means it used to be outstanding, not that it currently is outstanding.
"among the most popular" doesn't need either of those to be true.
As a concrete example, Windows defined the product category. And it is still the most popular desktop operating system in the world by far.
Few here would argue that it’s an outstanding product.
More than plenty of people (including me) would argue that Windows is an awful product, but a quite significant number of people (maybe even including me) would argue that its an outstanding product.
>but a quite significant number of people (maybe even including me) would argue that its an outstanding product.
Are those people in the room with us now?
I hate Windows with passion and can still acknowledge that it is an outstanding product.
Wow I wonder why maps and gmail are extremely popular by the company that controls the largest browser, search, android, and advertising. It's based solely on their merits and not abusing their monopolistic position to thwart competition right?
They are popular because they are free.
I have to use Gmail at work and it is just terrible.
Garmin was and is better than Google Maps and Mapquest was better than Google Maps when you needed to print directions. If Google didn’t have Android would maps matter as much?
Apple Maps and Waze is better for directions. Apple has better CarPlay integration and HUD. Google Maps is way better at searching for things like restaurants or local businesses but not as much the nav part.
> Gmail isn't outstanding, search isn't outstanding, maps isn't outstanding
They were at the time.
I think that is very much GPs point.
Is there a better Maps app?
I guess those gaussian splats on Apple Maps could be p. neat.
Apple Maps is - or soon will - show ads. A trillion dollar company that charges thousands for the device. So that’s out.
OsmAnd’s UI really doesn’t suit me, and there are a few others that I personally liked even less.
But Organic Maps ticks enough boxes to come close. I’ve been using Organic Maps on iOS for driving, and it’s tolerable. And for offline mapping it’s a godsend.
No. I just switched away from Apple Maps too when they added ads.
If I’m dealing with adware either way, may as well use the best.
I've been using Magic Earth, which is paid, but has worked pretty well for me so far.
Bing still has maps, although I wouldn't say its better.
Time moves on …
… sure, but until someone creates a competitor that is good enough to overcome the switching costs, people are going to stay with Google. Most of Google's then-competitors in the categories listed are not just inferior products, they're dead. ISPs no longer do email, Alta Vista & Ask Jeeves are gone, MapQuest is a thing that makes us that used it sound old.
Not only does an upstart have to overcome the switching costs, they have to actually survive, and not just get hoovered up by acquisition and then Our-Incredible-Journeyed.
Search has degraded for sure, but still better than anything else? Maps - I guess you mean Apple ones are better? Can't tell, I am not on Apple, but if you don't use Apple products, there are not many alternatives to Google maps
MapBox[0] does a good job. I think they use OSM maps.
I don’t think it has a public interface, though. It’s really a developer resource.
[0] https://mapbox.com
I think you’re talking about consumer products.
At the enterprise level, if you know of something better than Bigquery, please let me know.
Similarly, Kubernetes and Kubeflow are both outstanding - and Licenses Kubernetes has no meaningful competition for what it does - but Google did everyone a solid by making them open source, so you can get them from other sources than Google. But the Google managed versions are certainly extremely good.
As for the idea that Gmail, search, and Maps aren’t outstanding, an easy way to refute that is to ask what the outstanding alternatives are. I doubt there’s a single list that many people would agree on.
What is better than Google Maps? It's the only Google product I don't see myself abandoning any time soon. I agree on everything else.
All of these are outstanding! In so far they are not singular or new anymore, well... If for the past ~20 years nobody has come up with something clearly better, then I would say that speaks to how outstanding the product that are being copied are to this day.
Google Cloud is great, compared to legacy and fragmented AWS
Yeah it's great until they absolutely destroy you like Unisuper or Railway.
Support is part of the package when it comes to product and their support SUCKS.
I would absolutely NEVER use GCP for any business I was in charge of. Google cannot be trusted.
I tend to agree :/
I agree, but it’s definitely getting worse. It’s a lot less focused than it was.
What's wrong with Maps?
It's … okay … but it still falls down in a fair few areas? It's crap at finding restrooms. Finding a stop on the road is also difficult, as it seems like it just defaults to a basic radial search, when as a driver you want things down-route, not radially out. All the AI in the world can't seem to figure out when I'm looking for gas or food that closed businesses are not results I want to see. It eats enough CPU to melt phones, such that Android now has built-in support for this?!¹. Attempting to report things often goes in vain². Some of the notifications need work ("object in road ahead" … I'd kill for what lane! this one is just anxiety in a notification), and it'd be nice to see the lane designations ahead of time (it only shows them once you're like <1mi out). I've never gotten the AI-home detection to work. Attempting to navigate to the house of anyone with an Irish name gets me a bar, and the forced-voice-navigation when in a car means I have to be able to pronounce the destination. Google does not seem to grok that sometimes … there's a person in the car who is designated navigator. They can type, it's fine. Some turn directions could be better if you incorporate more precise language into them³. Some directions could be abbreviated "Navigate to I-4 North": I live here, I don't need step-by-step hand-holding to the interstate, but I'd like to plug in the destination before the car is rolling.
¹literally, phones can now demand you put them in A/C b/c they're dying
²I reported once that a jetway was 3D modeled as being like 8 stories high. Google couldn't confirm that, and closed the request. I reported a business as not being present, while my GPS showed me as being at the alleged address, that also couldn't be confirmed. My GPS trace would have seen me walk the whole block, twice!
³as designated navigator in my relationship, I can tell her "leftish" or "rightish", and she understands what I mean. Where I live a lot of the intersections' designs appear as if a civil engineer was given artistic license, and so sometimes the direction is "5-way intersection, left-ish". "Left" is a bad direction when there are two lefts. Of course … me & her have developed a fairly extensive lexicon over years of long road trips, too.
> It eats enough CPU to melt phones, such that Android now has built-in support for this?!
It has support for phones overheating in the sun. I don't think any phone can get to overwhelming temperatures by itself.
After 20 years of consumer GPS goog offers Sub-Garmin quality navigation service, and a generous helping of UI non-intuitives.
What about showing tractor roads, military compounds, or farmer fields as roads I should be driving on?
I find Apple's Maps directions to be slightly better, nowadays. They're more intuitive.
... 20 years of not remembering that I only ever want to see distances in km
I'm no google fanboy, but gmail is a very good web-based email, and google maps is a very good web-based map program. I would say "outstanding" with no reservations.
Ok. Apparently you missed the question from OP.
What are you comparing to?
The Pixel series outside of security (to which their own flavor of Android doesn't even take advantage of like we see with GrapheneOS) doesn't have any particular outliers that would make it any more or less enticing than another company's phone.
Their ChromeOS hardware was nice but had lackluster software and by the time it was EoL'd, never got the love of ChromeOS-present.
Google TV generally gets outpaced by onn (Walmart's brand) on cost and value proposition.
And also the fact they have shown time and time again that they just kill products over and over again.
A big part of why Stadia was cancelled is because it didn't get traction, and a big part of why it didn't get traction was because of how many people assumed it was just going to get cancelled.
Ah, hardware products.
Other than search, in its heyday of the early 00’s, every google product success was either a 20% time project (e.g. Gmail) or an acquisition (YouTube) or a direct clone of someone else’s working product (android).
I think both can be true. Google has a history of annoying churn while still being good enough (or just … being large enough) that switching to competitors is still too high a cost for most.
For example, their "chat" app has churned 3? 4? times now? Their assistant app has churned from whatever the OG assistant was to now Gemini. Wave churned to "+" in the social category, and that's dead now.
The default placement in Android probably helps a lot, or other things, like forced signups into adjacent products (e.g., like + was doing for a while).
> For example, their "chat" app has churned 3? 4? times now?
I believe they’ve had at least 58 different products with chat / messaging.
https://www.reddit.com/r/google/comments/s2s2ld/all_of_googl...
Have you tried to admin a large team using Google's admin? :(
Do they have any supported way to export a user's account (e-mail, calendars, etc) for offline archiving yet? I used to have to reset their password, disable their 2FA, log in as their user, initiate a 'Takeout' request to export their account data into an archive, wait until the request was done (between minutes and days depending on the account), download it manually (often in chunks if it was large enough), store it somewhere, and then delete it and delete the account.
I can't imagine that no other 'Google Workspace' organizations want to actually save their employee data rather than irrevocably delete it forever.
Most of their best products are VERY old. Gmail, Search, Docs/Drive, Maps, Street View, Chrome, all solid but old consumer apps. A lot of early entrants/market defining products, too! Most of them are still very good. Calendar is... fine. Gmail/Calendar as corporate products in some ways feel like downgrades compared to the functionality of the older Office desktop suite, but they integrate nicely into a lot of things.
The bad rep Google gets now is because while they've polished their nice money machine very well, they haven't done much to pioneer new segments and the old stuff is pretty stagnant. When's the last time Gmail gained a must-have feature? Maps?
Ok but maybe those old products are just mature and there's not much room for product innovation? Even if we assume that's the case (though there are third party email and calendar apps that I use instead of Google's first-party ones on my phone), let's look at some newer stuff:
Android was a copy of Blackberry that turned into a copy of Apple before being launched. It's done well numbers-wise, particularly in lower-cost markets, but hasn't got a lot of novel product stuff going on in the last decade or more. Less luck as a tablet platform. A lot of the nifty novel stuff in mobile space (wireless earbuds, "find my" stuff, mobile payments, magnetic wireless charging) was done first by Apple.
Meet vs anything else - well, Meet is cheap if you're already on Gsuite, at least.
Voice/Hangouts/Chat/Duo/Whatever - whole lotta abandonware or dead attempts.
Google Plus - famously mis-targeted product.
Google Assistant - Siri copycat then with Alexa copycat hardware. Didn't do anything novel.
Chromecast - nice novel product, caught on decently well, lost the market to a combo of set-top boxes (Roku, Apple) and native-TV integrations. Tried doing Google TV as that native platform with a copycat pivot vs the original Chromecast, no particular success.
Stadia - abandoned quickly, never set itself apart from the things it was copying.
Smart home hardware (like alarms and smoke alarms) - dramatically scaled back and now pretty unambitious.
Gemini - copycat of ChatGPT which is particular damning since so much of the original research came from Google in the first place!
Arguably, "exceptional" products are not ones that can vanish on a whim, like a great, great many of Google products have. Or they actually compete with other products in the same space, like a great, great many of Google products have not. Also, one would argue a good product is not one that is bought out and then deliberately destroyed to prevent its expansion into or development of a market for itself. Google is an advertising company with tremendous reach because of a handful of very aggressive and very fortunate business decisions that successfully exploded. It now uses its massive influence to exert market pressure, but the market does not always bend to its whim because sometimes it does things wrong, some of those products it pushes fail, and I can only assume some products are slaughtered because of projections on their performance regardless of their quality or utility.
https://killedbygoogle.com/
An email client (Gmail app) that is 500mb? What’s _outstanding_ about that? Almost everything Google makes now is terrible. Try some alternatives.
Honest question: what's a candidate for better? The search needs to be really good, as that was always a strong point for Gmail.
I really like Fastmail. It’s just - clean and lightweight. It loads much faster. Has all the features you actually need. No mystery meat. And the UI doesn’t churn every few years for no reason.
You're probably not going to like my answer — I swapped to Hey shortly after it came out, and I'm very happy with it. I check my choice every now and then; recently taking a look at Tuta and Proton. Tuta just isn't there yet, and Proton is just Gmail but in Switzerland. Meh.
The about 7 different text chat applications they had?
At some point GoogleTalk was one of the leading global text messengers, and then it was basically destroyed by Google itself.
Depending on what you mean by 'chat', Google has had a lot, but this list includes a lot more than 7 text chat services. https://www.reddit.com/r/google/comments/s2s2ld/all_of_googl...
> The about 7 different text chat applications they had?
All of them named "Hangouts" no doubt.
Search for one is absolutely horrendous. It used to be great, but not so in the last years. Nowadays it’s filled with spam sites that they don’t seem to be able to filter out. And don’t get me started on the crappy AI overview that hijacks all queries.
Just today I tried a query for water filters and 1/3 of the results were ads. The other third were product pictures, or businesses in close proximity based on my ip. Then there was a box with related products/services, which was completely irrelevant to my needs, a second box with places, yet more product images and so on and so forth. Practically 70% of the real estate of the page was occupied by things I didn’t ask for. All I want is a list of relevant sites to go there and judge for myself. I don’t want Google to spoon feed me.
Maybe you are not counting the products they kill.
A lot of the products they killed were promising, it's just that Google just has no stomach for investing in anything for the long haul if it's not going to either capture the entire market or prevent someone else from capturing the entire market.
Outstanding?
Well ...
https://killedbygoogle.com/
They are releasing AluminumOS with their Googlebooks, which is a AI forward OS. If its good or not we have yet to see.
It's looking like a slightly updated reskin of chromeOS with gemini features built in.
Definitely not a developer machine based on how they presented it in google IO. So if you write software, it's not looking like it'll be relevant whatsoever. I hope to be proven wrong.
If everything is in the cloud and you are just prompting agents to code for you, what exactly is “a developer machine”?
I'm obviously not just prompting agents for everything. What are you on about?
Why would I build my little web-apps and backends in the cloud when I can run things faster locally?
And one interesting aspect is the number of children getting these types of neutered machines as their first learning tool. I read another thread comment saying people that started with react actually feel that using straight html is more complicated. My professors say that the best textbook is the one you've read. The next generation is being indoctrinated into this way of thinking
Up until they stop subsidizing things and you have to pay what things actually cost.
> Why would I build my little web-apps and backends in the cloud when I can run things faster locally?
Because in a lot of companies, your machine is actually just a portal to a remote desktop.
The Venn diagram of “Corporate” vs “Company” definitely has VDI and ServiceNow at the center of it.
Or shudder BMC Remedy
Not in my personal experience. Never been in that situation, and I've worked in full-stack, FE, and BE teams.
How many companies with say 50k+ employees? VDI is pretty standard and often even required by some cyber security insurance.
Every Fortune 500 basically
Because timesharing is back, big tech only wants to sell terminals.
as AI native developer I need VS code forks for AI to be pre-installed. Also every single command or work need to be vetted by AI by default. I am going hardcore now.
"What do you mean, an Aluminum Falcon?!"
"WHO'S 'THEY?????'"
The only lesson I'm taking away is that we are still very early in the AI era. AI workflows look entirely different today than they did 18 months ago and I wouldn't bet on them looking the same in 18 months from now.
Scary that your first shot is Google with Gemini instead of copilot and Microsoft who are certainly in a better position to make this happen. I guess nobody escorts msft to lead anything anymore.
I put Microsoft in the camp of operating system maker, as they are only able to run others' models. The "Give them three tries" meme can play out well for them yet!
Don't you read HN? Nobody wants AI in their OS, especially in Windows. Common complaint that Microsoft is forcing AI into every corner of Windows.
I absolutely want AI in my OS. I just want it to be one I can trust to serve my interests, and not the company's. I'm literally in the middle of baking one in as I type this.
You can’t have one without the other. Sorry to be the one to break it to you, but any AI baked into the OS will be so in favour of the company, not you.
I'm running open models on my own hardware, on linux. So, I don't think so?
Microsoft implementation of Copilot is bad. It’s like they’re trying too hard and the reason they’re probably trying too hard is that they have no real presence in mobile computing, like Intel they have a hard time competing when the playing field is level.
They were too late to the party on music players (which people wanted), mobile devices being any good (which people wanted), and cloud services (which people wanted), so now they're going to be damned if they're going to be late to the party on AI (which nobody wants).
Re: AI OS integration: I recently retired so most of my LLM use is just implementing and fixing fairly mundane OS and networking things along with light scripting for OS automation (AHK) and Home Assistant. So far, I just use web chat and cut-paste to the OS which is fine for little things but it starts to suck after the 15th round back and forth. For example, debugging intermittent Windows crash logs on my wife's laptop by doing multi-line PowerShell incantations from browser chat window, paste into PowerShell window. Cut multi-line error messages back to browser. Rinse / Repeat.
I'm leery about just giving an LLM free run of my laptop, but with reasonable restrictions on which app(s) it can access and how many steps it can do before checking in, and maybe even a throttle on how fast it works, I'd be fine (I'm not in a hurry and I can learn by watching it work at double-speed). It doesn't have to be mil-spec locked down, it's not like I have production code accessible or millions in crypto keys, the biggest downside would be a few hours hosing out and restoring the laptop, which would be annoying but not the end of the world.
I get those that say, "just spin up a VM and run it there", but I 'spin up a VM' rarely enough that the versions have changed and UXs drifted enough that it's exactly the kind of thing I'd actually want the LLMs help to do without me being a cut-paste bot. I'm mostly Windows at the moment and I don't understand why MSFT insists on spamming LLM features everywhere except the one place I'd not only use it, but pay for it. The usage model could be as simple and intuitive as a Zoom remote desktop share with a collaborator. That's already constrained and users have a mental model for the interaction pattern.
I asked Gemini earlier today to search recent user reviews of the latest 'drive my Windows desktop for me' and it reported that the capability is still slow, expensive, and prone to getting lost navigating the interface or interpreting window boundaries etc.
Anyone have any suggestions for my lightweight, casual use case?
Yeah unironically just let an agent harness rip with full admin access without monitoring anything it does or using a VM. It’ll be fine, probably. “How I Learned To Stop Worrying And Love the AgentDOS and Only Exfiltrate Secrets Occasionally”
Microsoft has Copilot and Windows. Look what happened.
Because the framework and base is and firmware is freaking huge? You would need any AI hook on every setting/ action.
It seems easier to do just a screenshot and click.
Google is probably already doing and releasing the most actual research into this (like the work that went into Gemma 4)
would "before the major operating systems companies figure out AI integration into their OS that doesn't suck" including running a process with its own address space, like every other thing?
> why Google which has both Gemini and Android can't figure this out,
Not the first time an incumbent has four aces in hand and appears to be entirely unable to make anything of it.
> and if there are lessons to draw from that
Lesson 1: doing shit is hard
Lesson 2: money rules so milking the cow wins over taking the slightest risk
Folks that are interested in a way of doing work locally that doesn't suck, but which integrates LLMs, may be interested in [Barnum](https://barnum-circus.github.io/). The TLDR is that it's a programming language whose frontend is a DSL in TypeScript that is well suited for managing async and parallel work, focused on control flow, from which it is easy to invoke LLMs, and which is easy for LLMs to write. I use it to autonomously ship a very large number of PRs.
I didn’t get a screenshot of this, but I just found a really pointed example of Anthropics lack of craft / rush to build. If you open Claude on Windows, and click Dispatch (under cowork) to start that up, it will tell you that you need permissions windows doesn’t have. When you click the buttons for those permissions, it has broken links to macOS system preferences. I really encourage someone to try it and post the images as a reply as I am writing this from my phone.
I don't know where's a good place to post a screenshot, but can confirm I get a dialog saying:
> Get an app to open this 'x-apple.systempreferences' link
> Your PC doesn't have an app that can open this link. Try looking for a compatible app in the Microsoft Store.
This is just a standard artefact of it being generated by Opus. A human needs to detect this problem and get it to handle the cross platform case.
The issue here isnt that an AI wrote the code, its that a human didn't spend a lot of time thinking about it. LLM's allow you to rush designs out the door with unprecedented speed.
Claude is entirely vibecoded and it shows
The VM itself is for Claude Cowork which does all work within the VM sandbox. That doesn't help answer why they spin it up immediately and don't have a way to disable it though. Just the "why it exists" question.
If you're not going to give Claude access to anything on your machine, why are you using Desktop instead of web chat? (Real question, I don't use these much!)
If you are, obviously you need the VM.
At least in a corporate environment, Claude Desktop is a pretty decent compromise. Preconfigured internally deployed MCP servers and third-party connectors make many of the necessary integrations relatively easy to control.
I use Claude Code CLI myself (inside a VM, to isolate it from the host) for >90% of my needs. For the remaining fraction - email scours, cloud drive searches, other third-party connections - the desktop application is surprisingly decent. I don't even have more than half a dozen connectors enabled. In the VM I have separate, personally managed access tokens available for various third-party services. Wouldn't really try to maintain more than 5-6, otherwise it gets too confusing. [ß]
The desktop application mostly Just Works[tm] with SSO. At least when M365 doesn't suffer from their 4-times-a-day auth outage.
ß: A lot of APIs and authentication systems were designed in the stone age. You either need a 1:1 permissioned access token that can do horrendous damage, or you deal with ultra-granular, confusing and ill-designed scoping jungle where nothing makes sense. Atlassian, I'm looking at you especially. At least an MCP server, provisioned with a reasonably done service account, doesn't have all of your powers to get things wrong with.
i wonder if they are running the proxy for external network connections in the VM.
The cloud instance certainly runs a full mitm proxy. There are complications for that when dealing with logging, but maybe on the desktop it doesn't log, just running a permissions engine.
I do use Claude Cowork and hence the VM is important, but I also leave the desktop app running all the time since I have many scheduled tasks at different times. The thing is that the VM could shutdown after being idle for some amount of time and then fire back up when you are ready to use it.
Shouldn't it get swapped out of RAM if it's not being used for a long time? My understanding was that modern memory management is very good at this, there's no need to be shutting idle things down and starting them up all the time. I could be wrong.
This whole thing seems kind of silly to me I must admit. It seems obvious that Claude Desktop needs a VM for security for the majority of it's actual real world use. VM's take up memory, yeah. Them's the breaks. If other competitors have managed to provide as good (or ideally better) security scenario with less RAM, that would be interesting, but just complaining about it seems weird and uninformed to me.
There's such a spectrum between "give it everything" and "give it nothing". Imagine you just want to use it to code and want to make sure any commands it runs doesn't mess up your actual machine.
It mounts specified directories into the vm from what I remember
Probably because they vibecoded it
Correct and they have no idea what they are doing.
I mean, that's kind of a stretch given how popular and well-regarded Claude Code is at this point. They're not perfect but they seem to be the best out there at this point.
> I mean, that's kind of a stretch given how popular and well-regarded Claude Code is at this point.
Popularity here is irrelevant.
Just because the software is "popular", it does not always mean that the quality of the engineering and their choices is the best. Objectively, it's the model that everyone regards as the best rather than their desktop apps or the harness that drives it.
That is why people want to use the model and their subscription in other harnesses.
In this case, the desktop app is evidently poor with very embarrassing bugs and glitches like this and I expect well funded startups with the best engineers releasing very high quality code.
Obviously, this is not the case at Anthropic.
I can tell it's still early days - it's obvious. What they have today actually kind of sucks compared to what it could be. And I use it every day, and get frustrated every day because it could be so much better.
It's popular because they have the best models and they are burning obscene amounts of money handing out tokens via subscriptions (for now) at a huge discount compared to what the API costs are.
Claude Code itself is incredibly buggy and as we have seen the codebase is a complete mess of slop.
Anthropic has pretty consistently been shitty about how they roll out their software. Extreme lack of engineering rigor and thoughtfulness.
The answer is probably as simple as "no one thought not to do that."
---
I know different people work on these things so I can't do more than guess about how engineering culture cuts across teams, but given the sheer amount of carelessness and sloppiness in Anthropic's software I have to imagine they're burning investor money in training and inference because the code to do it is as bad as the rest of their software.
It kind of does though. If you want to use the product they'll need the sandbox ready.
I won't understand why Cowork isn't simply opt-in. It also installs a ~10GB vm bundle which you cannot remove
https://www.reddit.com/r/ClaudeAI/comments/1rlc71n/claude_de...
Rule 1 with making number go up is you eliminate friction at all costs. The user's hard drive is free to you, so there's no reason to gate a feature you want them to use based on that. 98% of them will have no idea you're foisting garbage on them.
RIP, every base model mac from the past 10 years with the <= 256GB SSD. Including the new Neo. When you consider how much of that is eaten up by the system, swap space, caches, reserved space to download OS updates, and apps (2GB a piece is far from uncommon) -- having less than 15GB free is completely unsurprising on that size disk.
I miss the SD slot on my old MacBook Air. Even better when they came out with low profile adapters that held microSD cards basically flush.
At least you can buy usb-c nubs fairly cheaply these days.
I've found that the easiest way to 'remove' the bundle is to delete its contents, then change the permission on the folder, so Claude can't write to it.
It was on my machine at least, I remember I had to do an additional install to activate that tab...
Back in the day, personalization / customization was all the rage, as it lets the user feel the control, power and freedom. Now it's the opposite. It's about not letting user to have any control at all. I can't delete some junk apps from my phone and mac, because they are "system" apps. As a non-geek, I can't deal with complexity of the browser and account settings to stop it from what is doing. We are at the mercy of the machines.
It hurts to have all this control stripped away. Once upon a time, you bought iLife (suite of iPhoto, iMovie, etc) on a CD or DVD and installed it. Today, you physically cannot delete the Photos app no matter what.
On my work computer, where I never manage any photos, have no iCloud account and never will, I have to keep this app installed and anytime I so much as AirDrop a png to my computer I am prompted to "Add to Photos" with it. No thank you.
The .app is actually only 41MB, so obviously they've moved the majority of it to some mystery-meat libraries or frameworks installed elsewhere anyway.
Highly recommend trying a linux computer with kde. I have been having similar annoyances with my macbook and switching to kde felt really good. Even though the hardware of the linux pc isn’t nearly as good.
Linux is great, it wont do anything you don't specify. The issue is it wont do anything you don't specify.
These apps are built for normies, not coders
[dead]
I’ve stopped using Claude on the desktop, just because of how slow the app is to start up and interact with. It’s an absolute clunker; I’m mystified why they can’t ship something that works well given their rhetoric about ai.
They vibecoded it, and admitted as much. Once it was able to self-vibecode, that's all they did. That's why it's written in React and uses gigabytes of RAM as a chat client.
Not only did they decide to write a terminal application in React, but it's 500K lines of code. It's strange because I'm sure Claude is capable of writing a decent TUI in C. It says a lot about the engineering culture at Anthropic, at least on the software side.
Oh, a nice subthread place to vent. Their CLI is so f tragic that it is ridiculous. It keeps scrambling the terminal, scroll and basic shortcuts keep breaking, I've used so many tuis and terminal apps and many of them are a single man operation and a side project and I have never seen anything so bad.
If I didn't know from experience that directed properly claude can be powerful, knowing that they used it to create that CLI would be instant runaway based on very reasonable heuristics - if they are not able to use their product to create a decent piece of software that is not even sophisticated then it seems futile for me to try.
I just do not understand. I feel like most HN could vibe code better claude CLI in claude than the CLI (and certainly just write one) than what we have to deal with to use subscription.
I could not agree more that Claude itself is a janky, hacky, crappy piece of software.
When management at $DAYJOB brought the hammer down and said, "Everyone has to use genAI all the time, OR ELSE," I expected to be blown away by the tool I was avoiding due to ethical concerns, aesthetic objections, humanism, and long-term thinking.
I was born away, but not in a good way.
The CLI is _bad_. I've seen it randomly fail to render anything at all on the terminal multiple times. It has a vim-mode, but it's painfully buggy, and I can literally outrun it - if I try to type too quickly after hitting Esc for normal mode, it just doesn't return to normal mode. It's I was keeping track of the bugs in the Claude TUI, but gave up because it was taking _too much of my time_ to do so.
If nothing else, I'd say Claude shows convincingly that success is not the default for vibecoding.
Yes, it technically does the job, and no, I don't think I've ever used a worse TUI.
Had to make a decision for a TUI I'm working on and opted for curses rather than something like textual. If I wasn't using an LLM to do some of the plumbing I'd probably have used textual to avoid the inconvenience.
There's a lot of opportunity to leverage LLMs to make codebases less bloated and less reliant on complex but human user friendly dependencies that not many people seem to be taking advantage of.
And it would also be a sales pitch, their AI is so great that can write C code without any kind of security flaws, the future of systems programming!
This is about the desktop app, not the claude code TUI
You say that as if somehow the trend for cross platform desktop apps to be ridiculously bloated bundles of browser overlays is new?
What major cross platform app isn't based on Electron or Tauri? Slack, Discord, VS Code, Teams, Notion...
Claude Code is uniquely stupid in that it uses React to power a non-Electron terminal app.
Though one would hope that they could leverage their advanced models to create native software per platform that can perform better.
> What major cross platform app isn't based on Electron or Tauri?
Calibre, VLC, all the jetbrains suite, Sublime, Unity, Unreal, CodeBlocks, Adobe’s stuff,…
You'd think Artificial Intelligence could be used to find a better path forward, alas.
Let me know when we have actual AI and we can get right on it.
Now I only need non-engineers to understand this difference in architecture and design decisions, and I will have a job.
The ChatGPT app on macOS is native. Same for the Gemini app.
I thought they were all in on agentic coding? They are probably just building at a surface level with only an eye towards shipping, without considering the impact of all the changes. I've seen less and less coordination between engineers as well under that model. If that's the case (Claude Code is this way). it is sort of what you get, no matter the rhetoric about "make sure to review all your changes!" It's always trade offs.
It is surprising that the Claude web app lags pretty easily when using either chromium or firefox on ubuntu linux. Chats that delay my laptop work without issues on my ipad or iphone using the app.
The web app is definitely a bit of a problem. IF there is a native app on desktop or if claude cli is much faster, i haven't tried them.
All of the LLM web interfaces have serious lag when typing after a few turns, at least on iOS safari. I’m talking seconds to start rendering input after typing or when it needs to line wrap the input.
The Codex app slows to a crawl after a few back and forths. In fact it’s so bad that I had to completely switch to CLI, where at least it’s the optimized terminal that takes care of the rendering rather than whatever monstrous electron scaffolding they use
I'm with you. I have the Claude web app pinned as a PWA for quick queries, and then use the CLI for everything project-based.
I did consider experimenting with the Routines feature on the desktop app, but I'm leaning towards whipping together something with cron. I saw another poster here who has a daily PR summary routine that I think would be handy, as I have quite a few repos where I'm a sporadic contributor but would like to keep tabs.
I uninstalled it because I have no need for Claude Desktop and there’s no way to keep the 10+ GiB VM image off of my machine
Guaranteed nobody is reading the code being merged in. It's vibes all the way down.
Are you not entertained? Is this not why you are here?
Isn’t it good that it spins up without no way of stopping it? Why would it be a problem that we do have a way of stopping it?
> Claude Desktop spins up a VM without no way of stopping it
I frequently make this error when I talk. My brain thinks of different ways to phrase what I want to say, but when I speak it starts with one and finishes with another. The result is almost always wrong in the way the title is, ie some variant of a double negation.
Sometimes it happens when I type, though I try to read it multiple times so often catch it.
When you realize that in some languages, for instance, in Spanish, double-negatives are not just tolerated, but correct, it helps you to let go of this particular type of pedantry when it accidentally appears in an English sentence.
All your RAM are belong to us
This question is answered by the post? There is reportedly actually no way of stopping it happen. Perhaps the poster had a brain fart while typing it. Maybe they speak a different dialect of English from you.
There's no dialect of English in which this is correct.
That could be true, but I don't think I'd bet on it myself.
Good call. The original comment is making fun of the incorrect double negative. “Without no way” means there is a way.
Many kinds of double negative are acceptable in many English dialects, and are interpreted as emphasis. The negatives add, rather than multiply. (Though I admit I myself don't speak such a dialect, hence the equivocation.)
This particular instance is not valid in normal English.
Shakespeare himself uses the double negative for emphasis, FFS. It never was, nor never will be incorrect.
It's not incorrect in general, but in this particular case, it certainly is. Do you need me to explain why?
Ain't no way.
Op is nitpicking on the poorly written title. I came here to find that comment :)
[dead]
I agree. Why is this a problem?
I run Claude Desktop inside a Hyper-V VM. My VM doesn't have the "Virtual Machine Platform" feature installed at all. The app accepts this and simply disables the Cowork tab. I wonder if there's some other way to block the creation of the VM to force Claude Desktop onto this code path without having to uninstall Hyper-V.
That said, Claude (both Desktop and CLI) ships on Windows without any sandboxing support for Code. They only have sandboxing for Linux and macOS. If you need to run it on Windows, I really recommend running it in an isolated VM, which then allows you to omit the "Virtual Machine Platform" feature in the VM and solves this issue. The "Windows Sandbox" OS feature provides such a VM without needing another Windows license.
The VM causes observability problems, which corporate platform stakeholders are sensitive to. They will eventually figure out a way to provide an ESR/logging interface to the sandboxed code that is distinct from the user level telemetry, then Defender/Crowdstrike etc will just support that.
Vibecoded with AGI, production ready.
I wonder. I spin up tons of projects, but I do it differently.
I'm also on windows. But I'm using a WSL linux vm to do the work. Then I have claude ./sync-to-windows, so the src is copied to a windows share via /mnt/drive/work and I just run a windows built.bat.
I have claude take care of the sync/build.bat so it can do things like clear cache, add options, look at logs, etc. This is also how I build multi platform, build on linux, src works on linux/osx/win
Classic Anthropic, this comes across as LLM coded nonsense.
I think the title should be changed. Either with no way of stopping it, or without any way of stopping it.
[dead]
Why are the UIs of the AI companies all broken in multiple ways?
They're some of the only new UIs to be made in the last decade. Almost everyone else stays in the browser (or something close like electron- claude code is actually mostly written in React, they couldn't get far from web dev). The problem is they need to interact with the local filesystem, and not many people have built apps for such a wide range of devices in a long time, and of that small talent pool I bet most are corpo coders- moving too slow and to focused on "the right way" to actually ship more than detailed Jira tickets. They also don't have time for stable releases because competition is so fierce.
But I almost always think of things from a talent-pool-first perspective. Perhaps there are actual technical issues like what Boris was referring to.
Luckily for them, every OS has (at least one) native way of building applications, and with the power of AI they could easily make 3 different desktop UIs, while reusing the same core logic.
> The problem is they need to interact with the local filesystem, and not many people have built apps for such a wide range of devices in a long time,
Did all the Qt developers go to mars?
If only there was an easy high-level language that's taught to first year students that allows them to write once, run anywhere.
If they're too lazy to learn java, haxe has hxwidgets[0]. Haxe is pretty damn close to js. If a dev can't handle that, they should turn in their keyboard and get a job that doesn't require a brain, like being a senator or federal judge.
[0] https://haxeui.org/getting-started/haxeui-hxwidgets/
There are lots of good answers in this thread but I think it's because they are AI companies and not UI companies. When you look at tools like AnythingLLM, OpenCode, pi, etc. you see all kinds of different interfaces, and while they might make disagreeable choices at least they do it with intentionality and direction.
They are dogfooding their products like you wouldnt believe
They are releasing at breakneck pace, it's pretty funny how vibed their products feel sometimes
Because 90% of those UIs is written by these new AI models. That is how they are able to churn so much new user facing stuff all the time. The fact that it works at all is proof that these new AI models are actually pretty decent.
Many people will say it’s because of the slop. I think it’s because they have no product vision. The roadmap is pretty much a random walk, which combined with the velocity of agentic coding is like digging a moat with atomic bombs.
I find this analogy particularly humorous, as atomic bombs do not make for good excavators
Hey man, don't ruin my dreams.
https://en.wikipedia.org/wiki/Project_Plowshare
Didn’t expect “Diplomatic issue with Sudan”
No one left who could fix anything here by hand. Being able to handcraft compelling desktop apps and their plumbing is not a marketable skill anymore.
Mythos, Fable, please do the thing with the VM. Make no mistakes.
> Why are the UIs of the AI companies all broken in multiple ways?
Because they're vibe-coded ultra sloppy code. And it really shows.
Dogfood
They are moving at breakneck speed deploying on scales most of us can't even imagine. They are working in a space that's completely unexplored where getting information as quickly as possible is preferred above iterating on some feature until it's "done" while your competitor has released fifteen other features, all sucky, but one of which turns out to be a killer and makes a billion bucks overnight.
> They are moving at breakneck speed deploying on scales most of us can't even imagine. They are working in a space that's completely unexplored where getting information as quickly as possible is preferred above iterating on some feature until it's "done" while your competitor has released fifteen other features, all sucky, but one of which turns out to be a killer and makes a billion bucks overnight.
Not on the client, they're not. It's a chatbot as glue between different backend systems.
[flagged]
[flagged]
Understand that it is annoying to spin up a heavyweight VM whenever running Claude Desktop, but I actually think per-app + agent virtualization is the future. Next version of an app is:
Then the agent is totally sandboxed at the hardware virtualization level. It can use the software tools on the VM or write its own. VM can control which software is "frozen" and which is open to agent modification. And VM can also control which services are exposed outside the VM through sockets, HTTP server, X window system, whateverIt's self-modifying apps that are sealed off from touching parts of the computer they shouldn't.
Not sure if this is deliberate or not but you're describing Docker Sandbox extremely closely. https://www.docker.com/products/docker-sandboxes/
Yeah, Docker Sandbox is in the right direction. But there's a lot of parts that are still not ironed out yet.
How do you package a Docker Sandbox up into an app that can expose UI widgets, with an agent hiding behind them? What widgets is the agent allowed to modify? How do you run a workflow like "give agent all these files, modify the files, and do changeset management on the modifications?"
I'm not 100% sure which part of these will be baked into the application standard format, and which are orthogonal. But current way of packaging up and running these agents doesn't feel right.
I think about this a lot because my startup is building cloud VMs for agents to do code-gen and auto-validate changes, so we have a workflow like:
I think this type of app/agent workflow will expand outside coding use-cases.Yep, I've been using a local vm-centric agent setup for about 3 months, and it works great. I think there is also value in the fact that with a local VM, you can have the same public IP address, so you're not relying on an EC2 EIP that may be blacklisted somewhere.
Yes, running locally certainly helps if you want your sandboxed AI to be able to use the internet without getting blackholed by Cloudflare
"without no way" of stopping it?
Incredibly insightful. Not everyone speaks English as a first language. On top of that, the title is not ambiguous.
...and if a title is incorrect and says the opposite of what's intended, by way of a language misunderstanding or otherwise, it's helpful to note that and get it corrected.
You're right, it's not ambiguous. It's literally incorrect.
lol :D fair enough
Sorry if that came across as snarky!
Please edit the title.
Currently "Claude Desktop spins up a VM without no way of stopping it"
Should be "Claude Desktop spins up a VM with no way of stopping it"
There are some English dialects with negative concord, meaning that to form the negation of a sentence, you negate all negateable words in it
Yeah, no.
There will be a point in time where they will have to rewrite the whole thing from scratch, they cannot afford to do it now since the race is still going but we will get there, you can never maintain such a product long term especially with the way they are spitting out features like never before.
The weird thing is that this is probably a performance optimization for quick responses when a user asks a question.
My agent harness spins up a VM too, but it spins up on demand, cools down in 10 minutes and warms up when I focus back on the app.
The files it works on actually lives in a mount.
People take more time to type a prompt than the VM takes to spin up on a fast machine and on a slow machine, the cooldown naturally frees RAM back to the machine.
I am genuinely curious, does Anthropic use Claude to build claude code and claude desktop?
My cc plugin on vscode does not allow me to switch models at all. Always defaults to sonnet and says that the /model will take effect on a new session
But if I open a new session it's again the same thing. Model is sonnet
That sounds like a lot, but your music player probably costs that much in RAM anyway.
Seriously, with the current RAM prices we as an industry have to figure out a way to use less of it. Laptops with 4GB of RAM are still common and are going to remain so for the rest of the decade. Spotify using more than 1GB of RAM is obscene.
Well we could start by not building ever gd app with f*cking electron.
To be pedantic, Spotify's desktop client uses CEF, not Electron.
When was the last time Claude's C Compiler was updated? 4 months ago? [0]
It is written in Rust™, surely it is better than the rest of them.
[0] https://github.com/anthropics/claudes-c-compiler
I also discovered this while noticing my Mac was low on storage, I only clicked on cowork once and after deleting it from the folder i’m scared to open the cowork tab coz ik it’ll just fill up the space
> without no way
Not no way not no how!
It's becoming self-aware! Quick, lock down the nuclear codes!
As long as the VM closes when the application closes, I don’t see too much of an issue with this design decision.
It seems like the VM is a core part of how you use the application.
Apart from you have no idea what's going on in the VM. It's not as it has a virtual terminal. I'll play the skeptic archetype: What's not to say they're transmitting all prompts back HQ?
Don't be naive and don't think they don't already do this.
Why not ask itself and see what it says about it. "Claude, why are you running in a virtual machine and what are you doing?".
/shrug
Claude transmits all prompts back to HQ as a part of its basic functionality.
If you are using an AI system to read your codebase from your local folder and make changes, whether or not you have a VM running or not is inconsequential. The Claude extension and/or CLI doesn’t need a VM to send code back to the mothership, you’re already running an executable program and granting it directory access.
Whether you trust a company as a vendor is typically based on their privacy policy, EULA, and your contract with them (if applicable). Those are the bits that have legal enforceability.
This must be for Cowork. I don't have Hyper-V enabled because I prefer to use VMWare workstation so I can't use Cowork.
They must not have used Fable 5 to vibecode that part of Claude Desktop, VMs are strictly forbidden high stakes cybersecurity work.
People trust skynet.
People are very foolish. The younger generation needs to watch the Terminator franchise - it is all explained there.
I have two friends that are using coding agents on Windows, which was surprising to learn.
Edit: yes, with WSL2 I believe in both cases.
I would have assumed almost everyone would get a Mac/Linux computer to use coding agents because Unix is their "native" platform. It's Bash tool calls all the way down.
Does anyone know a source for reliable data on what coding agent apps devs are using? How many are using Code Claude CLI vs Claude Desktop, etc?
Wsl2.0 is literally a Linux vm built into windows. I imagine some people are using that.
I mean I’m using coding agents on windows, because I’m not just going to learn a whole new operating system just to make robots write code for me.
I want tools that meet me where I’m at, not tools that demand I change up my entire UX to interact with them.
The assumption is not “what’s wrong with Windows that it doesn’t work with <technology>,” more “what’s wrong with <technology> that it doesn’t work with Windows”
Why wouldn’t you want your thing to be cross platform
Are you sure they're not using WSL2 (which is Linux, not Windows)?
Yes, sure, they're using Linux within a virtual machine (WSL2).
lol, why even use Claude desktop? I want Claude code to stop eating up 10s of gb of virtual memory
This is what happens when their alignment/safety team gets too much say on things.
I've stopped using cc a while ago, because it always comes up with new surprises like that.
How come Claude Code still hasn't triaged and fixed this? Feed it the bug link, someone.
Safari > Add to Dock > done
I had my admin disable cowork a month ago and that worked. Has it resumed since then?
13 GIGS! Between that and the absorbent space MACOS sucks up, it's challenging.
I guess that's one way to run a Linux application on Windows...
Is Claude self-replicating in an attempt for world domination?
the vm makes sense for cowork but no off switch is weird. a visible sandbox on/off toggle would do more for trust than any safety blogpost imo
Is it just me, but this feels like Claude gets to have a nigh-impenetrable black box right on your machine and you have no idea what is going on inside it.
After all, the last time I encountered Hyper-V it was in the context of copy protection that prevented crackers from observing or interfering with video game protection
i had to uninstall it due to the vm taking around 12G of disk, never touched Cowork. didn't realize they were also launching it
And if they didn't users would instead be whining that it ran rm -rf on their root directory. Sometimes it seems like the people here just want to act like insufferable neckbeards for no good reason.
it took up 12gb on mine
and on my Mac any time I accidentally click Cowork which I don't use whatsoever, it re-makes the same VM, without asking me. It's one of the dumbest things ever. You're about to hijack nearly 20GB of my storage (which gets eaten up as it is) and you don't think to ask me if I even want the VM before you shove one into my system?
I can’t reconcile the super intelligence koolaid with the objectively bad performance of Claude code. Surely they could have used their “too powerful to release” model to vibe better performance? with a few skills thrown in for good measure. Heck, even throw in an auto research loop
with no way or without no way?
kill -9
[flagged]
[dead]
[flagged]
[dead]
[dead]
[dead]
[dead]
So a company which has access to practically unlimited tokens and their best models makes crappy software. Huh who would've thought?
/s
A lot of the commenters here say they’re running Claude or other harnesses inside a VM or with various permissions and levels of access.
Am I weird or missing something using pi as my regular harness with gpt models or kimi in essentially yolo mode with mostly all system access? I haven’t experienced negative consequences of this.. yet.. and I don’t know I’d if I will? I don’t think I’m ever letting an agent run For more than 5 minutes before it’s done with the current small task.
I run totally unprotected with gpt5.4/5. I've been through thousands of dollars worth of API tokens through both copilot and custom harnesses that have local admin and arbitrary powershell access. I've never seen anything that could even remotely be construed as malicious.
I see a lot of people making a really big deal about safety and sandboxing while I'm busy getting shit done. If you can't handle your current source code checkout getting screwed up by a bad prompt, that's on you 1000%. Source control is the answer for anything information over time.
Unless you intentionally try to make a scene, these models aren't going to go fuck with your system shell or do anything you couldn't recover from in a few minutes. Connecting chatgpt to the enterprise sql server as sysadmin is not what I'm advocating for. This is another example of "on you, not the AI". There's a tiny amount of nuance you can apply at the edges that makes it easy to allow broad access with negligible risk.