Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Leaked Samsung exFat driver relicensed as GPL

lists.gpl-violations.org

Post reply on HN
Print view View original

71–80 of 206 posts

Re: Leaked Samsung exFat driver relicensed as GPL

#71

Post by regularfry »

Nursie wrote: post #56
gngeal wrote: post #54

Earlier quoted context omitted.

Did Samsung, in fact, distribute the code in Android devices? If the answer is "yes", wouldn't that make it GPLd?

Not necessarily, no. There is a long history of people releasing closed source drivers, particularly on embedded platforms. So far as I can tell the kernel devs think it's harmful and A Bad Thing (TM), but not a violation. Also as has been pointed out below, it wouldn't automatically make it GPL. It would make it in violation of GPL and releasing under GPL would fix the violation, but it wouldn't make it GPL.

There's a 10-year-old post to LKML from Linus saying the GPL nature of kernel modules is a huge grey area. I don't think it's been resolved since then.

Re: Leaked Samsung exFat driver relicensed as GPL

#72

Post by dmytrish »

andreasvc wrote: post #46
effn wrote: post #34

Earlier quoted context omitted.

No. The authors of GPL-licensed code has no say in how other people license their code they don't own the copyright on. Derivatives of GPL-licensed code is either: 1. GPL-licensed 2. Illegal If it's not explicitly 1 by the license on the derived work, it becomes 2. It is not automatically "forced" into category 1.

By illegal you mean illegal to distribute. If Samsung used this code internally and it was leaked accidentally, they are not in violation of the GPL.

If Samsung ships binaries produced from this code to customers and this code is derivative from GPL code, they _are_ in violation of the GPL.

Re: Leaked Samsung exFat driver relicensed as GPL

#73

Post by regularfry »

agf wrote: post #29
lucian1900 wrote: post #23

Earlier quoted context omitted.

Those are technically GPL violations, but kernel devs haven't (yet) complained much.

That is absolutely not the community consensus. Linus and the kernel community as a whole have taken the position that this is specifically allowed by the GPL as using a public interface isn't creating a derivative work.

Not unless something's changed recently: http://lkml.indiana.edu/hypermail/linux/kernel/0312.0/0670.h...

Re: Leaked Samsung exFat driver relicensed as GPL

#74

Post by susi22 »

regularfry wrote: post #71
Nursie wrote: post #56

Earlier quoted context omitted.

Not necessarily, no. There is a long history of people releasing closed source drivers, particularly on embedded platforms. So far as I can tell the kernel devs think it's harmful and A Bad Thing (TM), but not a violation. Also as has been pointed out below, it wouldn't automatically make it GPL. It would make it in violation of GPL and releasing under GPL would fix the violation, but it wouldn't make it GPL.

There's a 10-year-old post to LKML from Linus saying the GPL nature of kernel modules is a huge grey area. I don't think it's been resolved since then.

Correct. Link:

http://lkml.indiana.edu/hypermail/linux/kernel/0312.0/0670.h...

Re: Leaked Samsung exFat driver relicensed as GPL

#75

Post by josteink »

girvo wrote: post #68
rquirk wrote: post #51

Isn't this is a general trend with Android/Cyanogenmod hackers? If you go on something like xda-developers then there's a sticky saying "please don't post GPL violations". Early Cyanogen releases were fast and loose with redistribution of Google proprietary apps, which got the author in a spot of trouble IIRC. Right now the distribution of these ROMS is IMO pretty suspect - they include whole swathes of binary blobs…

The ROM scene has always been a grey (verging on black) area when it comes to IP, however the enthusiast scene for the original WinMo and Android has helped the market in general. There has always been a bit of a gentleman's agreement with the manufacturers that if we went too far, we'd be slapped back, but thankfully this has only happened a couple of times. Indeed, the manufacturers themselves have been known to le…

> The ROM scene has always been a grey (verging on black) area when it comes to IP

Speak for yourself.

There are two scenes: The binary modder scene, where people retro-fit vendor-created and provided firmware-images with their own modifications. Commonly referred to as "Photoshop ROMs". This scene is clearly grey at best.

However, there is another scene: The one based on AOSP source and derivates of it. This scene lives on Github and similar places where open source is shared and reused freely, as intended.

This is driven by serious developers and focus is on doing things cleanly, properly and reliably.

Most successful and widely deployed ROMs (like Cyanogenmod) are based on this scene, and there is nothing grey or remotely shady about it.

Re: Leaked Samsung exFat driver relicensed as GPL

#76

Post by coldpie »

wisty wrote: post #52
maaaats wrote: post #43

Earlier quoted context omitted.

Wow, again. Also found: > Now, putting it publicly on github qualifies as distribution, and therefore this code is also under the GPLv2 license implicitly. The act has already been done, and even if it does contain copyrighted bits by MS, the released version is now forever GPL. so, yeah.. by this logic I can take my company's code, put it on GitHub under a GPL license and now everyone is free to use it under GPL. Do…

I can sort of see is logic. If a company has a private stash of GPL code (say, Google is running a special fork of Linux), and it gets leaked, even though the leak was illegal the code might still be distributable under GPL. But ... this code was never (afaik) legitimately GPL. Maybe Samsung had based it on GPL code (in which case, it's GPL), but we'd need some proof.

Here's a couple of related answers at the FSF GPL FAQ: http://www.gnu.org/licenses/gpl-faq.html#InternalDistributio...

Unfortunately, the question isn't really directly answered.

The most relevant passage seems to be, "If the [unwittingly distributed] version in question is unpublished and considered by a company to be its trade secret, then publishing it may be a violation of trade secret law, depending on other circumstances. The GPL does not change that."

So I think a court would rule that the original leak does not qualify as "distribution," and thus the company is not required to distribute the source and everyone distributing the leaked code is under copyright violation.

Re: Leaked Samsung exFat driver relicensed as GPL

#77

Post by belorn »

Nursie wrote: post #32
lucian1900 wrote: post #23

Earlier quoted context omitted.

Those are technically GPL violations, but kernel devs haven't (yet) complained much.

I've been able to find a statement from 2008 in which many kernel devs publicly said they think that closed drivers are harmful. But violations? I'm not convinced. Got a link that might convince me?

Depend which you want to listen to. If you want to listen to what lawyers think (fsf), then there is a GPL violation( https://www.gnu.org/licenses/gpl-faq.html#NonfreeDriverKerne...).

If you want to listen to kernel hackers, (such as Linus Torvalds), then he doesn't say if it is, or isn't a violation. Rather, he says that developers has a right to write non-free drivers to the kernel using modules (http://linuxmafia.com/faq/Kernel/proprietary-kernel-modules....).

So... what does that mean? Im not a lawyer, but it looks to me as an GPL violation which has been given permission to exist by one of the kernel developers. That it is the project leader that gives permission does lower the risk that any individual kernel developer will suddenly go out and start suing companies for GPL violations, even if they has a legal right to do so under copyright.

Re: Leaked Samsung exFat driver relicensed as GPL

#78

Post by serf »

reginaldjcooper wrote: post #31
bxc89 wrote: post #20

Earlier quoted context omitted.

There are those of us who completely disagree with copyright law. How many years have been wasted on projects like Samba because of draconian copyright? Much simpler and cleaner to acquire the source.

Wasting man-years on reimplementing software is not a large enough problem to decide that authors no longer have any rights with their product.

but it is a large enough problem.

Wasting man years is precisely the problem. Our collective knowledge is boundless, but time is finite. See the problem here? Why would we want to waste the effort of future generations with the problems of yesterday?

So that an individual can live more comfortably with exclusive privileges to something that benefits all of mankind? We shouldn't be working towards selfishness and arbitrary contractual law.

Re: Leaked Samsung exFat driver relicensed as GPL

#79

Post by belorn »

kristofferR wrote: post #41
agf wrote: post #29

Earlier quoted context omitted.

That is absolutely not the community consensus. Linus and the kernel community as a whole have taken the position that this is specifically allowed by the GPL as using a public interface isn't creating a derivative work.

Linus Torvalds disagrees with that: https://en.wikipedia.org/wiki/Linux_kernel#Loadable_kernel_m... https://en.wikipedia.org/wiki/Loadable_kernel_module#License...

If you use Wikipedia as source, please verify that the section you are referring has a source. Both now has citation needed tags on them.

Re: Leaked Samsung exFat driver relicensed as GPL

#80

Post by fulafel »

josteink wrote: post #75
girvo wrote: post #68

Earlier quoted context omitted.

The ROM scene has always been a grey (verging on black) area when it comes to IP, however the enthusiast scene for the original WinMo and Android has helped the market in general. There has always been a bit of a gentleman's agreement with the manufacturers that if we went too far, we'd be slapped back, but thankfully this has only happened a couple of times. Indeed, the manufacturers themselves have been known to le…

> The ROM scene has always been a grey (verging on black) area when it comes to IP Speak for yourself. There are two scenes: The binary modder scene, where people retro-fit vendor-created and provided firmware-images with their own modifications. Commonly referred to as "Photoshop ROMs". This scene is clearly grey at best. However, there is another scene: The one based on AOSP source and derivates of it. This scene l…

The open source subset of Android is not practically usable though. There aren't many (any?) devices that work usefully without a lot of proprietary bits.
Post reply on HN