HN Simulatornew | past | comments | lists | submit | p0w3n3d's commentslogin

Hmm Fortran example is so positive... It means that in 2090s there will still be programmers hired to understand the code created by AI in 2020s

Go home! Goober!...

I mean Google!


Paywalled...

My thoughts: I could pay a subscription that would be divided by the portals I visit. For example I don't want to pay 5$ subscription to new Yorker, 5$ to BBC 5$ to times, 20zl to Rzeczpospolita etc because it would take unreasonably large amount of money. If I could pay 100zl-20$ and use them to see some articles in those portals, I would be happy and I would pay, because now I just lick the window and go home



The architecture has been changed fundamentally. AFAIK the old windows 2000 had flaws beyond repair. On the other hand the UI overlay build on UI overlay built on UI overlay, making everything a html page etc was terrible direction

You might be thinking of win9x. Win2k had no flaws and was perfect in all ways. I will not be taking questions.

My bet is on Windows Millennium Edition (Windows ME). That was broken version. Windows 2000 was good.

While WinMe was 98 successor and meant for home use, Win2k was business oriented.


God WinME was such hot garbage. Windows 2000 Workstation ran everything that ME ran. I fucking gamed on it, and did everything. It was faster and more stable, than ME. I never understood why people I knew who were computer nerds ran that steaming pile of shit.

> Win2k had no flaw

Win2k could only display 16-color icons in its systray


Nobody needs more than 16-color icons.

—Bill Gates


Interestingly enough Windows 7 boots faster (in a VM) than both Windows 10 and Windows XP. It feels faster than Windows 10 while running too.

Trusted Platform Computing in 3... 2... 1...

We are already there and it is called Play Integrity (meets STRONG integrity) and Google is trying to get apps to use it to shut out competitors. They are succeeding to some extend (see the Volkswagen and IKEA Smart Home apps as recent successes for Google).

Google should not be allowed to abuse its duopoly position in this way. I recommend EU citizens to report every case they are (potentially) affected by to the EU DMA team.


> see the Volkswagen and IKEA Smart Home apps as recent successes for Google

Someone should make a complaint about ePD violation in relation to those. I doubt it's "strictly necessary" for those applications to read that information (if they even fulfill the "at distance" requirement).


I think that the EU shall render this practice (and the equivalent from Apple) illegal, stating that every citizen has a right to modify their own device without consequences, and that consequences shall include also having apps not working.

But to me big tech succeeded also lobbying the EU like they already did in the US, in fact they violate multiple EU rules (GDPR, etc.) without facing any consequence, and when they try to announce they would do something at the other side of the ocean they treat us that they will double or triple tariffs so...


> stating that every citizen has a right to modify their own device without consequences, and that consequences shall include also having apps not working.

iOS being so difficult to jailbreak means that most apps don't have any verification at all. Even most DRM doesn't even try to verify that the device isn't jailbroken in order to access the highest resolutions or whatever. But if we ever add a first-party option for the device to be modified, then we start running into the Android situation where every app has some invasive root/tamper detection thing. Some apps already have anti-debugging things, but the situation on iOS could be far, far worse.


iOS has its own Play Integrity equivalent with DeviceCheck (https://developer.apple.com/documentation/devicecheck) and banking apps on iOS use it widely just like on Android as well.

DeviceCheck is not about detecting if the device is jailbroken, it's about detecting if the app is being emulated or tampered with.

A distinction without a difference - the purpose is to verify that the device on which the app runs hasn't been tampered with (for Play Integrity or AppCheck) from OEM state.

That's not true. Jailbroken devices have been tampered with, but the Secure Enclave is still fully functional. Device Check is designed to catch non-genuine hardware or non-genuine app binaries, NOT tampered software.

Depends on the type of jailbreak

Hmm, you're right. It looks like some older devices were able to be jailbroken into a state where the Secure Enclave was in an untrusted state, so you couldn't use authentication, but the encrypted flash was still accessible as long as you didn't have a passcode set. I thought even that wouldn't work in an untrusted state, but it appears I was incorrect.

(I was reading recently about the SEP's behavior after a memory authentication failure -- it completely stops accepting requests until the processor is reset -- so I thought that applied to trusted boot violations as well. Seems not.)

I suppose that also explains why every time I open Settings, the Apple Pay section is missing for a moment before it shows up; the OS is apparently able to function without it. Interesting bit of trivia.


If AI is good enough to find RCE in google code, maybe it can crack TPM

This thinking never leads to anything good. We got to stand for our right before it's too late and everything requires hacking

Maybe the TPM developer should be checking this using AI?

This ideea haunts me in my nightmares.

Trusted computing is fine and even good only if the owner (the person/company that bought the device) is in control, but it's not the world we live in.

> owner (the person/company that bought the device)

At this point I'm willing to argue that the owner of a device is whoever is legally allowed to log in as root. When I pay for server colocation, I'm the owner - and not whoever runs the building - because I decide what runs on the box, and if they tamper with it they can go to jail.

With phones it's the same, except that you pay Google for the privilege for keeping their device safe, charged, connected to the Internet, and walking it around every day with you, so they can collect better data. The only thing they can't do is snatch it from your hands, but why would they - you're doing so much useful work for them.

And of course if you try to root it, you're breaking a digital lock, which is punishable by up to 10 years in prison and $1,000,000, so the colocation analogy holds there too.


cgi app is basically the same idea as a Lambda. Stateless app, that can be run virtually anywhere

The downloaded model is stored in

  Library/Application\ Support/ai.haku.eg2demo

To be honest these ubiquitous AI hardware news look to me more and more like Maybach commercials.

Not targeting me


You get an AMD GPU, put it in an external case ("enclosure") connect it with usb-c or thunderbolt and that's what's being used in the app. For inference of course

But then how do you even make the driver install on the ipad ? Aren't drivers supposed to run under system-level privileges (which iPadOS will never let you run under)

DriverKit

>the upstream Linux amdgpu driver embedded as DriverKit

https://developer.apple.com/documentation/driverkit

>The drivers you build with DriverKit run in user space, rather than as kernel extensions, which improves system stability and security. You create your driver as an app extension and deliver it inside your existing app.


AI will be used for total surveillance in the first place. So totalitarian governments, welcome on in


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: