zlacker

[parent] [thread] 11 comments
1. ranger+(OP)[view] [source] 2025-06-12 05:51:27
No it's not, they're just not releasing sources until a new version is actually released, and there's nothing inherently wrong with that.
replies(3): >>fc417f+Z >>charci+51 >>pzo+A1
2. fc417f+Z[view] [source] 2025-06-12 06:05:48
>>ranger+(OP)
Which makes life more difficult for downstream projects. However the recent release didn't include the usual device-specific sources.

> This means AOSP 16 cannot currently be built or run on any recent Pixel device easily just using official source. It’s unclear whether this is a delay or a policy change. Either way, it seriously disrupts custom ROM development and our porting efforts.

https://calyxos.org/news/2025/06/11/android-16-plans/

replies(1): >>immibi+g4
3. charci+51[view] [source] 2025-06-12 06:08:11
>>ranger+(OP)
The alleged claim being made is different than the previous official announcement.

The claim here is that AOSP will stop releasing sources publicly altogether similar to other versions of Android like Wear OS.

replies(1): >>flakin+K1
4. pzo+A1[view] [source] 2025-06-12 06:16:44
>>ranger+(OP)
seems not exactly true, they did release android 16 to AOSP but "Google did not publish any device-specific source code for supported, modern Pixel devices."

https://calyxos.org/news/2025/06/11/android-16-plans/

◧◩
5. flakin+K1[view] [source] [discussion] 2025-06-12 06:21:29
>>charci+51
FYI: https://source.android.com/docs/setup/about/faqs#android-lat...

> Google pushes the code for the next release to the latest public release branch and updates the android-latest-release manifest to point to that branch.

Along with https://source.android.com/docs/whatsnew/site-updates#aosp-c...

> The android-latest-release manifest is set to the latest AOSP release branch, android16-release

◧◩
6. immibi+g4[view] [source] [discussion] 2025-06-12 06:58:27
>>fc417f+Z
Once they give you the device, they're legally required to give you the source code. Most manufacturers don't, but you can sue them and win.
replies(2): >>rob_c+n4 >>flotza+c8
◧◩◪
7. rob_c+n4[view] [source] [discussion] 2025-06-12 07:02:47
>>immibi+g4
I think they're counting on that fight. Good luck, I look forward to seeing the gofundme...
replies(1): >>lipowi+k5
◧◩◪◨
8. lipowi+k5[view] [source] [discussion] 2025-06-12 07:17:42
>>rob_c+n4
This is more an opportunity for the EU to suddenly play harder ball than expected.
◧◩◪
9. flotza+c8[view] [source] [discussion] 2025-06-12 07:55:45
>>immibi+g4
Most of AOSP is Apache 2.0 licensed (permissive, not copyleft)
replies(1): >>fc417f+ba
◧◩◪◨
10. fc417f+ba[view] [source] [discussion] 2025-06-12 08:20:10
>>flotza+c8
I believe the device tree stuff is GPL due to the kernel. Might be mistaken though.
replies(1): >>immibi+Jn
◧◩◪◨⬒
11. immibi+Jn[view] [source] [discussion] 2025-06-12 10:47:41
>>fc417f+ba
A device tree is probably counted as data, so not copyrightable. I am not a lawyer and this is not legal advice. But there's plenty else that is GPL, and in AVM vs Sebastian Steck, it seems that even LGPLv2 requires the manufacturer to give you all the stuff you need to install a modified version of the software, i.e. it seems that all GPL licenses imply anti-tivoization.
replies(1): >>ranger+yN
◧◩◪◨⬒⬓
12. ranger+yN[view] [source] [discussion] 2025-06-12 13:40:45
>>immibi+Jn
> AVM vs Sebastian Steck

That case was dismissed because they settled privately, so technically it does not prove anything about LGPLv2, and Germany is not as big on blindly following precedent in law as other countries anyway.

[go to top]