LinuxTV ported to Windows

LinuxTV ported to Windows

[Abylay Ospan]

Hello,

We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista, 
XP, win7 tested). Results available under GPL and can be checkout from 
git repository:
https://github.com/netup/netup-dvb-s2-ci-dual

Binary builds (ready to install) available in build directory. Currently 
NetUP Dual DVB-S2 CI card supported ( 
http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).

Driver based on Microsoft BDA standard, but some features (DiSEqC, CI) 
supported by custom API, for more details see netup_bda_api.h file.

Any comments, suggestions are welcome.

-- 
Abylai Ospan<aospan@netup.ru>
NetUP Inc.

Re: LinuxTV ported to Windows

[Devin Heitmueller]

2011/11/30 Abylay Ospan <aospan@netup.ru>:
> Hello,
>
> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista, XP,
> win7 tested). Results available under GPL and can be checkout from git
> repository:
> https://github.com/netup/netup-dvb-s2-ci-dual
>
> Binary builds (ready to install) available in build directory. Currently
> NetUP Dual DVB-S2 CI card supported (
> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
>
> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI)
> supported by custom API, for more details see netup_bda_api.h file.
>
> Any comments, suggestions are welcome.
>
> --
> Abylai Ospan<aospan@netup.ru>
> NetUP Inc.

Am I the only one who thinks this is a legally ambigious grey area?
Seems like this could be a violation of the GPL as the driver code in
question links against a proprietary kernel.

I don't want to start a flame war, but I don't see how this is legal.
And you could definitely question whether it goes against the
intentions of the original authors to see their GPL driver code being
used in non-free operating systems.

Devin

-- 
Devin J. Heitmueller - Kernel Labs
http://www.kernellabs.com

Re: LinuxTV ported to Windows

[Andreas Oberritter]

On 30.11.2011 18:23, Devin Heitmueller wrote:
> 2011/11/30 Abylay Ospan <aospan@netup.ru>:
>> Hello,
>>
>> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista, XP,
>> win7 tested). Results available under GPL and can be checkout from git
>> repository:
>> https://github.com/netup/netup-dvb-s2-ci-dual

That's nice to hear, Abylay!

>> Binary builds (ready to install) available in build directory. Currently
>> NetUP Dual DVB-S2 CI card supported (
>> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
>>
>> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI)
>> supported by custom API, for more details see netup_bda_api.h file.
>>
>> Any comments, suggestions are welcome.
>>
>> --
>> Abylai Ospan<aospan@netup.ru>
>> NetUP Inc.
> 
> Am I the only one who thinks this is a legally ambigious grey area?
> Seems like this could be a violation of the GPL as the driver code in
> question links against a proprietary kernel.

Devin, please! Are you implying that the windows kernel becomes a
derived work of the driver, or that it's generally impossible to publish
windows drivers under the terms of the GPL?

> I don't want to start a flame war, but I don't see how this is legal.
> And you could definitely question whether it goes against the
> intentions of the original authors to see their GPL driver code being
> used in non-free operating systems.

The GPL doesn't cover such intentions.

Regards,
Andreas

P.S.: "The licenses for most software are designed to take away your
freedom to share and change it.  By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software--to make sure the software is free for all its users."

Re: LinuxTV ported to Windows

[Walter Van Eetvelt]

Nice!

How is the CI implementation?  Can both CI's be used by both tuners?  Or
is one CI bound to one tuner?  

Walter

On Wed, 30 Nov 2011 19:39:34 +0300, Abylay Ospan <aospan@netup.ru> wrote:
> Hello,
> 
> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista,

> XP, win7 tested). Results available under GPL and can be checkout from 
> git repository:
> https://github.com/netup/netup-dvb-s2-ci-dual
> 
> Binary builds (ready to install) available in build directory. Currently

> NetUP Dual DVB-S2 CI card supported ( 
> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
> 
> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI) 
> supported by custom API, for more details see netup_bda_api.h file.
> 
> Any comments, suggestions are welcome.

Re: LinuxTV ported to Windows

[Devin Heitmueller]

On Wed, Nov 30, 2011 at 1:02 PM, Andreas Oberritter <obi@linuxtv.org> wrote:
>> Am I the only one who thinks this is a legally ambigious grey area?
>> Seems like this could be a violation of the GPL as the driver code in
>> question links against a proprietary kernel.
>
> Devin, please! Are you implying that the windows kernel becomes a
> derived work of the driver, or that it's generally impossible to publish
> windows drivers under the terms of the GPL?

The simple answer is that "I don't know".  I'm not a lawyer (and as
far as I know, neither are you).  Nor have I researched the topic to
significant lengths.  That said though, whether it was the intention
of either copyright holder it's entirely possible that the two
software licenses are simply incompatible.  For example, while both
the Apache group and the FSF never really intended to prevent each
others' software from being linked against each other, the net effect
is still that you cannot redistribute such software together since the
Apache license is incompatible with the GPL.

>> I don't want to start a flame war, but I don't see how this is legal.
>> And you could definitely question whether it goes against the
>> intentions of the original authors to see their GPL driver code being
>> used in non-free operating systems.
>
> The GPL doesn't cover such intentions.

This isn't necessarily true.  Anybody who has written a library and
released it under the GPL instead of the LGPL has made a conscious
decision that the library is only to be used by software that is GPL
compatible.  By their actions they have inherently forbidden it's use
by non-free software.  You could certainly make the same argument
about a driver -- that they authors intent was to ensure that it only
be linked against other free software.

All this said, I don't really have a position one way or the other
(I'm not a copyright holder on the drivers in question).  But this
issue doesn't seem as obvious as you would make it sound.

Devin

-- 
Devin J. Heitmueller - Kernel Labs
http://www.kernellabs.com

Re: LinuxTV ported to Windows

[Andreas Oberritter]

On 30.11.2011 20:33, Devin Heitmueller wrote:
> On Wed, Nov 30, 2011 at 1:02 PM, Andreas Oberritter <obi@linuxtv.org> wrote:
>>> Am I the only one who thinks this is a legally ambigious grey area?
>>> Seems like this could be a violation of the GPL as the driver code in
>>> question links against a proprietary kernel.
>>
>> Devin, please! Are you implying that the windows kernel becomes a
>> derived work of the driver, or that it's generally impossible to publish
>> windows drivers under the terms of the GPL?
> 
> The simple answer is that "I don't know".  I'm not a lawyer (and as
> far as I know, neither are you).  Nor have I researched the topic to
> significant lengths.  That said though, whether it was the intention
> of either copyright holder it's entirely possible that the two
> software licenses are simply incompatible.  For example, while both
> the Apache group and the FSF never really intended to prevent each
> others' software from being linked against each other, the net effect
> is still that you cannot redistribute such software together since the
> Apache license is incompatible with the GPL.

Neither is Abylay distributing Windows together with this driver, nor is
this driver a library Windows links against, i.e. Windows is able to run
with this driver removed.

>>> I don't want to start a flame war, but I don't see how this is legal.
>>> And you could definitely question whether it goes against the
>>> intentions of the original authors to see their GPL driver code being
>>> used in non-free operating systems.
>>
>> The GPL doesn't cover such intentions.
> 
> This isn't necessarily true.  Anybody who has written a library and
> released it under the GPL instead of the LGPL has made a conscious
> decision that the library is only to be used by software that is GPL
> compatible.  By their actions they have inherently forbidden it's use
> by non-free software.  You could certainly make the same argument
> about a driver -- that they authors intent was to ensure that it only
> be linked against other free software.

That's something completely different than "being used in non-free
operating systems" and not necessarily comparable to a driver, which
implements a well-defined interface.

The point I'm trying to make: Someone made a presumably nice open source
port to a new platform and the first thing you're doing is to play the
GPL-has-been-violated-card, even though you're admitting that you don't
know whether any right is being violated or not. Please don't do that.
It's not very encouraging to someone who just announced free software.

Regards,
Andreas

Re: LinuxTV ported to Windows

[Devin Heitmueller]

On Wed, Nov 30, 2011 at 2:58 PM, Andreas Oberritter <obi@linuxtv.org> wrote:
> The point I'm trying to make: Someone made a presumably nice open source
> port to a new platform and the first thing you're doing is to play the
> GPL-has-been-violated-card, even though you're admitting that you don't
> know whether any right is being violated or not. Please don't do that.
> It's not very encouraging to someone who just announced free software.

To be clear, at no point did I say the GPL has been violated.  I am
*asking* if others think this represents a GPL violation since at
first glance it appears that this software very well might.

I don't really have a problem with GPL drivers running on Windows.
But if I were the original author of one of the drivers Abylay has
ported, I might think differently.

Devin

-- 
Devin J. Heitmueller - Kernel Labs
http://www.kernellabs.com

Re: LinuxTV ported to Windows

[Abylay Ospan]

Hi Walter,

On 30.11.2011 21:46, Walter Van Eetvelt wrote:
> Nice!
>
> How is the CI implementation?
it's ok. Working fine under windows including MMI.
Professional CAM's (with multi-PID descramble) are tested.

> Can both CI's be used by both tuners?  Or
> is one CI bound to one tuner?
First CI slot assigned to first tuner/demod and second CI slot assigned 
for second tuner/demod by hardware.
You can't share CI slots between tuners.

>
> Walter
>
> On Wed, 30 Nov 2011 19:39:34 +0300, Abylay Ospan<aospan@netup.ru>  wrote:
>> Hello,
>>
>> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista,
>> XP, win7 tested). Results available under GPL and can be checkout from
>> git repository:
>> https://github.com/netup/netup-dvb-s2-ci-dual
>>
>> Binary builds (ready to install) available in build directory. Currently
>> NetUP Dual DVB-S2 CI card supported (
>> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
>>
>> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI)
>> supported by custom API, for more details see netup_bda_api.h file.
>>
>> Any comments, suggestions are welcome.

-- 
Abylai Ospan<aospan@netup.ru>
NetUP Inc.

Re: LinuxTV ported to Windows

[Abylay Ospan]

Hi Devin,

Thanks for this idea. Need to investigate.
Currently we've made porting and released the results without any 
license violations in mind ...

On 30.11.2011 20:23, Devin Heitmueller wrote:
> 2011/11/30 Abylay Ospan<aospan@netup.ru>:
>> Hello,
>>
>> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista, XP,
>> win7 tested). Results available under GPL and can be checkout from git
>> repository:
>> https://github.com/netup/netup-dvb-s2-ci-dual
>>
>> Binary builds (ready to install) available in build directory. Currently
>> NetUP Dual DVB-S2 CI card supported (
>> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
>>
>> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI)
>> supported by custom API, for more details see netup_bda_api.h file.
>>
>> Any comments, suggestions are welcome.
>>
>> --
>> Abylai Ospan<aospan@netup.ru>
>> NetUP Inc.
> Am I the only one who thinks this is a legally ambigious grey area?
> Seems like this could be a violation of the GPL as the driver code in
> question links against a proprietary kernel.
>
> I don't want to start a flame war, but I don't see how this is legal.
> And you could definitely question whether it goes against the
> intentions of the original authors to see their GPL driver code being
> used in non-free operating systems.
>
> Devin
>

-- 
Abylai Ospan<aospan@netup.ru>
NetUP Inc.

Re: LinuxTV ported to Windows

[Laurent Pinchart]

Hi Andreas,

On Wednesday 30 November 2011 20:58:40 Andreas Oberritter wrote:
> On 30.11.2011 20:33, Devin Heitmueller wrote:
> > On Wed, Nov 30, 2011 at 1:02 PM, Andreas Oberritter wrote:
> >>> Am I the only one who thinks this is a legally ambigious grey area?
> >>> Seems like this could be a violation of the GPL as the driver code in
> >>> question links against a proprietary kernel.
> >> 
> >> Devin, please! Are you implying that the windows kernel becomes a
> >> derived work of the driver, or that it's generally impossible to publish
> >> windows drivers under the terms of the GPL?
> > 
> > The simple answer is that "I don't know".  I'm not a lawyer (and as
> > far as I know, neither are you).  Nor have I researched the topic to
> > significant lengths.  That said though, whether it was the intention
> > of either copyright holder it's entirely possible that the two
> > software licenses are simply incompatible.  For example, while both
> > the Apache group and the FSF never really intended to prevent each
> > others' software from being linked against each other, the net effect
> > is still that you cannot redistribute such software together since the
> > Apache license is incompatible with the GPL.
> 
> Neither is Abylay distributing Windows together with this driver, nor is
> this driver a library Windows links against, i.e. Windows is able to run
> with this driver removed.

But the driver can't run with Windows.

The important point to remember when discussing licenses is that the GPL 
license mostly affects distribution of binaries. Distribution of the source 
code isn't an issue in this case, as the code is clearly being redistributed 
under the terms of the license. Binaries, however, are a different story.

The resulting Windows driver binary is linked to GPL-incompatible code (namely 
the Windows kernel). I'm not, as most people here, a lawyer, but this can of 
situation always triggers an alarm in my brain. It might not be allowed by the 
GPL license, hence the comment about a *possible* GPL violation.

The binary might also violate Microsoft terms of use. I haven't studied the 
Windows DDK license, but I wouldn't be surprised if it forbade linking GPL 
code to the Windows kernel one way or the other. I won't be personally upset 
if someone violates the Windows DDK license, but it's worth a warning as well.

> >>> I don't want to start a flame war, but I don't see how this is legal.
> >>> And you could definitely question whether it goes against the
> >>> intentions of the original authors to see their GPL driver code being
> >>> used in non-free operating systems.
> >> 
> >> The GPL doesn't cover such intentions.
> > 
> > This isn't necessarily true.  Anybody who has written a library and
> > released it under the GPL instead of the LGPL has made a conscious
> > decision that the library is only to be used by software that is GPL
> > compatible.  By their actions they have inherently forbidden it's use
> > by non-free software.  You could certainly make the same argument
> > about a driver -- that they authors intent was to ensure that it only
> > be linked against other free software.
> 
> That's something completely different than "being used in non-free
> operating systems" and not necessarily comparable to a driver, which
> implements a well-defined interface.

It doesn't matter much if the interface is well-defined or not. What matters 
is the GPL license on one side, and the related Windows licenses on the other 
side. Distributing a Windows binary driver made of GPL code needs to comply 
with licenses on both sides.

Whether the original author intent was to forbid usage of the code in a 
proprietary operating system isn't really relevant from a legal point of view. 
Sure, it would be nice to take the original author opinion into consideration, 
but there's at best (or at worst, depending on the point of view) only a moral 
need to do so. When Google uses my kernel code in Android, with a proprietary 
(yet open) userspace that in my opinion hurts Linux, I'm not the happiest 
person in the world, but I live with it without complaining (OK, that's not 
completely true, I complain about Android creating a lot of new kernel APIs 
without any cooperation with the Linux community, but that's another story).

> The point I'm trying to make: Someone made a presumably nice open source
> port to a new platform and the first thing you're doing is to play the
> GPL-has-been-violated-card, even though you're admitting that you don't
> know whether any right is being violated or not. Please don't do that.
> It's not very encouraging to someone who just announced free software.

Thanks for pointing this out. I would have reacted similarly to Devin, not to 
discourage Abylay from working on this interesting project, but to warn him 
that there might be legal issues (I think we would all prefer early 
notifications from the community than late notifications from unfriendly 
lawyers :-)). You understanding this as an attack shows that we need to be 
more careful in the way we word our messages on license-related issues.

-- 
Regards,

Laurent Pinchart

Re: LinuxTV ported to Windows

[Steven Toth]

>> The point I'm trying to make: Someone made a presumably nice open source
>> port to a new platform and the first thing you're doing is to play the
>> GPL-has-been-violated-card, even though you're admitting that you don't
>> know whether any right is being violated or not. Please don't do that.
>> It's not very encouraging to someone who just announced free software.
>
> Thanks for pointing this out. I would have reacted similarly to Devin, not to
> discourage Abylay from working on this interesting project, but to warn him
> that there might be legal issues (I think we would all prefer early
> notifications from the community than late notifications from unfriendly
> lawyers :-)). You understanding this as an attack shows that we need to be
> more careful in the way we word our messages on license-related issues.

I've been silent as I wanted to see how the thread evolved. This is a
response in general to the group - not any individual.

Speaking as the maintainer and copyright owner I can say that it would
have been nice if someone had contacted me privately re the matter,
before hand. Not to assert any legal right, not for any approval,
simply as a courtesy and a perhaps a small 'Thank You'. NetUp could
have happily had my personal blessing on their project.

My first concern is that this only benefits NetUp on Windows, no other
company benefits on windows - as they all already have legal access to
the Conexant source reference driver. The Windows GPL driver
could/will evolve much faster than the Linux driver and that will suit
NetUp commercially and nobody else. Time will not be taken to
"backport" changes into the Linux driver and that's bad for the Linux
community. (Or, for commercial reasons, the backports will take longer
than expected)

My second concern is that NetUp have made it very simply for the
hundreds of no-name third party far-east companies (with zero
legitimate access to the Conexant windows source reference driver), to
take the windows driver, close source it, not distribute their changes
and compete against the few legitimate TVTuner companies left in the
world. If/when the one or two remaining TVTuner companies die because
their bread and butter Windows sales are being eroded to zero - how
does this help this community? It doesn't, it only helps NetUp.

I embrace open source, I welcome new developers, debate and growth....
I just think if you are going to get my 18 year old daughter pregnant
then it's courtesy to knock on my door and introduce yourself first -
regardless of my opinion or your legal rights.

-- 
Steven Toth - Kernel Labs
http://www.kernellabs.com

Re: LinuxTV ported to Windows

[Andreas Oberritter]

Hello Laurent,

On 01.12.2011 20:42, Laurent Pinchart wrote:
> Hi Andreas,
> 
> On Wednesday 30 November 2011 20:58:40 Andreas Oberritter wrote:
>> On 30.11.2011 20:33, Devin Heitmueller wrote:
>>> On Wed, Nov 30, 2011 at 1:02 PM, Andreas Oberritter wrote:
>>>>> Am I the only one who thinks this is a legally ambigious grey area?
>>>>> Seems like this could be a violation of the GPL as the driver code in
>>>>> question links against a proprietary kernel.
>>>>
>>>> Devin, please! Are you implying that the windows kernel becomes a
>>>> derived work of the driver, or that it's generally impossible to publish
>>>> windows drivers under the terms of the GPL?
>>>
>>> The simple answer is that "I don't know".  I'm not a lawyer (and as
>>> far as I know, neither are you).  Nor have I researched the topic to
>>> significant lengths.  That said though, whether it was the intention
>>> of either copyright holder it's entirely possible that the two
>>> software licenses are simply incompatible.  For example, while both
>>> the Apache group and the FSF never really intended to prevent each
>>> others' software from being linked against each other, the net effect
>>> is still that you cannot redistribute such software together since the
>>> Apache license is incompatible with the GPL.
>>
>> Neither is Abylay distributing Windows together with this driver, nor is
>> this driver a library Windows links against, i.e. Windows is able to run
>> with this driver removed.
> 
> But the driver can't run with Windows.

I guess you meant "can't run without Windows". It's probably safe to
assume that, but it's not relevant, unless you're questioning whether
Windows' licensing terms allow running free software or not.

> The important point to remember when discussing licenses is that the GPL 
> license mostly affects distribution of binaries. Distribution of the source 
> code isn't an issue in this case, as the code is clearly being redistributed 
> under the terms of the license. Binaries, however, are a different story.
> 
> The resulting Windows driver binary is linked to GPL-incompatible code (namely 
> the Windows kernel). I'm not, as most people here, a lawyer, but this can of 
> situation always triggers an alarm in my brain. It might not be allowed by the 
> GPL license, hence the comment about a *possible* GPL violation.

Do you also think it would violate the GPL to distribute binaries of
GPL'ed programs linked to android's libc or msvcrt, for example, because
they both have GPL incompatible licenses?

> The binary might also violate Microsoft terms of use. I haven't studied the 
> Windows DDK license, but I wouldn't be surprised if it forbade linking GPL 
> code to the Windows kernel one way or the other. I won't be personally upset 
> if someone violates the Windows DDK license, but it's worth a warning as well.

I'm used to SDK licenses that don't affect the license of the code
developed using it. Furthermore, only Abylay knows which rights his copy
of the DDK grants, so any public discussion about it is moot.

>>>>> I don't want to start a flame war, but I don't see how this is legal.
>>>>> And you could definitely question whether it goes against the
>>>>> intentions of the original authors to see their GPL driver code being
>>>>> used in non-free operating systems.
>>>>
>>>> The GPL doesn't cover such intentions.
>>>
>>> This isn't necessarily true.  Anybody who has written a library and
>>> released it under the GPL instead of the LGPL has made a conscious
>>> decision that the library is only to be used by software that is GPL
>>> compatible.  By their actions they have inherently forbidden it's use
>>> by non-free software.  You could certainly make the same argument
>>> about a driver -- that they authors intent was to ensure that it only
>>> be linked against other free software.
>>
>> That's something completely different than "being used in non-free
>> operating systems" and not necessarily comparable to a driver, which
>> implements a well-defined interface.
> 
> It doesn't matter much if the interface is well-defined or not. What matters 
> is the GPL license on one side, and the related Windows licenses on the other 
> side. Distributing a Windows binary driver made of GPL code needs to comply 
> with licenses on both sides.
> 
> Whether the original author intent was to forbid usage of the code in a 
> proprietary operating system isn't really relevant from a legal point of view. 
> Sure, it would be nice to take the original author opinion into consideration, 
> but there's at best (or at worst, depending on the point of view) only a moral 
> need to do so. When Google uses my kernel code in Android, with a proprietary 
> (yet open) userspace that in my opinion hurts Linux, I'm not the happiest 
> person in the world, but I live with it without complaining (OK, that's not 
> completely true, I complain about Android creating a lot of new kernel APIs 
> without any cooperation with the Linux community, but that's another story).

I'd go a step further and say that there's no moral need at all to do
anything not covered by the license. One cannot publish code under a
free software license and later expect people not to use that license
grant, because its use wouldn't make one of many original authors feel
good. Many people writing on this mailing list should think about that.

>> The point I'm trying to make: Someone made a presumably nice open source
>> port to a new platform and the first thing you're doing is to play the
>> GPL-has-been-violated-card, even though you're admitting that you don't
>> know whether any right is being violated or not. Please don't do that.
>> It's not very encouraging to someone who just announced free software.
> 
> Thanks for pointing this out. I would have reacted similarly to Devin, not to 
> discourage Abylay from working on this interesting project, but to warn him 
> that there might be legal issues (I think we would all prefer early 
> notifications from the community than late notifications from unfriendly 
> lawyers :-)).

I don't think it's applicable to this case, but a sentence like "I don't
see how this is legal" by Devin probably has potential to trigger more
unfriendly lawyers than a publication of source code together with
precompiled binaries. I mean, seriously, who's going to sue someone for
doing that - even if a license was violated - and what could be gained?
Last time I checked, the GPL was about freedom of source code and the
possibility to recreate (modified) binaries and not about which OS the
software may be used on.

Second, Abylay is cleary writing from a company email address, so
everybody could at least assume that a corporate law person acked the
publication. And if there was serious doubt about it, a friendly
copyright holder could have contacted him directly to resolve any issues
without needlessly involving the public.

Third, and this one upsets me most: Although we all seem to agree that
none of us knows much about international copyright laws and related
stuff in general, whenever LinuxTV code gets reused creatively, I hear
the same people calling for the GPL police. And those people seem to
think that their countries' law applies to every case in the world. And
if those people are unsure about something, their policy always defaults
to forbidden instead of allowed, as if they needed to protect "their"
GPL at any cost. But it's not possible to defend the GPL by mail. This
has to happen in courts. Sending seemingly random GPL-violation alert
emails to mailing lists just discredits the community and creates fear
among possible contributors.

> You understanding this as an attack shows that we need to be 
> more careful in the way we word our messages on license-related issues.

I'd prefer using no words at all and keep discussions at a technical level.

Regards,
Andreas

Re: LinuxTV ported to Windows

[Andreas Oberritter]

Hello Steven,

On 01.12.2011 22:18, Steven Toth wrote:
>>> The point I'm trying to make: Someone made a presumably nice open source
>>> port to a new platform and the first thing you're doing is to play the
>>> GPL-has-been-violated-card, even though you're admitting that you don't
>>> know whether any right is being violated or not. Please don't do that.
>>> It's not very encouraging to someone who just announced free software.
>>
>> Thanks for pointing this out. I would have reacted similarly to Devin, not to
>> discourage Abylay from working on this interesting project, but to warn him
>> that there might be legal issues (I think we would all prefer early
>> notifications from the community than late notifications from unfriendly
>> lawyers :-)). You understanding this as an attack shows that we need to be
>> more careful in the way we word our messages on license-related issues.
> 
> I've been silent as I wanted to see how the thread evolved. This is a
> response in general to the group - not any individual.
> 
> Speaking as the maintainer and copyright owner I can say that it would
> have been nice if someone had contacted me privately re the matter,
> before hand. Not to assert any legal right, not for any approval,
> simply as a courtesy and a perhaps a small 'Thank You'. NetUp could
> have happily had my personal blessing on their project.

you could have said thank you for porting the driver as well: The port
enlarges the user base, is likely to uncover bugs and you might even
receive fixes to those bugs for free (unless the ranting goes on).

> My first concern is that this only benefits NetUp on Windows, no other
> company benefits on windows - as they all already have legal access to
> the Conexant source reference driver.

Are you implying that
a) it's not the users who benefit most?
b) other companies won't be able to use this driver?
c) NetUp doesn't have legal access to the reference driver?

> The Windows GPL driver
> could/will evolve much faster than the Linux driver and that will suit
> NetUp commercially and nobody else. Time will not be taken to
> "backport" changes into the Linux driver and that's bad for the Linux
> community. (Or, for commercial reasons, the backports will take longer
> than expected)

Why don't you do the backports yourself? You want NetUp to do the work
for you? The code is published in a Git repository. You can easily track
any changes.

> My second concern is that NetUp have made it very simply for the
> hundreds of no-name third party far-east companies (with zero
> legitimate access to the Conexant windows source reference driver), to
> take the windows driver, close source it, not distribute their changes
> and compete against the few legitimate TVTuner companies left in the
> world. If/when the one or two remaining TVTuner companies die because
> their bread and butter Windows sales are being eroded to zero - how
> does this help this community? It doesn't, it only helps NetUp.

Any company doing that could use any existing binary driver as well.
Besides that, I'm sure it's no problem for them to get access to any
reference driver they want.

> I embrace open source, I welcome new developers, debate and growth....
> I just think if you are going to get my 18 year old daughter pregnant
> then it's courtesy to knock on my door and introduce yourself first -
> regardless of my opinion or your legal rights.

A very compelling analogy.

Regards,
Andreas

Re: LinuxTV ported to Windows

[Steven Toth]

>> Speaking as the maintainer and copyright owner I can say that it would
>> have been nice if someone had contacted me privately re the matter,
>> before hand. Not to assert any legal right, not for any approval,
>> simply as a courtesy and a perhaps a small 'Thank You'. NetUp could
>> have happily had my personal blessing on their project.
>
> you could have said thank you for porting the driver as well: The port
> enlarges the user base, is likely to uncover bugs and you might even
> receive fixes to those bugs for free (unless the ranting goes on).

I care not for a windows port, it's of no interest to me. I'm sure
NetUp Windows customers will find it useful.

>> My first concern is that this only benefits NetUp on Windows, no other
>> company benefits on windows - as they all already have legal access to
>> the Conexant source reference driver.
>
> Are you implying that
> a) it's not the users who benefit most?
> b) other companies won't be able to use this driver?
> c) NetUp doesn't have legal access to the reference driver?

I was simply stating my opinion, it not a list of points I wish to
debate with you or anyone else. Please don't take this comment
personally. You are welcome to your own opinion and draw your own
conclusions on how I feel about the matter.

>> The Windows GPL driver
>> could/will evolve much faster than the Linux driver and that will suit
>> NetUp commercially and nobody else. Time will not be taken to
>> "backport" changes into the Linux driver and that's bad for the Linux
>> community. (Or, for commercial reasons, the backports will take longer
>> than expected)
>
> Why don't you do the backports yourself? You want NetUp to do the work
> for you? The code is published in a Git repository. You can easily track
> any changes.

Yes, I could, thanks to github.

>
>> My second concern is that NetUp have made it very simply for the
>> hundreds of no-name third party far-east companies (with zero
>> legitimate access to the Conexant windows source reference driver), to
>> take the windows driver, close source it, not distribute their changes
>> and compete against the few legitimate TVTuner companies left in the
>> world. If/when the one or two remaining TVTuner companies die because
>> their bread and butter Windows sales are being eroded to zero - how
>> does this help this community? It doesn't, it only helps NetUp.
>
> Any company doing that could use any existing binary driver as well.
> Besides that, I'm sure it's no problem for them to get access to any
> reference driver they want.

I think I respectfully disagree with you.

>> I embrace open source, I welcome new developers, debate and growth....
>> I just think if you are going to get my 18 year old daughter pregnant
>> then it's courtesy to knock on my door and introduce yourself first -
>> regardless of my opinion or your legal rights.
>
> A very compelling analogy.

Best wishes,

- Steve

-- 
Steven Toth - Kernel Labs
http://www.kernellabs.com

Re: LinuxTV ported to Windows

[Issa Gorissen]

> Hello,
> 
> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista, 
> XP, win7 tested). Results available under GPL and can be checkout from 
> git repository:
> https://github.com/netup/netup-dvb-s2-ci-dual
> 
> Binary builds (ready to install) available in build directory. Currently 
> NetUP Dual DVB-S2 CI card supported ( 
> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
> 
> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI) 
> supported by custom API, for more details see netup_bda_api.h file.
> 
> Any comments, suggestions are welcome.
> 
> -- 
> Abylai Ospan<aospan@netup.ru>
> NetUP Inc.

Yes indeed, it is a pity but it seems this work is in violation of the GPL.

--
This General Public License does not permit incorporating your program into
proprietary programs. If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with the
library. If this is what you want to do, use the GNU Library General
Public License instead of this License.
--

[http://www.gnu.org/philosophy/why-not-lgpl.html]
[http://www.gnu.org/copyleft/gpl.html]

Re: LinuxTV ported to Windows

[Felipe Magno de Almeida]

On Fri, Dec 2, 2011 at 7:35 AM, Issa Gorissen <flop.m@usa.net> wrote:
>> Hello,
>>
>> We have ported linuxtv's cx23885+CAM en50221+Diseq to Windows OS (Vista,
>> XP, win7 tested). Results available under GPL and can be checkout from
>> git repository:
>> https://github.com/netup/netup-dvb-s2-ci-dual
>>
>> Binary builds (ready to install) available in build directory. Currently
>> NetUP Dual DVB-S2 CI card supported (
>> http://www.netup.tv/en-EN/dual_dvb-s2-ci_card.php ).
>>
>> Driver based on Microsoft BDA standard, but some features (DiSEqC, CI)
>> supported by custom API, for more details see netup_bda_api.h file.
>>
>> Any comments, suggestions are welcome.
>>
>> --
>> Abylai Ospan<aospan@netup.ru>
>> NetUP Inc.
>
> Yes indeed, it is a pity but it seems this work is in violation of the GPL.

The GPL has specific provisions for system libraries, which would IMO,
constitute the kernel AFAIU. So it would not violate the GPL.

> --
> This General Public License does not permit incorporating your program into
> proprietary programs. If your program is a subroutine library, you may
> consider it more useful to permit linking proprietary applications with the
> library. If this is what you want to do, use the GNU Library General
> Public License instead of this License.
> --
>
> [http://www.gnu.org/philosophy/why-not-lgpl.html]
> [http://www.gnu.org/copyleft/gpl.html]
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-media" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Felipe Magno de Almeida

Re: LinuxTV ported to Windows

[Rémi Denis-Courmont]

	Hello,

A GPL troll, as the "Vicious Nokia Employee [that got] VLC Removed from Apple 
App Store" I cannot resist...

Le mercredi 30 novembre 2011 19:23:26 Devin Heitmueller, vous avez écrit :
> Am I the only one who thinks this is a legally ambigious grey area?
> Seems like this could be a violation of the GPL as the driver code in
> question links against a proprietary kernel.

If you have any doubt, I would suggest you ask the SFLC. They tend to give 
valuable insights into that sort of problems. It might be intricate and/or not 
what you want to hear from them though (Been there done that).

> I don't want to start a flame war, but I don't see how this is legal.
> And you could definitely question whether it goes against the
> intentions of the original authors to see their GPL driver code being
> used in non-free operating systems.

As long as the distributed binaries do not include any GPL-incompatible code 
(presumably from Microsoft), there should be no GPL contamination problem. So 
it boils down to whether the driver binary has non-GPL code in it. I don't see 
how the license of the Windows code is relevant, so long as NetUp is not 
distributing the Windows OS alongside the driver (or vice versa).

And while I do not know the Windows DDK license, I doubt it cares much about 
the driver license, so long as Microsoft does not need to distribute nor 
certify the driver.


There may however be problems with the toolchain. The driver binary must be 
recompilable with just the GPL'd source code and "anything that is normally 
distributed with the operating system". VisualStudio is not distributed with 
Windows. In fact, it is sold as a separate product, except for restrictive 
freeware versions.

So unless this driver can be compiled with a GPL-compatible toolchain (and the 
toolchain is provided by NetUp), it might not be possible to distribute binary 
copies of the driver.

Then again, I am not a laywer. Someone that cares, please ask SFLC or friends.

-- 
Rémi Denis-Courmont
http://www.remlab.net/
http://fi.linkedin.com/in/remidenis

Re: LinuxTV ported to Windows

[Patrick Dickey]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 12/02/2011 12:03 PM, Rémi Denis-Courmont wrote:
> Hello,
> 
> A GPL troll, as the "Vicious Nokia Employee [that got] VLC Removed
> from Apple App Store" I cannot resist...
> 
> Le mercredi 30 novembre 2011 19:23:26 Devin Heitmueller, vous avez
> écrit :
>> Am I the only one who thinks this is a legally ambigious grey
>> area? Seems like this could be a violation of the GPL as the
>> driver code in question links against a proprietary kernel.
> 
> If you have any doubt, I would suggest you ask the SFLC. They tend
> to give valuable insights into that sort of problems. It might be
> intricate and/or not what you want to hear from them though (Been
> there done that).
> 
>> I don't want to start a flame war, but I don't see how this is
>> legal. And you could definitely question whether it goes against
>> the intentions of the original authors to see their GPL driver
>> code being used in non-free operating systems.
> 
> As long as the distributed binaries do not include any
> GPL-incompatible code (presumably from Microsoft), there should be
> no GPL contamination problem. So it boils down to whether the
> driver binary has non-GPL code in it. I don't see how the license
> of the Windows code is relevant, so long as NetUp is not 
> distributing the Windows OS alongside the driver (or vice versa).
> 
> And while I do not know the Windows DDK license, I doubt it cares
> much about the driver license, so long as Microsoft does not need
> to distribute nor certify the driver.
> 

I'm not sure about the Windows DDK license either, but I can tell you
that at least some of their licenses specifically forbid you to use
their libraries or source code in GPL-style programs.

I downloaded an iso of the Windows Driver Kit version 7.1.0, and what
I found in the Windows Development Kit license is this (concerning
redistributed code from the WDK)

> iii. Distribution Restrictions.  You may not alter any copyright,
> trademark or patent notice in the Distributable Code; use
> Microsoft’s trademarks in your programs’ names or in a way that
> suggests your programs come from or are endorsed by Microsoft;

********
> distribute Distributable Code to run on a platform other than the
> Windows platform;

********
> include Distributable Code in malicious, deceptive or unlawful
> programs; or

********
> modify or distribute the source code of any Distributable Code so
> that any part of it becomes subject to an Excluded License.  An
> Excluded License is one that requires, as a condition of use,
> modification or distribution, that the code be disclosed or
> distributed in source code form; or others have the right to modify
> it.

********

Which would tell me that you can't redistribute their code in any
product licensed under the GPL. So, if NetUP used any of the
redistributable code from the development kit in their port, it would
violate Microsoft's licensing. And any code that NetUP used cannot be
backported to Linux.  (The **'s are my emphasis of what I believe are
the relevant portions of the license)

> 
> There may however be problems with the toolchain. The driver binary
> must be recompilable with just the GPL'd source code and "anything
> that is normally distributed with the operating system".
> VisualStudio is not distributed with Windows. In fact, it is sold
> as a separate product, except for restrictive freeware versions.
> 
> So unless this driver can be compiled with a GPL-compatible
> toolchain (and the toolchain is provided by NetUp), it might not be
> possible to distribute binary copies of the driver.
> 
> Then again, I am not a laywer. Someone that cares, please ask SFLC
> or friends.
> 

I agree about contacting the SFLC. Also the copyright holder (Steven
Toth) weighed in about his concerns. So at the end of the day, this is
between him and the developers (NetUP, Abylay, etc).

GPL questions/potential issues aside, I can see some benefit from
this. Just in the idea that if the port works fairly decently, and
with the proper advertising, it might get the name "Linux" into the
average user's field of view (so to speak). Of course if the port is
crap, or if you have to pay for the product (or pay for a
spyware/adware free version), then it might have the opposite effect.
This is just my 2 cents worth, as an end-user mainly.

Have a great day:)
Patrick.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk7ZTLcACgkQMp6rvjb3CARnqwCgy6MqGTObMugv1S0v5gOTf/xx
f+sAn3hkImJvOCVMJlKcnV/b+VfI4wZL
=UJN4
-----END PGP SIGNATURE-----