Discussion:
Licensing stolen/leaked code as GPL
Andrei Frumusanu
2013-07-22 00:24:38 UTC
Permalink
Hello,

It has come to my attention that a recent internal leak of Samsung's
kernel-space exFat driver implementation has been making the rounds around
the web. Nothing we can do about that, what is out, is out.

However a certain user "rxrz" went a bit too far with his actions:

https://github.com/rxrz/exfat-nofuse

He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.

When confronted with the issue; he came with the most unbelievable retort:
https://github.com/rxrz/exfat-nofuse/issues/5

This went as far as being posted on Phoronix as a legitimate driver:
http://phoronix.com/forums/showthread.php?81642-Native-Linux-Kernel-Module-Is-Out-For-Microsoft-exFAT

Clearly this is a breach of the most severe matter. Samsung has been made
aware of it, but there has been no response on the matter yet. I expect
nothing less than a DCMA takedown of the repository.
"It's a leaked code of a proprietary exfat driver, written by Samsung,
Inc. It works, you can use it. What else do you want, a signed paper
from your parents on whether you can or can not use it? I'm a
programmer, not a lawyer. You got the code, now decide what to do with
it, it's up to you."
What is wrong with such persons? This is insane.

Andrei F.
Arnt Karlsen
2013-07-22 13:46:01 UTC
Permalink
On Mon, 22 Jul 2013 02:24:38 +0200, Andrei wrote in message
Post by Andrei Frumusanu
Hello,
It has come to my attention that a recent internal leak of Samsung's
kernel-space exFat driver implementation has been making the rounds
around the web. Nothing we can do about that, what is out, is out.
https://github.com/rxrz/exfat-nofuse
..does GitHub know now?
Post by Andrei Frumusanu
He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.
When confronted with the issue; he came with the most unbelievable
retort: https://github.com/rxrz/exfat-nofuse/issues/5
http://phoronix.com/forums/showthread.php?81642-Native-Linux-Kernel-Module-Is-Out-For-Microsoft-exFAT
..does phoronix.com know now?
Post by Andrei Frumusanu
Clearly this is a breach of the most severe matter. Samsung has been
made aware of it, but there has been no response on the matter yet. I
expect nothing less than a DCMA takedown of the repository.
..they probably want to be thorough before they get tough, ;o)
e.g., does Samsung really license and sell etc publish a closed
source patent licensed kernel space idiot stunt FS, when peer
review can be had on free userspace software, for free, fixing
_everything_, world-wide?
Post by Andrei Frumusanu
"It's a leaked code of a proprietary exfat driver, written by
Samsung, Inc. It works, you can use it. What else do you want, a
signed paper from your parents on whether you can or can not use
it? I'm a programmer, not a lawyer. You got the code, now decide
what to do with it, it's up to you."
What is wrong with such persons? This is insane.
..could be some sort of "black flag" trick to try paint us or
Samsung, Google, gpl-violations.org or "Linux" as code thieves.
Etc. Have the Nazgul smoke 'em like tSCOG.
--
..med vennlig hilsen = with Kind Regards from Arnt Karlsen
...with a number of polar bear hunters in his ancestry...
Scenarios always come in sets of three:
best case, worst case, and just in case.
Armijn Hemel
2013-07-22 14:12:30 UTC
Permalink
Post by Arnt Karlsen
does Samsung really license and sell etc publish a closed
source patent licensed kernel space idiot stunt FS, when peer
review can be had on free userspace software, for free, fixing
_everything_, world-wide?
Yes. Probably because of patents.

armijn
--
------------------------------------------------------------------------
***@gpl-violations.org || http://www.gpl-violations.org/
------------------------------------------------------------------------
Philip Kendall
2013-07-22 14:06:44 UTC
Permalink
On Mon, Jul 22, 2013 at 02:24:38AM +0200, Andrei Frumusanu wrote:

[ ... ]
Post by Andrei Frumusanu
He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.
If this is really what happened (and I'm in no way denying it is!),
this is clearly a violation of Samsung's copyright, but I'm not
seeing the *GPL* violation here.

Or in other words: is it worth us discussing this here? This is
something for Samsung, rxrz and maybe GitHub to sort out.

Cheers,

Phil
--
Philip Kendall <***@shadowmagic.org.uk>
http://www.shadowmagic.org.uk/
Eric Appleman
2013-07-22 15:31:26 UTC
Permalink
Post by Andrei Frumusanu
Hello,
It has come to my attention that a recent internal leak of Samsung's
kernel-space exFat driver implementation has been making the rounds
around the web. Nothing we can do about that, what is out, is out.
https://github.com/rxrz/exfat-nofuse
He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.
When confronted with the issue; he came with the most unbelievable
retort: https://github.com/rxrz/exfat-nofuse/issues/5
http://phoronix.com/forums/showthread.php?81642-Native-Linux-Kernel-Module-Is-Out-For-Microsoft-exFAT
Clearly this is a breach of the most severe matter. Samsung has been
made aware of it, but there has been no response on the matter yet. I
expect nothing less than a DCMA takedown of the repository.
"It's a leaked code of a proprietary exfat driver, written by
Samsung, Inc. It works, you can use it. What else do you want, a
signed paper from your parents on whether you can or can not use it?
I'm a programmer, not a lawyer. You got the code, now decide what to
do with it, it's up to you."
What is wrong with such persons? This is insane.
Andrei F.
I first learned of this driver from Phoronix, a Linux tabloid with very
low journalistic integrity.
Jeremy L. Gaddis
2013-08-16 22:48:43 UTC
Permalink
"Conservancy Helps Samsung Resolve GPL Compliance Matter Amicably"

"Conservancy's GPL Compliance Project for Linux Developers is pleased to
announce its role in assisting Samsung in a recent public compliance
issue. The compliance issue was brought to Conservancy's attention when
source code of an exFAT filesystem driver for Linux was unintentionally
released via GitHub, and Conservancy later determined that similar code
appeared in binary form only (thus violating GPLv2§3) in a Samsung
Linux-based tablet. Samsung has made a source release available on their
Open Source Release Center website."

-- http://sfconservancy.org/news/2013/aug/16/exfat-samsung/
Post by Andrei Frumusanu
Hello,
It has come to my attention that a recent internal leak of Samsung's
kernel-space exFat driver implementation has been making the rounds
around the web. Nothing we can do about that, what is out, is out.
https://github.com/rxrz/exfat-nofuse
He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.
When confronted with the issue; he came with the most unbelievable
retort: https://github.com/rxrz/exfat-nofuse/issues/5
http://phoronix.com/forums/showthread.php?81642-Native-Linux-Kernel-Module-Is-Out-For-Microsoft-exFAT
Clearly this is a breach of the most severe matter. Samsung has been
made aware of it, but there has been no response on the matter yet. I
expect nothing less than a DCMA takedown of the repository.
"It's a leaked code of a proprietary exfat driver, written by Samsung,
Inc. It works, you can use it. What else do you want, a signed paper
from your parents on whether you can or can not use it? I'm a
programmer, not a lawyer. You got the code, now decide what to do with
it, it's up to you."
What is wrong with such persons? This is insane.
Andrei F.
TJ
2013-08-17 10:02:22 UTC
Permalink
Post by Jeremy L. Gaddis
"Conservancy Helps Samsung Resolve GPL Compliance Matter Amicably"
"Conservancy's GPL Compliance Project for Linux Developers is pleased to announce its role in assisting Samsung in a recent public compliance issue. The compliance issue was brought to
Conservancy's attention when source code of an exFAT filesystem driver for Linux was unintentionally released via GitHub, and Conservancy later determined that similar code appeared in binary
form only (thus violating GPLv2§3) in a Samsung Linux-based tablet. Samsung has made a source release available on their Open Source Release Center website."
-- http://sfconservancy.org/news/2013/aug/16/exfat-samsung/
Wonderful - as much as anything else it authoritatively puts an end to the speculation, rumour, and innuendo surrounding the exFAT source.
TJ
2013-08-17 10:21:01 UTC
Permalink
Post by Jeremy L. Gaddis
"Conservancy Helps Samsung Resolve GPL Compliance Matter Amicably"
Thank-you for this Jeremy.

I went to the Samsung open-source site but, after clicking on one of the ZIP file downloads, was unable to download the source without first agreeing to an additional Samsung legal agreement. I
tried choosing "Disgree" and continuing but the site doesn't allow that.

As I understood it the GPL precludes distributors from imposing additional terms; surely that includes imposing a legal agreement in order to obtain the source?


From: http://opensource.samsung.com/reception/receptionSub.do?method=modal

"Please take a moment to read the legal agreement. If you accept the terms below, please click ‘Agree’.

Samsung Electronics, Co. Ltd. ("Samsung") is pleased to make available to you the various source codes ("Source Codes") for download from this website ("Download Service") at no charge. By using the
Source Code and/or the Download Service, you expressly assume all risk and liability associated with and/or caused by the same and complying with all applicable user agreements that accompany each
Source Code. To the extent any of the Source Codes are licensed under public licenses such as the GNU General Public License and/or GNU Lesser General Public License, such licenses can be found in
each of the files containing the Source Codes.

Samsung provides the Source Codes and this Downloading Service "as is" without representation or warranty of any kind and all such warranties, express or implied, are hereby disclaimed, including,
without limitation, warranties of merchantability, fitness for a particular purpose accuracy, completeness, currency, availability, title, or non-infringement. Samsung will not be liable for any
damages of any kind arising from the use of the Source Code and this Downloading Service, including, but not limited to direct, indirect, incidental, punitive, and support or assistance with respect
to the Source Codes or the Download Service.

This downloading service shall not affect not extend any warranty or disclaimer which Samsung makes in each of Samsung’s products in which the Source Code (or object/executable code based thereon)
is incorporated."
Armijn Hemel
2013-08-17 19:40:14 UTC
Permalink
Post by TJ
I went to the Samsung open-source site but, after clicking on one of the ZIP file downloads, was unable to download the source without first agreeing to an additional Samsung legal agreement. I
tried choosing "Disgree" and continuing but the site doesn't allow that.
Then get it from Gitorious where Bradley put it:

https://gitorious.org/binbang/exfat-linux

No extra agreements there...

armijn
--
------------------------------------------------------------------------
***@gpl-violations.org || http://www.gpl-violations.org/
------------------------------------------------------------------------
Bradley M. Kuhn
2013-08-18 01:45:42 UTC
Permalink
Post by Armijn Hemel
I... was unable to download the source without first agreeing to an
additional Samsung legal agreement.
..
Post by Armijn Hemel
As I understood it the GPL precludes distributors from imposing
additional terms; surely that includes imposing a legal agreement in
order to obtain the source?
You're correct that imposing additional restrictions on distribution is
prohibited by GPLv2§6. However, I didn't see anything in that
download agreement that imposed an additional restriction. Which
part do you think imposed an additional restriction?
Post by Armijn Hemel
https://gitorious.org/binbang/exfat-linux
You can feel free to take distribution of that code from me, under
GPLv2, via the aforementioned repository. :)
--
-- bkuhn
TJ
2013-08-18 23:55:03 UTC
Permalink
Post by Bradley M. Kuhn
Post by Armijn Hemel
I... was unable to download the source without first agreeing to an
additional Samsung legal agreement.
..
Post by Armijn Hemel
As I understood it the GPL precludes distributors from imposing
additional terms; surely that includes imposing a legal agreement in
order to obtain the source?
You're correct that imposing additional restrictions on distribution is
prohibited by GPLv2§6. However, I didn't see anything in that
download agreement that imposed an additional restriction. Which
part do you think imposed an additional restriction?
"...and all such warranties, express or implied, are hereby disclaimed, including,
without limitation, warranties of merchantability, fitness for a particular purpose accuracy, completeness.."

My eyes caught on:

"warranties" ... "disclaimed, including" ... "accuracy" ... "completeness"

which made me immediately wonder what a user's recourse would be if the download(s) from the Samsung open source centre are incomplete. It would seem to leave users without a remedy to obtain the
"complete corresponding machine-readable source code" if it did prove to be incomplete.
Post by Bradley M. Kuhn
Post by Armijn Hemel
https://gitorious.org/binbang/exfat-linux
You can feel free to take distribution of that code from me, under
GPLv2, via the aforementioned repository. :)
Indeed, after Armijn made me aware of it that is what I did, but there was no mention of that repository in the SFLC press release.
Bradley M. Kuhn
2013-08-19 13:28:31 UTC
Permalink
Post by TJ
Post by Bradley M. Kuhn
Post by TJ
As I understood it the GPL precludes distributors from imposing
additional terms; surely that includes imposing a legal agreement in
order to obtain the source?
You're correct that imposing additional restrictions on distribution is
prohibited by GPLv2§6. However, I didn't see anything in that
download agreement that imposed an additional restriction. Which
part do you think imposed an additional restriction?
"...and all such warranties, express or implied, are hereby
disclaimed, including, without limitation, warranties of
merchantability, fitness for a particular purpose accuracy,
completeness.."
The above reads to me as compatible with the terms GPLv2§11 and
GPLv2§12. Do you find some contradiction between the text quoted above
and those sections?

(Note that GPLv3§7(a) makes this sort of compatibility analysis of
warranty disclaimers even easier.)
Post by TJ
which made me immediately wonder what a user's recourse would be if
the download(s) from the Samsung open source centre are incomplete. It
would seem to leave users without a remedy to obtain the
It seems to me that it's still copyright infringement if they don't
comply with copyright, regardless of warranties being disclaimed. Are
you aware of some regulation or law in some jurisdiction which would
cause that not to be true?

Also, in the alternative, note that you don't get the binaries from that
site, only the sources. Thus, distribution of the binaries is not
covered by that specific warranty disclaimer anyway (although I suspect
there's a similar warranty disclaimer with Samsung's products).
Post by TJ
Indeed, after Armijn made me aware of it that is what I did, but there
was no mention of that repository in the SFLC press release.
I am not aware of any SFLC press release on this matter, nor of any work
done by SFLC on this matter. To my knowledge and belief, SFLC is not
involved in enforcement of GPL for Linux, BusyBox, nor Samba.

Perhaps you mean the press release by Conservancy, which is a different
organization entirely from SFLC (although SFLC used to be Conservancy's
lawyers, a long time ago)? Perhaps the fact that SFLC used to be, long
ago, Conservancy's law firm is what has you confused here?

Anyway, I want to note clearly that the git repository isn't mentioned
in Conservancy's press statement because I did that work on my own. I'm
personally distributing that software, not Conservancy. (Not that
Conservancy wouldn't do so, it's just that Conservancy didn't, *I* did.)
Anyway, I see no reason not to download the sources from Samsung's
website, as explained above.

I'd been planning to make a personal blog post about the git repository
and other personal thoughts on this matter, but I haven't gotten the
tuits yet.
--
-- bkuhn
TJ
2013-08-19 15:12:50 UTC
Permalink
"...and all such warranties, express or implied, are hereby disclaimed, including, without limitation, warranties of merchantability, fitness for a particular purpose accuracy, completeness.."
The above reads to me as compatible with the terms GPLv2§11 and GPLv2§12. Do you find some contradiction between the text quoted above and those sections?
(Note that GPLv3§7(a) makes this sort of compatibility analysis of warranty disclaimers even easier.)
My question was more from a user's perspective as to what remedies might be available.

The crux of what I was thinking about when I wrote the original question was:

1. The majority of the code in exFAT is copyright Samsung
2. Samsung license the code under GPLv2
2a. GPLv2§2 and §3, I assume, doesn't apply to Samsung since exFAT is not a modification nor a copy, it is an original work by Samsung.
3. Samsung imposes a supplementary contractual agreement on a user wishing to download the source code
4. That contract includes disclaimers as to accuracy and completeness
5. What would be the user's recourse if the source-code proved to be incomplete

In the devices exFAT is combined with the rest of the Linux kernel so for the binary distribution Samsung is both a distributor and an original author.

I think many of us regular developers find this area of the GPL difficult to understand because, it seems to me at least, there are two quite different readings of the license depending on whether
you receive the work and source directly from the copyright holder, or from someone who is copying and distributing.

Your contributions recently have helped me, at least, to get a better understanding. Thank-you.
I am not aware of any SFLC press release on this matter, nor of any work
done by SFLC on this matter.

Of course - it isn't hard when you're not familiar with the intricacies of the organisational acronyms to get SFC (Software Freedom Conservancy) and SFLC (Software Freedom Law Centre) confused!
Bradley M. Kuhn
2013-08-20 18:52:40 UTC
Permalink
Post by TJ
2a. GPLv2§2 and §3, I assume, doesn't apply to Samsung since exFAT is
not a modification nor a copy, it is an original work by Samsung.
Binaries are a modification of source. The module is derivative
of Linux for multiple reasons. GPLv2§2 and §3 both apply,
AFAICT, in all the real-world cases we're looking at regarding this
specific code.


Note, again, that users who bought the devices that include the
no-source-available exfat modules received distribution of binaries, and
have a right to source code.
Post by TJ
I think many of us regular developers find this area of the GPL
difficult to understand because, it seems to me at least, there are
two quite different readings of the license depending on whether you
receive the work and source directly from the copyright holder, or
from someone who is copying and distributing.
The license indeed works in different ways depending on whether you
receive source or not, and it's certainly true that some key provisions
of GPLv3 operates differently with regard to the copyright holder vs. a
redistributor (or, re-conveyor/re-propagator, to use GPLv3 terminology).

I admit that Free Software licensing is more complicated than it seems
on the surface.
Post by TJ
Your contributions recently have helped me, at least, to get a better
understanding. Thank-you.
You're welcome; I'm glad it's helpful.
Post by TJ
Of course - it isn't hard when you're not familiar with the
intricacies of the organisational acronyms to get SFC (Software
Freedom Conservancy) and SFLC (Software Freedom Law Centre) confused!
Of course. FWIW, I tend to call "Software Freedom Conservancy" by the
name: "Conservancy". "software freedom" is a generic term, after all,
like "Free Software". I sometimes call "FSF" "the Foundation", but that
has a semi-unintentional Asmiov references, so I don't do it often. :)
--
-- bkuhn
Thomas Charron
2013-08-20 22:08:13 UTC
Permalink
Post by Bradley M. Kuhn
Post by TJ
2a. GPLv2§2 and §3, I assume, doesn't apply to Samsung since exFAT is
not a modification nor a copy, it is an original work by Samsung.
Binaries are a modification of source. The module is derivative
of Linux for multiple reasons. GPLv2§2 and §3 both apply,
AFAICT, in all the real-world cases we're looking at regarding this
specific code.
A binaray is *not* a modification of the source, no more then a meal is
a 'modification' of a recipe.
Post by Bradley M. Kuhn
Note, again, that users who bought the devices that include the
no-source-available exfat modules received distribution of binaries, and
have a right to source code.
No, they do not. The code was released out of the kindness and as a
contribution to the community as a whole. It is still the property of
Samsung to do with what they please.

Samsung had written the code under a proprietary license. It was found
after a developer had improperly pushed it publically, and then was adopted
by other developers who put the GPL label on it *WITHOUT THE PROPER
PERMISSIONS OF THE OWNERS OF THE CODE*.

Samsung has decided to open source the implementation, which now allows
developers who ended up using the stolen code to now have a rightful
license to it.

-- Thomas
Jonathan Wilson
2013-08-21 02:47:17 UTC
Permalink
Post by Thomas Charron
No, they do not. The code was released out of the kindness and as a
contribution to the community as a whole. It is still the property of
Samsung to do with what they please.
The Samsung code contains statements like 'Some of the source code in this
file came from "linux/fs/fat/misc.c"' and 'Some of the source code in this
file came from "linux/fs/fat/file.c","linux/fs/fat/inode.c" and
"linux/fs/fat/misc.c"' and 'Some of the source code in this file came from
"linux/fs/fat/fat.h"'
I myself haven't done a comparison of the Samsung source code and the
kernel source but I have read of others who have done so and found the
statements in the Samsung code to be true (i.e. that the Samsung code IS
using code from other kernel source files).

Therefore the Samsung code is a derived work of kernel code not written by
Samsung and they are required to comply with the terms of GPLv2 as applied
to misc.c, file.c, inode.c, fat.h and any other source files they may have
copied from.
Bradley M. Kuhn
2013-08-21 15:19:57 UTC
Permalink
Post by Jonathan Wilson
I myself haven't done a comparison of the Samsung source code and the
kernel source but I have read of others who have done so and found the
statements in the Samsung code to be true (i.e. that the Samsung code IS
using code from other kernel source files).
I don't know if anyone has done this analysis on the new release (some people
did it on the "leaked" version not legitimately released from Samsung). But
Post by Jonathan Wilson
The Samsung code contains statements like 'Some of the source code in this
file came from "linux/fs/fat/misc.c"' and 'Some of the source code in this
file came from "linux/fs/fat/file.c","linux/fs/fat/inode.c" and
"linux/fs/fat/misc.c"' and 'Some of the source code in this file came from
"linux/fs/fat/fat.h"'
So we know that's one reason this is a derivative work of Linux.
Post by Jonathan Wilson
Therefore the Samsung code is a derived work of kernel code not
written by Samsung and they are required to comply with the terms of
GPLv2 as applied to misc.c, file.c, inode.c, fat.h and any other
source files they may have copied from.
They're were required to comply with GPLv2 for other reasons too, most
notably that they shipped a whole distribution of the kernel named Linux with
the binary exfat modules included. The whole work there was the
Linux+exfat_modules, and was a derivative work of all of Linux.

-- bkuhn
Bradley M. Kuhn
2013-08-21 18:39:05 UTC
Permalink
A binary is *not* a modification of the source, no more then a meal
is a 'modification' of a recipe.
Actually, in the USA it is. Admittedly, it may be handled differently in
different copyright jurisdictions.

Regardless, one needs permission under copyright to prepare and distribute
a binary work if that work builds, in part or in whole, from source code
copyrighted by someone else. That's true in pretty much all Berne Convention
countries AFAIK. (Can anyone identify a Berne Convention country where
this *isn't* true?)
Post by Bradley M. Kuhn
Note, again, that users who bought the devices that include the
no-source-available exfat modules received distribution of binaries,
and have a right to source code.
No, they do not.
Multiple investigators, including me personally, have confirmed the facts I
stated above and elsewhere in this thread.

I don't think Samsung distributed those modules with every single tablet,
though, only specific versions. If you have a Samsung tablet, it may or may
not have the exfat code in it.
The code was released out of the kindness and as a contribution to the
community as a whole.
I don't dispute that Samsung has handled this problem with kindness and
cooperation, but they *were* resolving a GPL violation.
It was found after a developer had improperly pushed it publically,
I also think rxrz's actions in this matter were deplorable and incorrect,
mostly because rxrz's actions made it *more difficult* than it would have
been to resolve the matter to a good resolution.

But please don't conflate rxrz's bad actions with the GPL violation
underneath. Samsung distributed the exfat Linux module with some of their
Linux-based tablet products. That's a violation of GPL and infringement of
copyright. Samsung since corrected that violation with their source release,
and, in my experience, they did so in a way that was kind, considerate, and
cooperative with the Free Software community.

rxrz thought it would help to "leak" the code. I agree with you that it
didn't help, and actually made things worse and more confusing. But that's a
separate issue.

-- bkuhn
Ralph Corderoy
2013-08-25 11:08:53 UTC
Permalink
Hi Bradley,
Post by Bradley M. Kuhn
I also think rxrz's actions in this matter were deplorable and
incorrect, mostly because rxrz's actions made it *more difficult* than
it would have been to resolve the matter to a good resolution.
It's annoying that rxrz slapped a GNU GPL licence on code that wasn't
his. Kind of sullies the GPL by association. The repo still sits at
github, along with its many fork-descendants. Github, understanding
what he's done, still say that they will only remove them at the request
of the copyright holder, and requested that they be contacted to file a
DMCA take-down; http://help.github.com/dmca-takedown. Would there be
someone at Samsung interested in doing this to tidy things up rather
than leave a not-properly-released "GPL'd" version detracting from the
pucker source? Presumably github zap all the descendants given the root
repo.

Cheers, Ralph.
Bruce Perens
2013-08-21 19:31:36 UTC
Permalink
Post by Thomas Charron
A binaray is *not* a modification of the source, no more then a
meal is a 'modification' of a recipe.
This analogy doesn't really work. A binary created using a compiler, and
potentially an assembler, linker, and other tools is a textual
translation of the source code. The artistic elements of the original
persist but in a different form. And of course we know all about how
that works legally due to CAI v. Altai. It is more than fair to call any
translation a modification of the original work. It might miss being a
/derivative/ work because there is no additional /art/ added by the
mechanical process of translation, and thus no new copyrightable elements.

In contrast, creating food from a recipie is not a textual translation,
it is the execution of a set of instructions regarding how to combine
materials.

But your entire argument is moot anyway, since the work appears to have
copied liberally from the Linux kernel and we have not established that
every element it copied was exclusively functional rather than expressive.
Eric Appleman
2013-08-17 18:08:48 UTC
Permalink
If you want to avoid the agreement for the sake of research, there's a sequence of exfat commits that go version by version on github.

https://github.com/AndreiLux/Perseus-UNIVERSAL5410/commits/perseus

Just scroll down a hair and you'll see them.

- Eric

-----Original Message-----
From: legal-***@lists.gpl-violations.org [mailto:legal-***@lists.gpl-violations.org] On Behalf Of TJ
Sent: Saturday, August 17, 2013 6:21 AM
To: Jeremy L. Gaddis
Cc: ***@lists.gpl-violations.org
Subject: Re: "Conservancy Helps Samsung Resolve GPL Compliance Matter Amicably"
Post by Jeremy L. Gaddis
"Conservancy Helps Samsung Resolve GPL Compliance Matter Amicably"
Thank-you for this Jeremy.

I went to the Samsung open-source site but, after clicking on one of the ZIP file downloads, was unable to download the source without first agreeing to an additional Samsung legal agreement. I tried choosing "Disgree" and continuing but the site doesn't allow that.

As I understood it the GPL precludes distributors from imposing additional terms; surely that includes imposing a legal agreement in order to obtain the source?


From: http://opensource.samsung.com/reception/receptionSub.do?method=modal

"Please take a moment to read the legal agreement. If you accept the terms below, please click ‘Agree’.

Samsung Electronics, Co. Ltd. ("Samsung") is pleased to make available to you the various source codes ("Source Codes") for download from this website ("Download Service") at no charge. By using the Source Code and/or the Download Service, you expressly assume all risk and liability associated with and/or caused by the same and complying with all applicable user agreements that accompany each Source Code. To the extent any of the Source Codes are licensed under public licenses such as the GNU General Public License and/or GNU Lesser General Public License, such licenses can be found in each of the files containing the Source Codes.

Samsung provides the Source Codes and this Downloading Service "as is" without representation or warranty of any kind and all such warranties, express or implied, are hereby disclaimed, including, without limitation, warranties of merchantability, fitness for a particular purpose accuracy, completeness, currency, availability, title, or non-infringement. Samsung will not be liable for any damages of any kind arising from the use of the Source Code and this Downloading Service, including, but not limited to direct, indirect, incidental, punitive, and support or assistance with respect to the Source Codes or the Download Service.

This downloading service shall not affect not extend any warranty or disclaimer which Samsung makes in each of Samsung’s products in which the Source Code (or object/executable code based thereon) is incorporated."
Jilayne Lovejoy
2013-08-20 23:12:25 UTC
Permalink
On 8/17/13 4:21 AM, "TJ" <gpl-***@iam.tj<mailto:gpl-***@iam.tj>> wrote:


As I understood it the GPL precludes distributors from imposing
additional terms; surely that includes imposing a legal agreement in
order to obtain the source?

That is correct in so far as section 6 of GPLv2, which states, ". . . You
may not impose any further restrictions on the recipients' exercise of the
rights granted herein. . ." but I don't believe the Agreement you've
included below imposes any further restrictions - see comments.


From: http://opensource.samsung.com/reception/receptionSub.do?method=modal

"Please take a moment to read the legal agreement. If you accept the
terms below, please click ŒAgree¹.

Samsung Electronics, Co. Ltd. ("Samsung") is pleased to make available to
you the various source codes ("Source Codes") for download from this
website ("Download Service") at no charge. By using the
Source Code and/or the Download Service, you expressly assume all risk
and liability associated with and/or caused by the same and complying
with all applicable user agreements that accompany each
Source Code.

This, to me, is like restating the disclaimer that the GPL already
includes. I'd also consider this a "gentle reminder" that there may be
compliance actions associated with the licenses that the source that is
being provided.

To the extent any of the Source Codes are licensed under public licenses
such as the GNU General Public License and/or GNU Lesser General Public
License, such licenses can be found in
each of the files containing the Source Codes.

No restrictions here, I refer to this as an "FYI" clause, as in, I don't
really have to tell you this, but I'm going to point it out to make sure
you notice.


Samsung provides the Source Codes and this Downloading Service "as is"
without representation or warranty of any kind and all such warranties,
express or implied, are hereby disclaimed, including,
without limitation, warranties of merchantability, fitness for a
particular purpose accuracy, completeness, currency, availability, title,
or non-infringement. Samsung will not be liable for any
damages of any kind arising from the use of the Source Code and this
Downloading Service, including, but not limited to direct, indirect,
incidental, punitive, and support or assistance with respect
to the Source Codes or the Download Service.

Again, this is reaffirming the "as is"/no warranty nature of the software
being provided, which is what GPL states to begin with.

This downloading service shall not affect not extend any warranty or
disclaimer which Samsung makes in each of Samsung¹s products in which the
Source Code (or object/executable code based thereon)
is incorporated."


Samsung is certainly allowed to choose when or how any warranty they offer
to customers directly expire or apply. Again, I don't see this as
imposing any further restriction on the GPL code being provided.

While it may seem uncomfortable to have to "agree" to any kind of an
agreement to get to the source code, so long as the terms that one must
agree to do not violate section 6, then I don't see that as being a
conflict. Of course, I'd be curious to hear if Bradley, Armijn, and
Harald agree :)

Cheers,
Jilayne (first time posting!)
Vinz Focker
2013-08-18 12:48:46 UTC
Permalink
Post by Andrei Frumusanu
What is wrong with such persons? This is insane.
What's wrong with Samsung I'd rather ask ... that mail has backfired,
hasn't it ?
However, glad that it got resolved.
Post by Andrei Frumusanu
Hello,
It has come to my attention that a recent internal leak of Samsung's
kernel-space exFat driver implementation has been making the rounds around
the web. Nothing we can do about that, what is out, is out.
https://github.com/rxrz/exfat-**nofuse<https://github.com/rxrz/exfat-nofuse>
He basically has stripped the original code clean of all Samsung
proprietary license marks and threw GPL tags on it.
https://github.com/rxrz/exfat-**nofuse/issues/5<https://github.com/rxrz/exfat-nofuse/issues/5>
http://phoronix.com/forums/**showthread.php?81642-Native-**
Linux-Kernel-Module-Is-Out-**For-Microsoft-exFAT<http://phoronix.com/forums/showthread.php?81642-Native-Linux-Kernel-Module-Is-Out-For-Microsoft-exFAT>
Clearly this is a breach of the most severe matter. Samsung has been made
aware of it, but there has been no response on the matter yet. I expect
nothing less than a DCMA takedown of the repository.
"It's a leaked code of a proprietary exfat driver, written by Samsung,
Inc. It works, you can use it. What else do you want, a signed paper from
your parents on whether you can or can not use it? I'm a programmer, not a
lawyer. You got the code, now decide what to do with it, it's up to you."
What is wrong with such persons? This is insane.
Andrei F.
Jilayne Lovejoy
2013-08-28 18:19:26 UTC
Permalink
Post by Ralph Corderoy
Post by Bradley M. Kuhn
I also think rxrz's actions in this matter were deplorable and
incorrect, mostly because rxrz's actions made it *more difficult* than
it would have been to resolve the matter to a good resolution.
It's annoying that rxrz slapped a GNU GPL licence on code that wasn't
his. Kind of sullies the GPL by association.
I agree, Ralph, although, I'm not sure it so much sullies the reputation
of the GPL specifically, so much as Free Software and developers
generally, by creating an impression of a rogue developer with no regard
for the very law that provides the framework to protect free software and
effect such licenses as GPL. Of course, not all free software developers
are heroes either, but such a public and negative impression can often get
imputed broadly. I find it especially disappointing that Wired magazine's
piece called this person "Robin Hood" and then allowed the person to be
quoted in Wired's article without verifying identity. I would
characterize the former as sensationalism in the wrong direction and the
latter as simply poor journalism. *sigh*
Post by Ralph Corderoy
The repo still sits at
github, along with its many fork-descendants. Github, understanding
what he's done, still say that they will only remove them at the request
of the copyright holder, and requested that they be contacted to file a
DMCA take-down; http://help.github.com/dmca-takedown. Would there be
someone at Samsung interested in doing this to tidy things up rather
than leave a not-properly-released "GPL'd" version detracting from the
pucker source? Presumably github zap all the descendants given the root
repo.
I can't blame GitHub for taking this stance - it would not be wise of them
to take down something without going through the proper procedures under
the DMCA, as that could set a dangerous precedent and impression for their
customers (even if this seems like a clear cut case of wrong-doing).
Remains to be seen whether Samsung will file the request, so we shall
"stay tuned"!

Jilayne
Bradley M. Kuhn
2013-08-30 19:32:09 UTC
Permalink
I'm not sure it so much sullies the reputation of the GPL
specifically, so much as Free Software and developers generally, by
creating an impression of a rogue developer with no regard for the
very law that provides the framework to protect free software and
effect such licenses as GPL.
It seems quite far to go to say that this situation "sullies the
reputation of Free Software and developers generally". Occasionally, a
lawyer and/or judge is corrupt and does something wrong. Would you say
that "sullies the reputation of lawyers and judges generally"?
--
-- bkuhn
Bruce Perens
2013-08-30 21:24:00 UTC
Permalink
Would you say that "sullies the reputation of lawyers and judges
generally"?
Well, Free Software Developers have higher standards. :-)
Ralph Corderoy
2013-08-30 21:42:33 UTC
Permalink
Hi Bradley,
Occasionally, a lawyer and/or judge is corrupt and does something
wrong. Would you say that "sullies the reputation of lawyers and
judges generally"?
Unfairly perhaps, but yes. Have all those that work in the banking
industry be sullied by the actions of a small percentage? A matter of
degree, yes, but r[xz]r[xz]'s actions have gained publicity and his
responses to criticism may have encouraged others as repercussions are
nil?

Cheers, Ralph.

Loading...