Hacker Newsnew | past | comments | ask | show | jobs | submit | asnelt's commentslogin

That's the premise of Tau Zero, a very good novel:

https://en.wikipedia.org/wiki/Tau_Zero


I hear you. This is exactly the reason why I have a digital signage display rather than a Smart TV.

> Can we take this as an official statement that Nvidia supports local models?

The first reply on the X thread is by Gerardo Delgado, who is "Sr. Director of Product - Local AI, Creators, Developers @NVIDIA" according to his profile. In his post, he says "My team's goal at NVIDIA is to grow Local AI."


Even with a burner sim, there is the International Mobile Equipment Identity (IMEI) number, which is tied to the phone, and is known to all apps with the android.permission.READ_PRIVILEGED_PHONE_STATE permission.


That can't be true? <https://grapheneos.org/faq#hardware-identifiers>

  As of Android 10, apps cannot obtain permission to access non-resettable hardware identifiers such as the serial number, MAC addresses, IMEIs/MEIDs, SIM card serial numbers and subscriber IDs. Only privileged apps included in the base system with READ_PRIVILEGED_PHONE_STATE whitelisted can access these hardware identifiers. Apps targeting Android 10 will receive a SecurityException and older apps will receive an empty value for compatibility. The currently enabled carrier-based messaging app for SMS/MMS/RCS is a special case that's given access to certain device identifiers including the IMEI. This is normally the GrapheneOS fork of AOSP Messaging but can be changed to another app by the user.

  Since these restrictions became standard, GrapheneOS only makes a small change to remove a legacy form of access to the serial number by legacy apps, which was still around for compatibility. It used to need more extensive changes such as disallowing access to the serial number but those restrictions are now standard.

I don't know however if sandboxed google play is such a privileged app.


I couldn't immediately find whether GrapheneOS grants READ_PRIVILEGED_PHONE_STATE to Google Play. It might very well be that the GrapheneOS sandbox spoofs a fake IMEI, and I do hope so.

In any case, my parent comment was meant for stock Pixels, as mentioned by armadyl further up in this thread.


https://grapheneos.org/usage#sandboxed-google-play

> Google Play receives absolutely no special access or privileges on GrapheneOS as opposed to bypassing the app sandbox and receiving a massive amount of highly privileged access.

It doesn't mention IMEI here, but hopefully READ_PRIVILEGED_PHONE_STATE is included in "privileged access."


There is an AppStore app, I am not sure if this is the one we are talking about? <https://github.com/GrapheneOS/AppStore/blob/main/app/src/mai...>

That one lists:

  ACCESS_NETWORK_STATE
  ENFORCE_UPDATE_OWNERSHIP
  FOREGROUND_SERVICE
  FOREGROUND_SERVICE_SPECIAL_USE
  INSTALL_PACKAGES
  INTERNET
  POST_NOTIFICATIONS
  QUERY_ALL_PACKAGES
  RECEIVE_BOOT_COMPLETED
  REQUEST_DELETE_PACKAGES
  REQUEST_INSTALL_PACKAGES
  UPDATE_PACKAGES_WITHOUT_USER_ACTION


That's grapheneos's own app, separate from the play store or play services.


True. I think this one is closer to the truth: <https://github.com/GrapheneOS/platform_packages_apps_GmsComp...>

There is no READ_PRIVILEGED_PHONE_STATE mentioned there.


That's also incorrect, because the gmscompat app is just a helper app. Play services can and does request additional permissions. Those permissions are handled by the OS under the play services app, not gmscompat. If you want RCS for instance, you must grant play services and google messages phone and ICC auth access, which isn't seen in gmscompat at all.


  > That's also incorrect, because the gmscompat app is just a helper app.
Hmm, ok. I was reasoning the helper app was needed to get around the default assumptions from Google Play Services.

  >  Those permissions are handled by the OS under the play services app
Yes, but I assume you don't mean that as that GOS makes special hard-coded provisions for the play services. GOS claims to run Play Services like any other unprivileged app, and so any additional permission it would want would have to be consented by the user and should be visible to the user. If not, then GOS wording would be quite a bit unfortunate at least.

EDIT: "GmsCompatConfig is the text-based configuration for the GrapheneOS sandboxed Google Play compatibility layer. It provides a large portion of the compatibility shims." [1] This seems to indicate that the permissions requested by Play Services are being honored with the shims from the helper app. That would alleviate the permission problem.

1. https://github.com/GrapheneOS/platform_packages_apps_GmsComp...


Google Mobile Services apps installed on GrapheneOS including Play services run as regular sandboxed apps. They receive absolutely no special access compared to other apps by installing and running them. There are the standard permission toggles for granting those but none of those are required for typical usage to provide compatibility with many apps from the Play Store depending on their services.

There are additional special permission toggles for RCS and Android Auto. The issue with RCS is mainly that they split the implementation across Google Messages and Play services. Android's standard permission model gives special access to the app selected by the user as the messaging app but Google Messages expects Play services to have special access too.

The shims defined with GmsCompatConfig are a small subset of the overall compatibility layer. It has many shims which need to actually implement the functionality such as remapping the Play Store using privileged installation APIs to the regular ones available to user installed app stores. It has to remap the APIs used by dynamite modules to ones not requiring privileged SELinux policies too. It has a mix of shims which simply stub out the functionality and many which need to handle it as a regular sandboxed app would need to do it.


Thank you for clarifying this.


The only way to install Google Mobile Services apps including the Play Store on GrapheneOS is as regular sandboxed apps. They aren't granted any of the large number of usual privileged permissions, don't run with the usual far more privileged SELinux policies, aren't used by the OS as a backend for anything and aren't otherwise allowed to do special things by the OS in the usual many ways that is granted. They're regular sandboxed apps on GrapheneOS.


That's the application software side. I would assume the IMEI and IMSI are both going out to the cell network though, and I would presume that it's trivial to tie a phone number to those with how the mobile industry generally sells subscriber data to various data brokers. The only question is how permissive those data brokers are (their major constraint is how much most people become aware of this dynamic), but when dealing with a major APT like Google I'd assume they're tuned into the best ones with songs about bona fide purposes.


Are you talking about the US here? I am hoping this would be off-limits in Europe.


Yes I am talking with a US perspective. I would hope the GDPR would prevent such things in (most of) Europe. But I also personally wouldn't assume so given that there are still the same dynamics of keeping the info flows private to avoid scrutiny, and claiming plausible "legitimate purposes" and "consent".


Play services isn't an "app", it's a highly privileged system service that has access to everything and some more.


The only way to use Play services on GrapheneOS is as a regular sandboxed app. It runs in the standard app sandbox without any of the usual privileged permissions, privileged SELinux policy and many other forms of special access. It also doesn't get used by the OS as a backend for anything. Our sandboxed Google Play compatibility layer implements this by remapping APIs to ones available to sandboxed apps and stubbing out many which aren't needed.


yes, but not on GOS.


The outcome of this vote is not set in stone. If the situation changes, there can be another general-resolution vote to adapt to the new situation. I think the result of this vote is appropriate for the current degree of automation when used responsibly.


Exactly. Just tweaking the order of suggested models for a search query could already have a huge impact on model popularity.


That's an interesting video. If the skyscrapers were flickering just a little faster, the sky would be the color of a television tuned to a dead channel.


(For those who did not get it, it’s the first line of Neuromancer).


And from John's parent comment, "The future is here, it's just not evenly distributed." is a quote attributed to William Gibson.

We're all just quoting authors now.


That's just a normal night. For special events, the skyscraper lighting is much more dynamic.[1]

[1] https://www.youtube.com/watch?v=FDNkcKiD2Ow


> If it bears the mark, it is likely generated (or someone deliberately made it look generated).

One could even say, the mark is load-bearing.


Given the context, I had to read this twice to understand that PR is not referring to Pull Request here.


Reminds me of the floating umbrella that appeared on Caladan in the 1984 Dune movie.


Do you mean the glowglobe rather than an umbrella? The glowglobes were used for lighting.


Yes, I do mean the glowglobe. As they are floating right above the person, they should also protect from rain. And there was a scene where one was floating through the rain.


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

Search: