I don't quite understand this sort of thing happening. Wasn't the whole point of the internet to be a self-healing network where we route around severed cables, etc?
Or is it that these ATC networks are their own air-gapped network with less redundancy? That just doesn't add up. Or maybe there was only one line going to the ATC, with no multiple "ISPs" like a datacenter would have?
* Yes, TRACON have their own dedicated links. You can look into ASTERIX and STARS to learn some of the cursed ways the data processing and dataflow work.
* There is supposed to be a primary and a secondary link, in this case the primary failed and the fail-over also failed. It's unclear from the reporting if they were damaged in the same incident or if the failover was not tested or monitored adequately.
The Philadelphia TRACON site has been notoriously unreliable and was supposedly improved in 2025, it's also unclear if these issue actually could stem from that implementation.
Reporting is saying that the backup was cut a while ago and they only discovered it when the primary failed. Apparently nobody was ping testing the backup link.
I’ve had enough double-redundant links fail - and I mean pretty thoroughly double-redundant, different provider, different media, different physical path - that I’m quite surprised something like the air traffic control system only has two links.
Contractually dual-path, or actually dual-path? My understanding is that there's enough infrastructure horse-trading going on behind the scenes that it's very difficulty to be certain that two circuits between points A and B don't share the same infrastructure somewhere in between.
First big oops of this form that I remember:
"In December 1986, the ARPANET had 7 dedicated trunk lines between NY and Boston, except that they all went through the same conduit -- which was accidentally cut by a backhoe. "
One thing that needs to be reviewed is if they are dual links or redundant links.
In any system where both sides of redundancy always carry traffic then loss of one link can cause congestion failure if any link fails. This is a very common means of failure in electrical networks that requires load shedding. Well, you can't load shed air traffic.
A more complex system that I like, but comes with it's own set of constraints and implementation issues is a system where both lines carry all the traffic at all times. This way the default state of the system is always working and your first failure isn't invisibly critical.
But this is very hard as we see in TCP when things get out of order and high latency creeps in. You have to manage a lot more state at the data level.
Wow, at first, I thought "this is wonderfully silly" but within about 45 seconds of playing with it, I instantly wanted the window seat and thought, wow... I could see what it's like to fly over so many cities from above, just like with a window seat.
I'm so excited about this sort of stuff. ForgeCAD [1] for me was the first mind-blowing application of this a few months ago, where I'm able to just have claude (or codex) write JS that creates 3D Models that I can then print in my 3D printer.
- Built a ring for my compost bin that can properly hold down the plastic bag. This was originally a multi week project because I was also learning Fusion 360 in small 2 hour chunks over many weekends, then with ForgeCAD I had a good prototype in 60 mins with multiple iterations.
- I also built a cover for my overheating Logitech Circle doorbell, snapped a photo, gave it to claude, and claude found all the dimensions on Logitech's website, and then we iterated with 3-4 prints until we had it just right.
This is a whole new set of workflows that are unleashed with agents, even for hobbyists.
Yeah I was skeptical about AI until I learned (through hands on testing) that an AI can fail 9 times and get it right on the 10th time - in like 1/10th of the time it takes a human to get it right. It's all in the iteration.
Are those not the exact same workflows, just replacing "I spent time learning" with claude?
I find that if you don't skip the learning step, you can then use the tool more efficiently(e.x. you learn to measure twice and verify so you can iterate in 1-2 prints instead of 3-4)
I am not averse to learning (I spent hours learning Sketchup many years ago when redesigning my garage).
But the time freed up not learning Fusion360 freed up time to learn about different kinds of fasteners, clasps, how to arrange magnets, etc, so I moved the learning to a different place.
Plus avoiding the "wife is angry cuz this project I over-engineered has gone on for weeks unsolved" factor is always a nice bonus.
I feel pretty good about the 27 series in a way I haven't in the past (I've been running iOS 27 and it's working well), except for one thing, they remove x86 support (no more Rosetta), so just keep that in mind in case that might disrupt anything for you.
Yes, 27 is the final Rosetta version. The warnings are worded in a way that makes sense if you already know that but are easy to misunderstand if you don’t. I think they initially didn’t want to commit to the next version being called macOS 28. The final 27 release has started explicitly saying it’s going away in 28 which makes it a lot more straightforward.
I actually think this is what they ended up doing - just regular conversations with tool calling into your various apps, behind a Siri AI API. not HTTP based mcps (for now) but rather local ones.
We did the same thing - had 90% of it overnight. Then spent a few days in the background tweaking for polish.
Our app is smaller, and has about 15-20 screens. I started at about 12:30am by giving codex a goal and it inventoried every screen based on the react native code, then created android and iOS directories, used maestro (I had already set up this tooling for a previous personal app build a few weeks prior), and had the whole thing working in android and iOS in the morning. Took it about 6 hours while I slept.
The app is way smaller, launches instantly, and the android app is (supposedly) native looking. I say supposedly because I don't use android phones. But it's using Jetpack Compose and Kotlin.
And I don't know Swift or Kotlin. I honestly don't see the point of React Native anymore. I know Expo is doing very cool agentic stuff, but I'm just not sure why I'd need any of it when I can write a native app.
We have people on the team that are android users. I just meant that I don't want to evaluate whether it feels "native" as I personally am not an Android user.
You just install it on your phone and use the app.
Maintainability concerns are entirely overblown by people who don't use agentic AI to develop large mobile apps, but anyway give their opinion as if they had that experience.
I put in a few hundred hours, and I reached the same conclusion as Shopify. With reviews from other models and then a manual QA pass the result is fully usable.
I work as a professional app developer. And I find this take to be naive.
Most of the time when I review code from AI, there is always something to improve.
It’s either a maintenance issue. e.g., Opus recommended and implemented a fix for a database corruption crash. This was ~400 lines of code with many moving parts. I reviewed, and found out Android Room library already handles this recovery case, and all I needed was a 10 liner PR that catches this exception and ignores it.
The maintenance is not only the burden on the human and LLM. With too many moving parts, it becomes harder and harder to build and verify the correctness of future features. Yes you can write test for this and that, but it didn’t need to exist in the first place.
The second problem is correctness issues. Especially the edge cases. You cannot just manually test out a race condition on a phone! Sometimes it happens! Sometimes it doesn’t! If it leads to a visible signal like a crash, then yes, you can try to reproduce it. But there are a lot of these that are “silent” and would just lead to bad experiences.
We already had a software quality crisis! And I think such views only exacerbate the situation! Quality matters!
And this is not an anti-AI stance. I vibe code personal projects where I don’t even look at the code. But when I use AI as a professional engineer, I act like a professional. Because these products do have an impact on people’s lives.
All true (and thanks for posting a concrete example rather than "LLMS suck"). But my take is that none of this is much different than before times when I had teams of developers creating applications. They would often make similar mistakes which I would either need to catch or which would flush out in the field. Where it seems that LLMs are not excellent is where the person driving it is also the senior domain expert so can immediately spot pitfalls. But typically using humans to develop software this was really not often the case. Those people get promoted so they're no longer cutting the code. Under that scenario (replacing subordinate humans) I find the current models are either on-par or somewhat better (specifically because the models can also act like a peer senior dev, discussing approach options etc).
So you don't use agentic AI to develop a large mobile app and you think my take is naive?
I also used to work full time as a Android developer for five years, and I'm pretty sure I know better than you about the quality of my app that I work on everyday.
> It’s either a maintenance issue. e.g., Opus recommended and implemented a fix for a database corruption crash. This was ~400 lines of code with many moving parts. I reviewed, and found out Android Room library already handles this recovery case, and all I needed was a 10 liner PR that catches this exception and ignores it.
I understand this, but I just can't bring myself to care. I've been doing professional software work for almost two decades. These sorts of improvements/time savers are great without AI. With AI? Whatever. It's fine.
When the underlying lib has an issue, it'll be quicker to debug with the whole thing in context.
You do realize that letting the LLM produce more output means that maintenance will be more expensive? I can easily see a world where claude and gpt are producing more tokens to sell you more tokens.
Just the other day I burned through my 5h quota twice in a row because I had opus spawn a review session on a medium sized PR and I don't know what happened but I told it to summarize to me and it said it spent 100M tokens throughout 50 subagent sessions.
And more recently its been recommending that i install this new browser called Aside. I did, and it almost felt like I was installing malware so I Uninstalled it fairly quickly (it also was not a great browser)
Well yeah, that's the name of the game of most software companies. Anthropic has been fairly good up until the last week or so when it started needing more hand holding to not do things it didn't normally do.
It depends on how hard to maintain the code added is, how likely it needs to change in the future, and most importantly on the cost.
If reviewing and manually improving the code takes hours, the cost may already be in the thousands.
That buys you a lot of AI usage, roughly a few months of continuous work.
You have to balance this with the chance that the suboptimal code the AI generated is actually fine and maintainable enough, and also the chance that during further work on that code a model might implement the same optimization on its own.
Same result here just using plain Opus 4.8+. I had a web ap with a PWA approach. Now I have an iOS app written in Swift/SwiftUI and an Android app in Kotlin in the appstores. I do not know how to code a single line of Swift or Kotlin. You just test the app and iterate with the AI over it until it is stable and does what it should.
Why would you asume that, and how do you define serious? Ever since the iOS appstore presence I get more signups from organic app store searches + installs, which results to some percentage in new daily+monthly active users. Meaning: People use the native app and keep using it. That is how I would define have value, as it provides value to the users - else they wouldn't be using it.
My main agentic coding side project is something that I can't justify paying the apple developer license for. If I was an Android person or I didn't have to pay the developer license, maybe I'd just go for it.
Android got wirse than Apple. It costs 25 dollar now to get ID verified (mandatory) and before you can publish in the Ply store you need 12 betatester that install the app from a special link - those testers must keep the app installed 14days. Only then will you become visible publicly in the play store.
So you end up paying people on Fiverr to do it which costs more than 99€
I have been getting these comments often here, including concerns about my non existent backend's security.
Last time, when I pointed out that the attack surface for mobile apps is typically very small, some users started to talk about zero day vulnerabilities in the OS's media handling, as if it was a concern for my app implementation.
You sound like a person who's never had their app pen tested. The attack surface is anything but small if you're working with any kind of sensitive data.
Are there cybersecurity concerns in the frontend? I would have thought you have to assume the client is untrusted and only do security work on the backend
My favourite is a logout button with a logout API that fails. (Not a huge pratical concern, I admit, because it's a local attack.) Nobody ever notices because it still shows the logout screen, which hides the API error toast (if errors were even displayed). The still valid refresh token stays in sessionStorage (or even localStorage) while the app displays "logged out". (Bonus points if you cleared the access token in the error handler but not the refresh token, and on page reload you ask the user to log in again despite having a valid token.)
Or a login form that gets hidden after login, but clears the username and password only when you click "login back in". (Bonus points if the backend also enforces a 5min session timeout "for security".)
Storing private secrets in your public client is easy to avoid for anyone halfway competent. We are all professionals here.
Turn on the secrets scan in GitLab, and put in your release checklist to have the AI audit the usage of secrets in your app, and this is basically guaranteed not to occur.
I doubt current models even make such a mistake in the first place, and particularly so if you use reviews at all.
WebViews are not an inherent problem, it's the system browser embedded in your app.
Where it gets tricky is if your use case involves authentication in the browser. Together with the authentication in your app this is the one area where you need to focus on security.
The case where a SDK update is needed to prevent weaknesses of the OS seems rather unlikely.
Doing anything right on web is 10x harder and more complex. The problem is the browser, once you use it to deliver anything you have to buy into all of it’s bullshit. CORS, XSS, headers, caching. All that just goes away (outside your backend API, if you even need one) when you ship a native app
This is the most naive take on security ever. For the backend, you assume your client is compromised, but you still don't want to allow your client to be compromised.
If your clients are compromised then what's even the point of backend security. Users will login and do legitimate actions while their compromised client does whatever behind their back, while still looking normal. And the backend can't tell the difference.
You don’t believe how often people leave secrets in the app or use webviews and iframes badly, misconfigure OAuth in client side and so on. There are many issues where secure API does not help.
I am using Gemini as well as Opus on a somewhat small project in React Native and I can not imagine this thing being able to build the whole thing on its own without it being a dumbpster fire.
Can you share some details of how you work? What models? What harness?
I use Codex and Claude Code desktop apps. I generally use only the SOTA, now Astra and Fable 5.1, Opus 5 when Fable runs out.
I don't know if Gemini is suitable.
I had few issues with my native iOS app, the results are just decent after a few iterations, the models do what I ask them to do. Where do you see the problem?
The LOC for my app is now at almost 200k + 110k lines of test code.
We use CC with Fable(Opus before that) continuously on a rather large project, everything is tested, we maintain high verified test coverage, we ship features x10 faster than when we started(pre Claude-everything era 2-3 years ago). I never worked with RN before and I ship features now. LLMs allowed us to find issues within RN itself, that thanks to some patches, improved lower end Android experience by a lot. We just use all the Claude defaults with claude.md that evolved over last year.
> Are there cybersecurity concerns in the frontend? I would have thought you have to assume the client is untrusted and only do security work on the backend.
No, they said they don't use android so don't know the native UX. You can test your app on the platform and confirm that the functionality all works, but how well it adheres to the platform's design language is subjective and hard to say if you aren't used to the platform.
I used the emulator - but just like I can use an iOS app for 30 seconds and tell you whether it feels native or not, I can't do the same for Android, since I'm not a daily user of Android phones.
And in the past, I didn't care because when I was manually building the app, I would just do my best with react native. But now that I can actually sweat the details (with the help of agents), I do want to hear from android users and use as many OS-native APIs and features.
Others on our team use Android phones. So when I said that we spent the next few days actually polishing it, that's where others came in, providing feedback when they used it.
I could only sweat the details on liquid glass, etc because I'm a daily iOS user.
That's a recipe for regressions as the amount of surface you have to cover with "just...use the app" gets bigger and bigger.
You can write more automation to test it. But that's also how you end up with ever-growing test run times.
There are much better ways that aren't just "throw out the LLM" either. You just need to be more focused on throughput. Requiring manual validation can pretty rapidly require more hours than just sanity-checking code by hand, even (and I'm not advocating that for every use case, either.)
I can't afford manual QA passes if I'm gonna go as quickly as I want to.
This is wildly overblown. I've been working with agents for a good while, read tens of thousands of generated Python and the language factor is actually the part they get right that humans don't.
Seriously! What a bonkers thing to claim. "I had codex use maestro so I assume it made Android work well and idiomatically".
It's a fine project to do but clearly they put zero value on being familiar with the project's codebase/stack and ecosystem, which makes me feel fear in my heart when I imagine the first "production is down" page coming in. I already hated mobile because it's so much harder to maintain than web (and I don't do any spyware or IAP so no benefits for me there); this yolo approach would give me constant dread.
It shows that at least some software development is moving away from code and to product management instead. I'm not passing judgement on that; I actually think that's great for a lot of software. It is interesting to see the shift happening though and will be fun to see if the general quality of software noticeably changes over the next few years.
> It's a fine project to do but clearly they put zero value on being familiar with the project's codebase/stack and ecosystem,
As much as I dislike it, I think that's the future of _all_ non-critical software (think social media, crms, CI, food delivery etc). Leadership in many companies is explicitly asking employees to have multiple agents running through the day and that will lead to this.
Read this for example: https://www.uber.com/in/en/blog/efficient-software-factory/ . A very useful system, I am sure. But when you have AI at every layer from code to review to triaging, rest assured AI is the only know who knows your system. And you better hope it's not telling you that something is load bearing during an incident.
I read the article and couldn't understand it. I asked Gemini; Pareto things definitely sounds like science.
I read the article again:
- We had bugs. We let agents try to fix the bugs. True positive (fixed bug) is the F1 score. Here are the results for different models. Fixes worked 50% of the time.
- We needed to find a query. Without a knowledge graph it took 20mins, and didn't work. We used a knowledge graph. It was fast (20s), and found the right thing.
I gave my LLM my summary of the article and apparently I "hit the nail on the head".
I answered elsewhere that I don’t personally use Android phones but we have people on our team that do. Which is why I don’t want to personally claim that it feels native.
But yes we are having real Android users test it.
So it’s quite the contrary. I care MORE what real users say. I can only guarantee that the app does things when I tap. So I’m not trusting the agent on UX, only on functionality. But whether it feels native, I am relying on those users in our team.
I wrote every single line of the react native app we ported, and maintained it for 9 years. So suffice to say, I'm familiar with the code.
But I am going to always prioritize the user experience over a developer (like myself)'s need for satisfaction to see code. And a pure native app is _always_ going to behave better than react native.
> And a pure native app is _always_ going to behave better than react native.
Maybe iOS is better about consistency, but I've used enough horribly made Android apps that I would not expect one that's vibe coded to behave better than a professionally made react native version.
You are familiar with the React Native code, you are not familiar with the Swift or Kotlin code, and likely nobody on your team is, since the AI wrote all of it for you.
Presumably a sarcastic post, but this is actually going to be how things are done soon. I had a box that OOMed and needed to be rebooted every few weeks. It was a disaster recovery standby box so figuring out what was going on never rose to the top of my priority list. So I asked Claude to dig into it (proxying the commands it wanted to run through me) and in an hour it had diagnosed the problem, fixed it, and taught me a bunch about memory usage in our system on modern kernels.
Not sarcastic. I have a moderately successful app and I haven’t once looked at the code (though I am capable, so far atleast). The only time I open Xcode is if I have to modify signing certificates. I did set up a slack workflow with a “fix it” button that basically tells Codex to fix the issue, run comprehensive tests, manually UI test for that particular issue with computer use and then deploy it.
It is. For my day job, I’m a software engineering director and I probably won’t have a job in under five years. For low to medium complexity apps, even Codex before Astra was capable of doing it completely by itself from scratch. For testing I would bring up the app after each atomic change and then manually try it out. I also use my own app heavily multiple times a day and I have found dozens of issues but I just ask Codex to fix it on the spot.
React native is an obstacle compared to what clear Swift/Kotlin code may produce. Swift is very powerful and Kotlin, in all honesty, is the first reasonable and very useful thing to come to the JRE ecosystem (save for Scala, which is, well, quite complex still).
Myself turned some python code to Swift, and keep doing so, without trouble or pressure. Of course, I've been doing fair amount of systems programming for 20 years now, so not sure what to advice newcomers. But this approach to dev DOES work for me very well.
Sounds like the advice to newcomers is to not worry about trying to learn a programming language. There already is an llm to program for you, and do it better than you could, so just learn how to talk.
While I mostly agree with you, this is the kind of thing that might not have been done if the AI couldn't do the heavy lifting.
They had previously chosen React Native because it was easier for programmers to keep it updated, since it was really just 1 codebase. With AI, those same programmers can do the more difficult version, keeping the native apps updated separately.
So did it kill a job, or did it make it possible for the existing programmers to do it better?
It's the kind of thing that is really hard to determine in general, but here, it really does sound like they only made the change because it became possible with existing resources. They would not have done it otherwise.
That's an example of AI growing the sector that can lead to more jobs. Because suddenly a lot of tasks that weren't economically viable are now going to be in demand. Custom software for small businesses, platform specific optimized code instead of cross platform software, etc...
In case of my company, the port was planned for this year, with a 9 month timeline and about 10 FTE team. It ended up being done in 3 months, with 2 engineers. (Sure, plus testers and internal bureaucracies, but that 2 FTE for 3 months vs. 10 FTE for 9 forecasted is apples-to-apples). Sure its eating up SWE jobs.
Of course it would. Supply and demand. Some companies would decide not to bother. Others would decide it was worthwhile. The limited pool of supply (app developers) would be distributed across demand.
>> Some companies would decide not to bother. Others would decide it was worthwhile.
The parent's point is that AI lowers the cost/benefit ratio drastically, by reducing the cost. So companies that would have shied away from projects like this pre-AI are now pulling the trigger without much hesitation.
And what a lot of people seem to miss is that with AI, there is going to be (already is?) orders of magnitude more software in this world. I'm sure a lot of jobs will be eliminated, but new jobs will be created as well. Hopefully enough to balance things out, but we'll see.
I agree that AI is suppressing developer wage growth and taking real jobs. The opposite argument is being made elsewhere in the thread, and I think that argument is wrong.
Previous company I worked for would have (and did) hire dedicated swift/java mobile developers to build and maintain ios and android native versions (largely porting functionality from an existing web application)
What's more likely (as others have said) is that the other thousand companies that can't afford to have dedicated staff would've just used React Native. So no jobs were lost, they weren't there in the first place. The places that have dedicated Swift/Java devs can now be more ambitious in what they build.
Really? I’ve worked with plenty of companies that had separate native iOS & Android teams. I don’t know any that use a compatibility layer. Unless by compatibility layer, you mean a web view.
This app has 15/20 screens, an app this small wouldn't employ many people to begin with. Before there was one guy (op) maintaining it in react native and now he maintains it native. I don't see much change tbh.
Just like this has made Web devs into Mobile devs, so could Mobile devs leverage AI to work in Web shops.
I'm not an optimist but there could be an increase of available apps being created which would still keep people employed. Theoretically the cost of creating an iOS app for a company without an engineering team dropped from several hundreds of thousand to a few thousand or hundreds, which could be paid out to a freelancer with AI. More people would go into freelancing for industries which were not contemplated before due to cost.
This is an incredibly boring task. Nothing new, just rewrite everything to just see it all rewritten again in 1 year. Perfect for LLMs and something humans shouldn't do.
While I agree with that statement -- that is also a lot of jobs. We have a lot of humans -- not every single developer sits in the innovation seat. The fewer the jobs available the less employable humans.
I think that rewrite - if AI enabled - owes its thanks to the legion of individuals who put their code up on the internet in the first place.
We all want super exciting jobs. But plenty have boring jobs like this. Between not having a job or having a boring job that pays well, the choice is obvious for many of us.
This is just not true. In the olden times (pre-claude) devs were constantly asking to do full rewrites of legacy code. This kind of project is exactly the kind of work I've trained for and have loved to do for the last 15 years as a mobile dev.
We did the same with our apps used by a few hundred thousand people a day. We are a slow, boring company, so the port took about 4 weeks of engineering work, and maybe 3 months taking into account release process and change management. But in fairness, we had the first working version after a day as well. This was in February, too, with substantially less capable models, unable to do overnight runs.
I thought the same, but then I build a few complicated apps and thought I could keep the different codebases stable but still was a headache, so switched web, ios, android all back to expo/rn. I love that testing is only (almost) only thing now, instead of 3 platforms.
Yes uniffi created swift and kotlin bindings. The app I did with this is just a sudoku game https://www.puzzlesight.com . Before AI I would have reached for something like Flutter to avoid doing all work twice but with AI doing two apps makes more sense.
Okay, makes sense, the whole thing looks and screams vibe coded. On a beefy machine, navigating between your website pages takes 2-3 seconds. So still a long way to go with AI.
Interesting, it's instant on mine. The landing page is just Astro with cloudflare in front of it, the origin server is a hetzner box in Germany though.
So now you have two vibe coded applications you don’t understand. I’m not an advocate of making “job security” decisions but this definitely goes to red flag territory. Was it your job to maintain the RN codebase or do you have other functions there as well? I’m not sure I would keep a native noob on a vibed native codebase. What is your take?
People routinely decompile and recompose existing binaries, and if you use "obfuscation" in your app it's still a small bump.
In today's age, IP is no longer a tangible concept that gives any advantages in software. What differentiates two businesses isn't their IP, it's their relationships and their moat.
Most vendors that had protected proprietary IP are now irrelevant within their vertical. What saves them are the protection provided by patents, which is public.
Definitely agree. Most developers could build a basic Twitter or Facebook clone. The hard part is getting the users, content, and relationships that make the product worth coming back to.
It is both tbh. Anyone can build a poor clone of an application. However there are many things that you don't realise is happening beneath the scene.
It is all the small things e.g.
YouTube Mobile website works really well when you have a inconsistent connection compared to the alternatives such as Odysee, Rumble, Kick and Twitch.
I can listen to a live stream or a video style podcast in the car and YouTube will resume the connection properly as long as the browser tab on my phone hasn't gone to sleep. Kick will just stop, if it is a replay it will resume from the start of the stream after rewinding the stream a few seconds while attempting playback.
While the code to do this isn't that difficult. YouTube has bothered to deal with the edge case of someone like me driving through an area with inconsistent connection while running their mobile site (not even their app).
Whenever people try to use alternatives, they often complain about poor reliability of the app. A lot of the clones are losing users and I doubt they really even know they are doing it.
I've had it work well on both (iOS and Graphene OS) using Brave browser on both. It is simply much better when you are on a unstable connection than any of the other sites.
Isn't ZDR only the case if you "opt out" through a setting? Lots of opportunities to make a mistake here, or for the LLM provider to play games. You could be unaware of the setting. The provider could reset the setting upon subscription lapse/renewal, application update, model release, etc. A developer could accidentally use their personal subscription (w/ setting on) on a work codebase.
Also what's the timing on this setting's effectiveness? What if before you knew to "opt out" you used the agent for a refactor and your entire repo got sucked up in inference? But then you opt out the next day? Is it too late?
That’s the case with a personal subscription directly with OpenAI or Anthropic, but enterprise customers are opt-in, AFAIK.
And there are solutions around it from other providers and model routers. OpenRouter lets you explicitly say on a per-request basis you don’t want to be routed to a provider that trains on your input, for example.
The mistakes you describe are only possible if the company doesn’t really think there’s proprietary IP in the codebase.
If a company really believed there’s valuable IP in code, there would at a minimum, ban laptops or ban code on laptops[0], provide and maintain a centrally managed LLM gateway[1] that handles authentication, billing and model choice, have MDM to prevent you from using your own Claude login, etc.
[0]: For example when I worked at Google, the proprietary google3 codebase cannot exist on laptops because there are no tools to download it to your laptop.
I'd wager many software startups out there would consider their code valuable IP (even if in the AI era it's not), and don't have these types of IT controls in place. Startups I worked at gave you an email account, github access to their private repos, and that was that.
Heck, you don't even need to mix up a personal and business AI subscription. For the vast majority of folks one day their IDE "updated" and a bunch of AI features suddenly became available. Free, no sub needed. A dev starts using them (because, why not?) and lo and behold the company's code got shipped out in an inference call and is now scheduled to be in the next training run. Oops.
You are merely describing companies that aspire to consider their code valuable IP, not companies that actually do so.
IDE updated? It’s the company IT’s job to perform testing before distributing those updates. And also their job to use whatever managed settings to disable those unmanaged AI features.
Yeah man I think we just have experience in two different worlds.
A single digit headcount startup with a "company IT" team testing program updates before distributing them??? Um, no. You just download VS Code (or whatever IDE you want) on the Macbook they give you and it updates when Microsoft pushes a public update.
How's that any different from, say, Windows updating and backing up the company's code to OneDrive?
Actions speak louder than words. If the company doesn't even supply and require work devices, they don't effectively consider their code to be valuable IP.
True, but shops are exposing everything to Claude or Open AI. It's akin to outsourcing all manufacturing to subcontractors in China. That was cheaper, but extremely short-sighted. Now so many products have cheap Chinese knock-offs that are nearly the same as the originals because the subcontractors made both.
I think probably GPT 5.5, or 5.6 Sol - it was ~ 2 months ago.
Each time I get access to a new model, I do two things on all our active codebases:
- Security review of all the code and vulnerabilities (new models will find new stuff)
- Code review of test suite quality, and idiomatic patterns/code for the language of that codebase.
And in the case of swift and kotlin, it's even more important that an agent helps me with the code quality since I don't know what "idiomatic" swift or kotlin looks like, the way I do with JS, Elixir, C#, Ruby, etc.
Next up, designing an even higher level language which will be used by LLMs to compile to high level languages like kotlin or swift. Just store instructions for LLMs in repos.
I often make the argument about whether metric is "better" being split by "is it better on paper" vs "is it better in day to day life" - and most metric stuff breaks down in the latter, meaning, the non metric systems seem to be perfectly fine in day to day life, but almost always better on paper.
Like Celsius - sounds great on paper, not any better (and in some ways, worse) in day to day weather/temperature.
One of the few exceptions I find is metric paper sizes. It's such a beautiful way of designing the sizes that is not only perfect on paper (pun proudly intended), is also totally practical in real life.
Speak to a carpenter or construction worker from any country with the metric system. Measure anything by simple powers of ten and simple decimal shifts makes life so much easier. Quick, what is half of a yard-length beam less 12 1/16"?
But realistically - most things are standardized. Nominal sizes. 2x4's, a sheet of plywood or drywall is 4x8 feet, etc.
(In fact, the metric equivalents of construction units are all based on feet/inches converted to a round mm because construction sizes are much older than metric)
So yes maybe you're messing with lengths, etc. 100% true. But unless you're doing fine carpentry, it doesn't matter.
The other thing is that it's a system that divides by 2 multiple times - it's not totally random. But metric only divides by 10 and 5 cleanly. You can't do a "quarter" of anything. Which is fine... but I'm saying that both systems have their advantages and disadvantages.
Assuming you mean (3 ft / 2) - 12 1/16 inches, it's about 6 inches, which is probably the accuracy you need. But it's exactly 5 15/16 inches, which is also easy to do in your head. Carpenters do it all the time, and even a doctor can.
My head did 36" divided by 2, so 18", minus 12 1/16", so 5 and 15/16". That is harder than working with quarters or fifths of meters, but easier than thirds or sixths of meters.
I don't know which is better. Considering how nothing stops you from doing decimal inches or fractional meters, but people rarely do that, both ways seem fine.
33.3 cm is easy to measure out, and 99.9 cm is exactly 1 mm short of 1 meter, well within tolerances for carpentry. Narrower than the mark left by your pencil.
Not sure how Celsius is worse? Its unit is the same of Kelvin, just shifted by 273. 0C is freezing, 100C is boiling. Some people say Fahrenheit is more representative of how people feel, but that is very subjective and biased towards the West and Northern hemisphere.
Celsius in a lab (or in cooking) is perfectly fine.
But as a way to describe daily weather (and thermostats) it's woefully inadequate.
The sum total of "comfortable" temperatures in celsius has to be squeezed between maybe 17 to 24 - why? you never ever use anything above 40 except when talking about people that died from a heatwave.
It's based on the freezing and boiling point of water _at sea level_ which almost none of us ever encounter (whether it's sea level, or the boiling point of water).
Fahrenheit is a much more intuitive scale (again... to describe daily weather or human temperature) - 0 - 100 better models the temperatures we experience daily, and you can intuitively separate into 10s, the way you "feel"
For example...
100s - honey, please, let's move to a cooler place.
90s - uncomfortably hot
80s - hot in the sun, not too bad at night
70s - you'd probably want a light sweater at night
60s - light sweater all the time
50s - medium sweater
40s - sweater, maybe a coat
30s - coat
20s - sweater and coat
10s - why am i living in this region?
0s - honey, please, let's move to a warmer place.
The point is that there's enough resolution to describe the actual human experience with temperature.
For everything else, celsius is perfectly adequate, totally agree.
FYI, none of this has ever made any sense at all to somebody who didn’t grow up with Fahrenheit. Numbers don’t need to go roughly between 40 and 100, and then occasionally 212 and 450.
The difference between 20 C and 20.5 C is not generally noticeable. The difference between 20 and 25 really is.
You’ll just have to face the fact that you feel comfortable with a really weird system that nobody else can see the point of, but it’s what you’re comfortable with. That’s OK, you’re allowed.
It's common here to say something like "this week it's in the 70s." In places that use Celsius, "in the 20s" must be too broad, so do they instead say something like "high 20s"?
Here in a metric country (Australia) they use words like cold, cool, warm, hot.
Otherwise people turn to BOM (Bureau of Meteorology) hour by hour / day by day records and forecasts of temp. and rain with radar cloud coverage - these just use nearest whole Celsius numbers, eg: https://www.bom.gov.au/location/australia/western-australia/...
Almost everything you've conjectured about "how they do it in a metric country" has been wrong; builders don't use a third of a metre, for example, everything is typically in millimetres - 1200, 1600, etc (aka, 1.2 metres, 1.6 metres) along side usages of 1 mm and 5 mm for small tolerances, expansion gaps, etc.
They don't use a third of a meter because, like you said, you can just use some number of mm or cm that is divisible by 3. We don't use decimal inches because fractions are easy too.
Why would you assume that this particular turn of phrase would be common at all?
People usually just say the temperature they’ve heard in the weather report, without giving a range. If you say “over 20”, people don’t assume “close to 30”, because then you would have said that.
Because it's really common here. This week it's in the 90s, which is very hot. The temperature changes day to day, so how do you give the approximate temperature for the week?
You’d just say “over 30” for that kind of temperature, or “almost 40” if it’s at the higher end. I think you’re taking your own unfamiliarity and just assuming that we’re all struggling because you don’t know how to translate what you’re used to, but I promise you nobody who is used to it is finding it inconvenient or difficult to use.
They also accidentally got MPH right for cars. Fastest highway is like 85. 100 is the fastest a normal car can go and also a reasonable point to arrest the driver.
I regularly see people going over 110 MPH on European highways. And no, I don't consider it reasonable to arrest people just because they reached middle RPMs on the top gear of their cheap SUV.
Celsius has plenty of resolution for this; in fact I would say Fahrenheit has too much. You can feel the change of one degree Celsius, you can’t really with Fahrenheit. To illustrate this, not only is your scale in tens, but I often only hear people quote the Fahrenheit temperature in tens. In Celsius we almost always give the whole number, because it makes a difference.
The widest range of temperatures found in any country is the USA. In the continental US alone the temperatures easily exceed -50°C and 50°C in both directions (much colder if you include Alaska). The USA also has the largest temperature change ever recorded in a 24 hour period, at 57°C.
Merits of specific measurement systems aside, the USA is arguably the most representative single country on Earth when it comes to temperature and "how people feel". It has among the hottest, coldest, and most extreme temperature ranges ever recorded.
It doesn't mean that. F is better than C for weather no matter what kind of climate you live in.
A few comments up, someone was saying F only makes more sense than C because you live somewhere with specific temperature ranges. But the US has all sorts of ranges.
Celsius isn't granular enough (people can perceive 0.5˚C changes), and anything above 50˚C is rare or nonexistent. 0-100˚F pretty much lines up with the extremes.
Empirically, my wife can perceive a 0° change on a bad day. The same temperature can be too hot and too cold at the same time. That said, I have never encountered a situation where 1°F makes any difference. If somebody feels bad, it takes at least 2-3°F change to make them feel good again. Celsius is granular enough for that. All workplace safety and food safety regulations in Europe are expressed in whole, non-fractional degrees Celsius.
What's so special about the number 50? Why aren't you comparing to 40°C instead? 40°C is a more useful boundary than both 50°C and 100°F.
Thermostats normally use 0.5˚C increments because people can feel the difference. Then they disable the decimals when in F mode.
50˚C is the highest temperature that regularly occurs in the hottest place I can think of, Death Valley. The point is, half the 0-100 range never gets used even in extreme settings. But yeah, in most places, more than half is unused.
Thermostats are made with way more granurality than needed, like body scales with 100 gram/quarter pound precision or battery testers with millivolt sensitivity. Most people have single zone HVAC with one controlling thermometer, and it's literally impossible for the whole house to have uniform temperature. And yet, most people are fine around the whole house, not just where the controlling thermometer is.
I still don't get why are you so fixated on 50°C. Why not 125°F? You'd get even more unused range if you go 0-125.
The drift is consistent if your doors aren't randomly open/closed. If I set my upstairs thermostat to 73˚F, the master bedroom with a closed door is always 71˚F. Downstairs thermostat is 70˚F, office is also 71˚F. We don't care about ±1˚F when working but do care when sleeping.
Also, I get some measures being pointlessly precise just cause it lines up with the units, but they deliberately add a decimal place to the displays to support ˚C, even in countries where ˚C is default. The same display could support 0.5˚F increments, but they intentionally don't.
> I still don't get why are you so fixated on 50°C. Why not 125°F? You'd get even more unused range if you go 0-125.
The 0-125˚F range gets used though. Usually 0-100 though. It's perfect.
To be clear, I don't have strong feelings. Obviously in a lot of science/engineering fields we use Celsius degrees a lot. But we also us Kelvin absolute scale a lot which would be ridiculous for day-to-day usage.
I am used to Fahrenheit but it is also a finer scale and a nice round number for boiling is largely irrelevant for the vast majority of purposes. Yes, knowing the freezing point of water can be important but memorizing one simple number probably isn't a huge stretch especially that the reality differs based on a number of factors. Arguably, knowing that 0 degrees F means fricking cold is sort of a higher latitudes bias but I'm not sure the degree to which Celsius simplifies.
The problem with "0F is fricking cold" is that so is 5F, and 10F, and even 15F to many people; and the difference between, say, -10F and +10F is pretty much none for practical purposes. Whereas knowing whether it was +1C or -1C during the night can have major impact on road safety.
With literally everything else about those two scales being a matter of preference and getting used to, this is the only objective difference with real world impact. Very tiny impact but still. Focusing on this one thing is the only way to answer the question of which is better with anything other than "whichever you're used to, it doesn't matter whatsoever". And if you focus on this one thing, then by the slightest of margins, Celsius is objectively better.
As someone who lives in a northern US state, there are so many other factors that affect winter road safety that aren't a binary above/below freezing temperature at sea level. I'm not really going to argue against Celsius but F is a very workable scale that has some (minor) benefits at human scale.
As far as I can tell, literally every single benefit of Fahrenheits I've ever heard of boils down to being used to dealing with numbers around 100 but not around -30. I have yet to hear about any such benefit that isn't 100% attributable in its entirety to being taught °F before °C at school.
All those other reasons are exactly why I said the difference is minimal. But that minimal difference is not zero. If you add every single such difference that exists between °C and °F, of which there's literally just one, then Celsius very clearly comes ahead, although by an amount way too small to matter.
Black ice is always dangerous. No matter how many other dangerous road conditions you can come up with, black ice is still one of them. And black ice happens exactly when the temperature crosses 32°F boundary multiple times. It's far more important to know if it was 31°F than to know if it was 30°F.
> Its unit is the same of Kelvin, just shifted by 273
You mean shifted by 273.15 because Kelvin doesn't have tidy fixed points at the freezing and boiling point of water, unlike Celsius and Fahrenheit.
Celsius and Fahrenheit are both utilitarian systems. Kelvin is the scientific system. Arguments about the relative utility of C° and F° are subjective and neither is particularly optimal for science.
I have a kitchen thermometer for making caramel (can't remember exact number but ballpark is 200C). I occasionally use my oven at 250C. My wife fires her ceramics in a kiln at around 1000C. I don't feel like I'm "wasting" anything here.
> Celsius isn't granular enough.
This argument would be reasonable if we didn't have to cut inches into 16 or 32 parts to approach decent granularity. At least we use decimal for cutting Celsius up...
Yup, because neither has significant advantage over the other. The only place where Celsius imo has a slight advantage comes from how often the phrase "below freezing" is used when talking about weather. That to me suggests that 0C/32F has some special significance, and thus makes Celsius a bit more elegant because it acknowledges that point. The granularity doesn't matter too much, as long as we agree to use decimals for ease of calculations, instead of using 1/32nds or some other barleycorns.
Only Americans feel this way about Celsius. The only downside is that Celsius requires 3 digit precision to really accurately characterize what temperature a room feels.
I would be willing to bet humans are woefully inadequate at knowing the temperature of the room they are in anyways
I do concede the overwhelming majority of people that argue that one is better than the other (both ways) are basically just arguing "the thing I grew up with is better" - because when you probe further, there isn't really any supporting argument, just "I understand this system better because I grew up with it" but it's equally vacuous coming from people who grew up with Celsius.
I came to the US when I was 12, so I grew up with Celsius but didn't pay attention to the weather as a young child. But still, I've lived in both systems.
As an American, I find Fahrenheit much better for ambient temperature and body temperature, but I hate seeing it for things like melting points of materials and the like. I'm always thrown off by statements like 800-degree pizza ovens, because once you get above 500°F it is out of the realm of everyday measurements, and I have trouble mentally separating the values for metals melting in °C and "gee whiz" temperatures in °F.
All scientific and engineering use should probably eschew the customary measurements, even in the news. For example, in aviation, they use °C now unless your airplane is very old. Unfortunately, in the US we still use inches of mercury for pressure, both knots and MPH but never km/h or m/s, etc.
Because in Celsius, 0 lets you know the water outside will likely be frozen and 100 lets you know the water in your pot is likely boiling. This is utterly useless information.
It is nowhere near as useful day to day as knowing 0°F means a very specific saltwater brine mixture will likely be frozen, and 90°F is kind of as hot as your insides. Or 96. Or 98.6 as the third revision decided.
If you actually measure a boiling pot (assuming you aren't at the beach), it will likely not be exactly 100 degrees, but less (or more, haven't had my coffee so I can't calculate).
So for most day to day stuff, it's equally arbitrary. And pretty much _nobody_ needs a thermometer to tell if something's boiling, you wait for a rolling boil. Or frozen, you wait until it looks like ice.
So again, great on paper, but in day to day use, not necessarily more useful than F.
I find F way more intuitive for weather. The larger scale helps. 100 is hot. 0 is quite cold. Boiling point is irrelevant. 32 freezing is the only thing that feels arbitrary.
50 hurts to the touch, your chip has likely broken and given off its black smoke
30 is good for swimming in cold water
>25 is worth worrying about sun stroke. make sure to drink a lot of water
24-26 is a nice summer day
18-22 is comfortable room temperature
>15 is good enough shorts weather
0-10 needs at least a light jacket
<0 you have to be attentive about frost bite, but not worried
-5 either wants gloves or good pockets
-15 wants a sweater and jacket
< -15 you should probably plug in your car
-10 - -25 is good downhill skiing weather
-20 - -27 is right cold, full winter gear: toque, scarf, mittens, thick socks, the whole kit and kaboodle, but still chic
-30 - -40: ski version of the full winter gear, including snow pants, double socks, and good boots. both your warmest sweater and ski jacket
----
its really just what you grew up with that will make sense.
like, wtf is 72 and 54 and mid 80s? i have no backing reference to put them into scale, even after living in the states for a decade. 500, 275, 425, 150 do make sense in F for me because thats what ovens list
I find that increments of 1C are too small to be useful in daily life but increments of 5C are quite big. As a result, you can't really round degrees Celsius to a multiple of 5 whereas full precision integer temperatures are too precise. So for example 15C and 17C feel like different temperatures, but each 1C step from 15C to 16C to 17C is not really useful.
In contrast, it's very reasonable to round temperature to increments of 5 degrees Fahrenheit. So for example 58F and 60F are basically the same temperature as far as my perception is concerned.
It's literally does not matter in day to day life. In the end you get to X degrees is cold, Y degrees is perfect and Z degrees is hot regardless of C or F.
I grew up using C and now I'm using F, if I move to a country that uses C, I will switch again. Its meaningless.
This is the correct take. The better system is the one that you have intuition in - just keep using that one if building up intuition in a different system is not worth the cost to you. At least when it comes to temperature. :)
What are you talking about? The same can be said about 59F (15°C) and 60F (15.56°C)
"One fahrenheit increment doesn't feel like anything"
At least Celsius are grounded in reality, should I be worried about ice? Just check if temperature is below 0°C. But hey you may need to check if your solution of water, ice and ammonium chloride is frozen, so I just need to check if below 0F. That's probably something you do often right?
theyre both grounded in reality because the represent real temperature conditions.
theyre both just arbitrary degrees though, since 0 isnt actually 0. they both have better alternatives too - Kelvin and Rankin. your criticism about freezing applies to celsius too. freezing is at a certain pressure and temperature, so both the 100 and 0 celsius only sometimes describe boiling and freezing
...making the point that Fahrenheit's 0-100 range is better suited to normal day to day measurements than 0-100 Celsius. The numbers that feel easiest to communicate with are scaled nicely relative to the range of temperatures you most often want to communicate about.
> It is more precise. Fahrenheit has more frequent degrees, allowing for greater resolution with analog thermometers.
this is really about the graduation and size of the thermometer? you can print a useful line pitch as you like, and have it represent what you want, where thats a 0.25C, 3F, 10c or whathaveyou.
everyday temperatures are pretty well centered around 0C, half the year above, the other half below, and the 10, 20, 30, and 40 increments feel quite similar to each other in severity of the season. +30 is damn hot and -30 is damn cold. +/-10 is all mildly warm or cold, with +/-20 really being where you do seasonal acitivites like swimming or skiing.
as far as western units go, C fits fantastically for the the traditional western culture with 4 idealized seasons. I could see F being better for places outside the idealized west - eg. the american south/sun belt, the PNW, southeast asia, south america, a lot of africa, etc where the seasons are different
The only connection off the top of my head is that mL of water increases by 1˚C if heated with 1 calorie. Which is annoying because 1 Calorie is 1000 calories.
I don't agree with "nor should it be" - the market decided it was clearly needed. And this isn't unique to Tailwind. For decades, companies in many ecosystems (like .NET, JS, etc) have made lots of money selling component libraries that save developers and companies a lot of time. And I even bought UI.sh in April, and it really was great until the models just natively got much better.
I happily paid for Tailwind UI because it was worth it. I paid for ui.sh and got plenty of my $130 (or whatever) value. The market spoke in a good way initially, and now it's sending a different message.
They've just got to do the next thing, which is find an AI-resistant steward of the project now.
"nor should it be" is connected to "being worth millions". It's not that it isn't worth money, but it's more of a lifestyle business than the sort of thing that will drive a tech company with employees, marketing, and conference budgets.
I think if you said "couldn't be a venture backed startup", I'd 100% agree with you. Can't support employees or a marketing budget is a much lower bar, and I think that's perfectly achievable by something like Tailwind (or I guess was prior to AI)
There are several graphing libraries that have been businesses for years. Like Highcharts. A frontend framework used by millions and millions certainly has some value to loads of people.
"The market can stay irrational longer than you can stay solvent" applies to running a business as well as trading. There is no certainty when you're dealing with the individual whims of millions of people.
I’m not sure what their valuation was but single digits millions sounds totally reasonable for a company you’re describing. Just to hire 10 engineers you’ll need to be in low millions
I don't want to diminish the work that any developers do, but a headcount of 10 seems high to me. (Unless some of those are simply "support engineers" who do nothing but support paying customers, but that would go up and down with revenue)
To hire 10 engineers you only need millions in yearly revenue. And millions in revenue does not imply that the business is worth millions. It very well, quite easily, could be worthless.
Just for reference, in the .NET ecosystem there are multiple vendors selling UI component libraries for sums ranging from lower three digits to upper three digits per dev, per year. This was generally considered good value (might no longer be great value post-AI)
As an engineering leader back in the day (2000-2015) I considered it terrific value to license Telerik and others. In fact, it looks like Progress has acquired a large catalog of the most common .NET dev tools over the past ten years.
> I happily paid for Tailwind UI because it was worth it.
The business was worth something. But, as the earlier comment questions, was it worth millions? Or was that the free money, tech hype talking? Maybe in a more normal market it would have only been a simple mom and pop.
I think that “lots of money” needs to be put in perspective when thinking about software that’s used by developers. When you compare the size of the market for these kinds of tools compared to a truly massive market for something like an office suite, it looks pretty puny.
Developers are also very fickle and the one constant is change. Sometimes it seems like change for changes sake alone.
I’m not saying that any businesses easy just that it seems to me after seeing a lot of companies come and go, but this is one of the more difficult markets to capitalize on.
It’s hard to name a more pathological market than office suites. You have Microsoft which is lackadaisical about the subject, like maybe they think Office is a place to stuff with thoroughly unwanted AI features. The rest of the market is German zealots who spend all day with bash and emacs who lobby legislatures to subject public servants to off-brand office suites and then the public sector unions go on strike to get Microsoft Office back.
Bundling is one reason who can’t have nice things. Like in my opinion Excel is a lot worse than the others but if customers are thinking about a “suite” they are still going to buy Office to get Word and PowerPoint and can’t save money by switching to a different CSV wrangler or financial analysis tool.
Funny, my opinion is the opposite. Excel is a much better spreadsheet than Word is a word processor or than PowerPoint is a presentation tool.
Of course it’s got quirks and unfortunate relics of its long history, but I reach for it willingly, and I can’t say that about the rest of the Office suite.
The market didn’t decide it was needed, the users just downloaded a bunch of free files and free docs. There isn't any indication there was ever a real market.
Using something free and paying for a product are entirely different things.
Hey do you have any opinion on the current utility of ui.sh? I wanted to buy it but I'm a bit wary of it just not being useful and they don't have a preview of a single skill.
"The market" for Tailwind only existed due to ignorance and obfuscation fueled by hype for web dev jobs.
Completely artificial market built on propaganda and low interest rate money printing. Same as the rest of SaaS. Nothing about the tech industry since the '08 crash has been organic investment. Was 100% collusion between government and private interests to manipulate public opinion.
> They've just got to do the next thing, which is find an AI-resistant steward of the project now.
Aside; all the leaked emails, chats, and going back further in time, telegrams, letters, travel logs on display in court yet the myth the economy is an organic free market this whole time persists
Such belief is nought but a quirk of biology that allowed humans to latch onto the idea Benevolent but Punitive Sky Wizard watches out for them; total cope. In Wealth of Nations Adam Smith called it the "invisible hand" and later came to realize he was wrong. He reconsidered and concluded the avarice of man will not be stopped by philosophy of national unity.
No, I didn't mean ai-resistant as far as "don't need coding agents" - I use agents to do all my coding now. I'm saying, AI-resistant in the sense that a company like Shopify isn't going anywhere even in the AI world. If anything, they will thrive with more AI-assisted shops coming online.
That gives the Tailwind folks a more steady stream of income.
this. leftpad as a service bullshit SaaS companies somehow raising millions was a temporary insanity driven by ZIRP and incestuous VC money currents. we see the same thing happening now with companies whose products are paper-thin vibecoded wrappers over someone else's models.
Unless this ban actually resembles something like global nuclear non-proliferation treaties, it would make absolutely no sense for us to cripple ourselves when someone like China continues full speed ahead.
I don't know what the solution is, but what I do know is almost nothing good will come out of _just_ the US pausing.
Or is it that these ATC networks are their own air-gapped network with less redundancy? That just doesn't add up. Or maybe there was only one line going to the ATC, with no multiple "ISPs" like a datacenter would have?
reply