General Discussion about GPLness

General Discussion about GPLness

[Stephan von Krawczynski]

Hello all,

you may have already heard about it or not (several times in the past),
non-kernel devices run into a symbol export problem as soon as something is
only exported GPL from the kernel.
Currently there is a discussion regarding zfs using this call chain:

vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) -> percpu_ref_tryget
(kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)

where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL exported)
percpu_ref_tryget.

That this popped up (again) made me think a bit more general about the issue.
And I do wonder if this rather ideologic problem is on the right track
currently. Because what the kernel tries to do with the export GPL symbol
stuff is to prevent any other licensed software from _using_ it in _runtime_.
It does not try to prevent use/copy of the source code inside another non-gpl
project.
And I do think that this is not the intention of GPL. If it were, then 100% of
all mobile phones on this planet are illegal. All of them use GPL software
from non-gpl software, be it kernel modules or apps - and I see no difference
in the two. The constructed difference between kernel mode software and
user-space software is pure ideology. Because during runtime everything is
just call-chained.
Which means if you fopen() a file in user-space it of course uses GPL symbols
down in the chain somewhere. The contents of the opened file are not
heaven-sent.
If you/we follow the current completely ideology-driven GPL strategy then I am
all for completely giving up this whole project. In real world you simply
cannot use such a piece of software. The success of linux during the last
years (i.e. decade) is not based on the pure GPL strategy, but on the
successful interaction between linux and non-GPL software.
Just think of the billions of smartphones all using a non-gpl firmware
(underneath, and there is no GPL version at all), the kernel (with non-gpl
modules) and apps (quite some of which are non-gpl).
This is only one prominent example, but there are lots of others.
In the end it all sums up to one simple question:
Can one _use_ GPL software during runtime as a base for own projects of any
license type or not? We are not talking about _copying_ gpl code, we are
talking about runtime use.
If runtime use is generally allowed, then the export gpl symbol stuff inside
the kernel code is nonsense. Because to use the kernel you must be allowed to
call it, no matter from where.
Hit me.

-- 
Regards,
Stephan

Re: General Discussion about GPLness

[Bernd Petrovitsch]

Hi all!

On 24/02/2020 10:29, Stephan von Krawczynski wrote:
> On Mon, 24 Feb 2020 00:46:03 +0100
> Bernd Petrovitsch <bernd@tuxoid.at> wrote:
[...]
>> On 23/02/2020 21:47, Stephan von Krawczynski wrote:
[...]
>> ... publish implicitly with a GPLv2 or compatible license ...
> 
> Ok, you cannot really mean this. Because what you say is that no company
> making some hardware is allowed to produce a closed-source kernel driver for

It's not that simple - in neither direction (and no, I won't give guidelines
or reproduce the lame excuses and/or other reasons for creating "proprietary
Linux kernel drivers". They can very probably be found via $SEARCH_ENGINE
anyways).

The law term is "derived work" (you will probably find lots of
discussions if you $SEARCH_ENGINE it) and it's up to the module/driver
author to prove that the module/driver is not derived work from the
GPLed software it interoperates.
If you copy a skeleton driver (or cannibalise another driver) it's hard
to say it's not derived.

For the syscall layer, there exists POSIX, SUSv3 et.al. as well as lots
of more or more-the-like less identical implementations long before the
first line of .c of the Kernel.

One ugly and nasty thing about law is that it's not automatically
correct just because one (company or person) does/claims it. And it
doesn't make it right, legal, etc.
Not sueing that one also doesn't change a bit of the legal status.

Another even more ugly and nasty thing about a given law is that it's
bound to it's jurisdiction. Yes, some countries/goverments/
administrations want to extend it beyond their own and several can
actually do that but we are again in political swamp.
In short: Law people basically ignored the globalization as I see almost
no progress in the last decades in that area - just lame excuses and
propaganda-style discussions about the "law less internet" - there is
no "law less" area on the world as everywhere is at least one
jurisdiction.

> it. You know who are the last ones on the planet taking that approach? It's> China. If you want to build some car there, you have to give them all the

It is neither technically nor law-wise relevant and/or interesting or a
point for or against anything if some powers-that-be plain simply ignore
laws. At most, it's a political thing ....

> details of everything and only then are allowed to do a joint-venture with
> chinese where you are not allowed to hold a majority.

And every sane company which sees this should think about it and then
decide if they want to go there. I wouldn't do that (and hadn't do that
15 years ago).
Obviously a lot of them have no problem with these kind of "giving away
intellectual property, know-how and God knows what more". There are many
more stories about "what happens after moving $FACTORY to China" which
are off-topic here ....

> You are saying that every company creating new hardware must open up the
> details to it so that they themselves or someone else is capable to write a
> GPL-driver for it. Is that the world you want to live in? China? Really?

For one that's IMHO the basic intention of the GPL (read and think about
"the four essential freedoms" on
https://www.gnu.org/philosophy/free-sw.html and their implications).
And one main reason is that the user of the driver withthe 4 freedoms is
capable of fixing errors in it - ask RMS about it;-)

And it's IMNSHO *not* the intention of China to produce an open, better
world - quite the contrary - so please don't try compare that or use that
in any discussion/analysis.

BTW it's IMHO a major difference between publishing documentation for
everyone just for creating interoperable software and forcing others to
release (all?) information and allow major influence for nation-controlled
bodies to company/project/organizations as such.

>>> defined and open interface for interaction. No kernel code is modified in
>>> that  
>>
>> And that's a point which is completely wrong: the only *defined*
>> interface of the kernel is sys-calls, /proc and similar. And all of this
>> is from user-space to kernel-space and back.
> 
> This is "said" to be so. But since I wrote several kernel drivers I know that
> there is at least a skeleton of what a driver should look like. And this means
> that there is a definition. I know what you mean. It is said that none of the

No, that's only an implementation model - even if it's published in a book.
The point is: It may change with any git-commit so it's a total different story
than syscalls et.al. (which are set in stone and it may produce "funny"
discussions if people want to bug-fix something there).

And you always have implicit skeleton (e.g. simple) drivers (think "/dev/null"
if you are looking for a minimalistic char driver) which people take as model
or even copy and edit them.
BTW if you copied a GPLv2 driver/skeleton/enough source code, you also copied
it's license too. Authors rights and copyright 101 .....

> interfacing is defined as some standard written in stone. But we all know the
> reason: the project wants to stay evulatory and wants to be able to implement
> new - possibly better ideas. But only because its flowing does not mean there> is none. Else there would not be lots of people working on different drivers
> filesystems and the like. They all have to meet at some point in the code. And
> that meeting point(s) is(are) the open interface. And it's open because

But it's not a *defined* interface in any technical interpretation. Do you
consider an interface which changes in every release (without backward
compatibility) a "defined interface"? I don't ....

At most it's a documented interface ...

If you want a *defined* driver interface, go to the Windows world as M$FT is
committed to maintain kernel-internal interfaces across major versions (yes,
they don't do it in 100% of all cases - just talk to Windows driver
maintainers which have to touch their driver with at least every version
...).

> everyone is allowed to read the source and use the exported functions (be them
> GPL-defined or not) to write some _extensions_.
>  
>> There is no (and never was) a "defined interface" within the kernel
>> (inter-operating in kernel space) in any direction simply because the
>> kernel-internal (infra)structure changes more or less constantly - some
>> may call that evolution;-)
[...]
>> And I don't get what an "open interface" here couls seriously mean. You
>> surely don't want to call the list of the GPL_EXPORTed C functions
>> (which may change from one kernel version to the next) an "open
>> interface" (whatever that should suggest).
>>
>> At most the list of GPL_EXPORTed C functions is a de-facto interface and
>> that may change from one git-commit to the next.
> 
> The lifetime of the details of this interface is not relevant. Some last
> years, some maybe only days. Nevertheless the EXPORT per se means that someone
> _else_ should and can use them. Otherwise there would be no EXPORT definition.

EXPORT has the same semantics from the (running) kernel to modules that
"extern" has in .c from one .o to another .o as you can't use the function
in a module without EXPORT - not more, not less.

>>> sense. But you fail to understand that.
>>> Hopefully others here do. I do not expect them to stand up and jump into a
>>> discussion where you are a part of. But I hope people start to think about
>>> it  
>>
>> And you are barking up the wrong tree. The discussion has to happen in
>> the ZFS-world so that they fix their license if they want to interface
>> with any GPL software (without sys-calls etc. in between - WTF it works
>> via FUSE) like they seen to do now.
> 
> Ok, now please stay serious. FUSE is nothing more than a hack for people who

I'm dead serious. I actually thought we talked about license/law stuff and
runtime situations which have no relevance when it comes to license/law issues.

> want to make case studies on projects but is completely useless in production.
> Because it is DEAD slow. If you want to check out, look at glusterfs (that
And it solves the license issue. That's the price to pay, it's one possible
solution to your license problem, it's up to you where you find the - Your -
least cumbersome compromise ....

> never managed to be implemented as kernel driver). And nobody used ZFS on FUSE,
> because it was unusably slow. Since it has become a kernel space driver it is
> amongst the best fs around.

That's all completely irrelevant for license and law discussions.

> But zfs is not the main issue. It is only an example of the problem. 

So what's the main issue that you abuse ZFS for your agenda?

You feel limited by the GPLv2 (or any othre license)? Then don't use such
software. Problem solved.

Hell, there are people out there who will help only in GPL projects (and not
BSD and near-BSD as that's basically - and intentionally(!)- a "proprietary
license on call").

> Companies and other entities must be allowed to make kernel drivers and
> therefore interface with the kernel in kernel space no matter what licensing

Everyone can integrate ZFS on their own and themselves in their product
without any license issue.

> they prefer. This is my opinion. I would love to hear if this really is the
> projects' policy or not though. This is what this thread is all about.
> 
> Of course, I would prefer a GPL driver as well. That's why I use ATI and not
> nvidia. But nevertheless I think nvidia and all the others should be allowed
> to make their drivers. I'd say the rest is market/customer choice.

Market/customer choice is completely irrelevant for license and law
discussions.

Well, please search in the LKML archives - we have such discussions every
other year here and it won't (and can't) change the GPLv2 (or other versions)
as such.
And for a license change of the Linux kernel, you need the explicit agreement
from all contributors (at least for the more than trivial patches like "fixed
typo" etc. - that's another law discussion when my patch is not trivial enough
so that authors rights/copyright is actually applicable),

MfG,
	Bernd

PS: For all others: That's the end of the discussion from my side - I already
    wrote more than enough and I don't think can (or want) write anything not
    already written in this thread (by /me - of course;-).
-- 
"I dislike type abstraction if it has no real reason. And saving
on typing is not a good reason - if your typing speed is the main
issue when you're coding, you're doing something seriously wrong."
    - Linus Torvalds

Re: General Discussion about GPLness

[whywontyousue]

I see that you're just going to ignore
Universal City Studios Inc v Reimerdes

Instead you address your emotional issues. This is the making of a 
beast: which is what you are. Driven by your instincts.

You want to violate the linux copyright. You announced your intent 
(which increases the damages you are liable for under US jurisprudence.)

The facts are simple: You may not execute the scheme you are suggesting. 
If you do execute the scheme you or suggesting, or if you assist others 
in doing so then you are liable for copyright infringement (direct or 
contributory). If you make over 1000 dollars off of this infringement 
you are liable for criminal penalties as-well.

The courts see unlicensed modification of another copyright-holders 
work, in memory, while running, as the creation of an unlicensed 
derivative work.

It does not matter how you try to sugar coat it or change your 
description, nor do your complaints of me CC'ing relevant parties (a 
diversion technique by you) who could explain this all to you in ways 
you could understand, change anything. The courts in the USA see this 
type of action you suggest as the creation of a derivative work.

Thus you must follow the permissions and must not overstep them. That is 
all. Having non-gpl modules work as you describe violates the linux 
kernel copyright just as the plugin in the Reimerdes case violated the 
RealPlayer copyright: in both cases the situation involves "modules" 
extending the running program in real-time, in both cases the "modules" 
are not permitted by the copyright owners to do as they wish: and the 
Court agreed.

On 2020-02-24 14:01, Stephan von Krawczynski wrote:
> On Mon, 24 Feb 2020 07:46:33 +0000
> whywontyousue@waifu.club wrote:
> 
>> > Hm, really it is quite hard to stay calm reading your constant insults
>> > on
>> 
>> If such is so, it is shown to all that you are a stupid white man, 
>> what
>> can I say. You can't separate facts from fiction (opinion)
>> [...]
> 
> Please stop spamming people completely unrelated to this thread.
> And please stop spamming this thread with insults and blatant 
> ignorance.
> Thank you.
> No more answers from me to you.
> --
> Regards,
> Stephan

Re: General Discussion about GPLness

[Stephan von Krawczynski]

On Mon, 24 Feb 2020 07:46:33 +0000
whywontyousue@waifu.club wrote:

> > Hm, really it is quite hard to stay calm reading your constant insults 
> > on  
> 
> If such is so, it is shown to all that you are a stupid white man, what 
> can I say. You can't separate facts from fiction (opinion)
> [...]

Please stop spamming people completely unrelated to this thread.
And please stop spamming this thread with insults and blatant ignorance.
Thank you.
No more answers from me to you.
--
Regards,
Stephan
 

Re: General Discussion about GPLness

[Stephan von Krawczynski]

On Mon, 24 Feb 2020 00:46:03 +0100
Bernd Petrovitsch <bernd@tuxoid.at> wrote:

> Hi all!

Hello Bernd,
 
> On 23/02/2020 21:47, Stephan von Krawczynski wrote:
> [...]
> > I do understand why you cannot enter a discussion with your real name, as
> > most  
> 
> I ignore folks who don't use their real name - this is since Usenet
> times an indication for a troll ...

Well, yes. But I don't want to start a topic and ignore the first answer right
away ... Call me naive
 
> > of your input is of zero quality - and below.  
> [...]
> >                                                              But our story
> > is about kernel modules, something everybody is free to write and publish,
> > with a  
> 
> ... publish implicitly with a GPLv2 or compatible license ...

Ok, you cannot really mean this. Because what you say is that no company
making some hardware is allowed to produce a closed-source kernel driver for
it. You know who are the last ones on the planet taking that approach? It's
China. If you want to build some car there, you have to give them all the
details of everything and only then are allowed to do a joint-venture with
chinese where you are not allowed to hold a majority.
You are saying that every company creating new hardware must open up the
details to it so that they themselves or someone else is capable to write a
GPL-driver for it. Is that the world you want to live in? China? Really?
 
> > defined and open interface for interaction. No kernel code is modified in
> > that  
> 
> And that's a point which is completely wrong: the only *defined*
> interface of the kernel is sys-calls, /proc and similar. And all of this
> is from user-space to kernel-space and back.

This is "said" to be so. But since I wrote several kernel drivers I know that
there is at least a skeleton of what a driver should look like. And this means
that there is a definition. I know what you mean. It is said that none of the
interfacing is defined as some standard written in stone. But we all know the
reason: the project wants to stay evulatory and wants to be able to implement
new - possibly better ideas. But only because its flowing does not mean there
is none. Else there would not be lots of people working on different drivers,
filesystems and the like. They all have to meet at some point in the code. And
that meeting point(s) is(are) the open interface. And it's open because
everyone is allowed to read the source and use the exported functions (be them
GPL-defined or not) to write some _extensions_.
 
> There is no (and never was) a "defined interface" within the kernel
> (inter-operating in kernel space) in any direction simply because the
> kernel-internal (infra)structure changes more or less constantly - some
> may call that evolution;-)

Yes, read above.
 
> And I don't get what an "open interface" here couls seriously mean. You
> surely don't want to call the list of the GPL_EXPORTed C functions
> (which may change from one kernel version to the next) an "open
> interface" (whatever that should suggest).
> 
> At most the list of GPL_EXPORTed C functions is a de-facto interface and
> that may change from one git-commit to the next.

The lifetime of the details of this interface is not relevant. Some last
years, some maybe only days. Nevertheless the EXPORT per se means that someone
_else_ should and can use them. Otherwise there would be no EXPORT definition.
 
> > sense. But you fail to understand that.
> > Hopefully others here do. I do not expect them to stand up and jump into a
> > discussion where you are a part of. But I hope people start to think about
> > it  
> 
> And you are barking up the wrong tree. The discussion has to happen in
> the ZFS-world so that they fix their license if they want to interface
> with any GPL software (without sys-calls etc. in between - WTF it works
> via FUSE) like they seen to do now.

Ok, now please stay serious. FUSE is nothing more than a hack for people who
want to make case studies on projects but is completely useless in production.
Because it is DEAD slow. If you want to check out, look at glusterfs (that
never managed to be implemented as kernel driver). And nobody used ZFS on FUSE,
because it was unusably slow. Since it has become a kernel space driver it is
amongst the best fs around.
But zfs is not the main issue. It is only an example of the problem. 
Companies and other entities must be allowed to make kernel drivers and
therefore interface with the kernel in kernel space no matter what licensing
they prefer. This is my opinion. I would love to hear if this really is the
projects' policy or not though. This is what this thread is all about.

Of course, I would prefer a GPL driver as well. That's why I use ATI and not
nvidia. But nevertheless I think nvidia and all the others should be allowed
to make their drivers. I'd say the rest is market/customer choice.

> MfG,
> 	Bernd

-- 
Regards,
Stephan

Re: General Discussion about GPLness

[whywontyousue]

> Hm, really it is quite hard to stay calm reading your constant insults 
> on

If such is so, it is shown to all that you are a stupid white man, what 
can I say. You can't separate facts from fiction (opinion)

The FACT relevant to this discussion is how the US Courts see this 
issue: and they see a Copyright violation when your Work modifies 
another protected Work without the Copyright owners permission.

You (the creator of the impinging Work) create a derivative work (yes, 
running) this way. You must have permission from the Copyright holder of 
the work you are modifying. You don't like that: I know; you've made it 
clear.

The Opinion that the Linux Kernel Copyright holders are weak scum has 
nothing to do with that and you should simply glide over those passages: 
they really shouldn't upset you.

What should Upset you is that they refuse to ever sue entities like 
OpenSourceSecurity (GrSecurity) and Bradly Spengler, to enforce their 
Copyrights. That has a "lets all violate the license" effect. The GPL 
_IS_ dead because of the INACTION of the linux scumbags and of the FSF 
(which has been captured by industry).

You killed the dream Linux copyright holder: Yes: Fuck you (unless you 
sue: then you are the greatest and I wish for you total and complete 
Victory.)

You asked a question "Why can't we violate the linux copyright in this 
way". Your question was answered.

What more do you want?
Well I gave you more: lots of opinion.
You want more or less?

How about YOU write the response you want?
Go to another email address, and respond to yourself
"Hey, yes, let us violate the linux copyrights, fuck the US courts, it 
doesn't matter"

Go on, I don't want to be the only other "person" in this thread.

Oh, and I've been contributing the Free Software for about 20 years, 
full time, so I haven't done nothing.

On 2020-02-23 20:47, Stephan von Krawczynski wrote:
> Hm, really it is quite hard to stay calm reading your constant insults 
> on
> people that have quite sure done a lot more for free software than you 
> have.
> I do understand why you cannot enter a discussion with your real name, 
> as most
> of your input is of zero quality - and below.
> Unfortunately you did not even try to understand what the true issue is 
> all
> about. It is one story to modify running code that probably was never 
> ment to
> be patched that way (which would involve re-engineering it). But our 
> story is
> about kernel modules, something everybody is free to write and publish, 
> with a
> defined and open interface for interaction. No kernel code is modified 
> in that
> sense. But you fail to understand that.
> Hopefully others here do. I do not expect them to stand up and jump 
> into a
> discussion where you are a part of. But I hope people start to think 
> about it
> and realise - like I did - that this train is on the wrong track.
> After all I do believe that constructive interaction of software is 
> still
> better than destructive building of hurdles and walls.
> Because in the end people are only suffering from this approach and 
> nothing is
> protected.
> --
> Regards,
> Stephan
> 

Re: General Discussion about GPLness

[Bernd Petrovitsch]

Hi all!

On 23/02/2020 21:47, Stephan von Krawczynski wrote:
[...]
> I do understand why you cannot enter a discussion with your real name, as most

I ignore folks who don't use their real name - this is since Usenet
times an indication for a troll ...

> of your input is of zero quality - and below.
[...]
>                                                              But our story is
> about kernel modules, something everybody is free to write and publish, with a

... publish implicitly with a GPLv2 or compatible license ...

> defined and open interface for interaction. No kernel code is modified in that

And that's a point which is completely wrong: the only *defined*
interface of the kernel is sys-calls, /proc and similar. And all of this
is from user-space to kernel-space and back.

There is no (and never was) a "defined interface" within the kernel
(inter-operating in kernel space) in any direction simply because the
kernel-internal (infra)structure changes more or less constantly - some
may call that evolution;-)

And I don't get what an "open interface" here couls seriously mean. You
surely don't want to call the list of the GPL_EXPORTed C functions
(which may change from one kernel version to the next) an "open
interface" (whatever that should suggest).

At most the list of GPL_EXPORTed C functions is a de-facto interface and
that may change from one git-commit to the next.

> sense. But you fail to understand that.
> Hopefully others here do. I do not expect them to stand up and jump into a
> discussion where you are a part of. But I hope people start to think about it

And you are barking up the wrong tree. The discussion has to happen in
the ZFS-world so that they fix their license if they want to interface
with any GPL software (without sys-calls etc. in between - WTF it works
via FUSE) like they seen to do now.

And no, I'm not a lawyer so I won't comment more on the law-aspects -
the above is my short summary of discussions hereover about license
clashes etc. over the last decades ...

[ Full quote deleted - pls do not top-post]

MfG,
	Bernd
-- 
Bernd Petrovitsch                  Email : bernd@petrovitsch.priv.at
                     LUGA : http://www.luga.at

Re: General Discussion about GPLness

[Stephan von Krawczynski]

Hm, really it is quite hard to stay calm reading your constant insults on
people that have quite sure done a lot more for free software than you have.
I do understand why you cannot enter a discussion with your real name, as most
of your input is of zero quality - and below.
Unfortunately you did not even try to understand what the true issue is all
about. It is one story to modify running code that probably was never ment to
be patched that way (which would involve re-engineering it). But our story is
about kernel modules, something everybody is free to write and publish, with a
defined and open interface for interaction. No kernel code is modified in that
sense. But you fail to understand that.
Hopefully others here do. I do not expect them to stand up and jump into a
discussion where you are a part of. But I hope people start to think about it
and realise - like I did - that this train is on the wrong track.
After all I do believe that constructive interaction of software is still
better than destructive building of hurdles and walls.
Because in the end people are only suffering from this approach and nothing is
protected.
--
Regards,
Stephan





On Sun, 23 Feb 2020 16:24:28 +0000
whywontyousue@waifu.club wrote:

> I gave you an example of a court immediately finding that modifying a 
> running program is not allowed without permission. Creators of Non-gpl'd 
> programs don't have permission to interact with a GPL'd work by 
> modifying or extending it. If the principal stands for RealPlayer, it 
> stands for Linux.
> 
> The GPL is a copyright license; law is the domain this is in.
> 
> In the domain of programming: you can do whatever you want. No technical 
> measure is stopping you.
> 
> In the domain of what is "right and wrong" but excluding law, since you 
> mentioned it and thus opened the door. Well... YHWH allows men to have 
> female children as brides:
> >>>> The Torah explicitly allows men to marry female children, including 
> >>>> in cases of the rape (tahphas) of the girl child: Devarim chapter 
> >>>> 22, verse 28. Key words: Na'ar (child (hebrew masoretic text)), 
> >>>> Padia (child: padia+philos = paedophillia (greek septuagint)) Puella 
> >>>> (young girl (latin vulgate))
> >>>> Nachmanides points out that a child may be called na'ar from the 
> >>>> moment he is born.  
> White idiots (such as the Linux programmers like Linus Trovalds) do not: 
> White idiots worship white women. This is a problem for Free Software 
> because White idiots (like Linus Trovalds) will do /anything/ to make 
> money for "Muh whoite wuhman". That includes cowering in the face of 
> being "blackballed" from the industry if they DARE enforce their 
> copyrights.
> 
> In the domain of willpower: the linux copyright holders, atleast the 
> programmers who are copyright holders, are NOT going to sue you for 
> violating the copyright license permissions. They are pieces of shit who 
> don't believe in "Copyleft". They believe in the BSD license, and have 
> effectivly made Linux a BSD-type work since they never enforce the GPL.
> 
> OpenSourceSecurity (GRSecurity) is blatantly violating section 4 and 
> section 6 of the linux copyright license and the Linux copyright-owning 
> programmers would rather punish anyone who brings it up than sue the 
> violators.
> 
> They are weak feckless people: more concerned about making money for 
> "Muh wuhman" than anything else. Pay them no mind. They are scumbag 
> idiots who won't know what they had until it's gone (it is).
> 
> I think that covers all the topics, right?
> 
> So, by law may you do what you suggested: No: it is a copyright 
> violation in the US.
> Is anyTHING going to stop you: no.
> Is anyONE going to try to punish you for it: no.
> 
> Remember: linux copyright holders are either stupid white nerds (if they 
> were smart they would have listened to their parents and become doctors 
> and lawyers, and have tech as a fun hobby. Instead they are wageslaving 
> for "MUUH WHOITE WUUHHMAN": same as any white idiot worker)
> OR corporations who like the BSD license better.
> 
> You don't really have much to worry about. Because linux programmers are 
> what they are. Sheep that thought themselves lions after RMS' win vs 
> Cisco (that put the fear of the violating the copyright license (GPL in 
> that case) for awhile).
> 
> On 2020-02-23 14:39, Stephan von Krawczynski wrote:
> > Hello again,
> > 
> > at least you are beginning to sound a bit more like being able to 
> > discuss
> > something ;-)
> > The thing about a lawyer (I learned you are) is that they judge the 
> > world
> > according to lawsuits. You can learn from the history of my home 
> > country that
> > laws and courts are no measure for moral and right behaviour.
> > So my whole purpose of the thread is not to find out how the laws in 
> > the US
> > are judging something. It is all about how _we_ (we meaning all the 
> > people
> > contributing) are judging the issue. Do we really think that it is the 
> > right
> > thing to do to prevent people from feature-extending linux (and 
> > distros) in
> > general? The zfs case is special in that it is a simple "free license 
> > clash".
> > Which means all involved parties agree on the free and open software
> > principle, only the licenses (i.e. paper) disagree on the usage of it - 
> > to a
> > certain extent. But lets not discuss legal details. In the end it all 
> > comes up
> > to this simple question: what do we really want with the project?
> > Because in the end even you have to agree that there is a whole lot 
> > more world
> > outside the US. If people from other countries agree on something which 
> > gives
> > a better performance in some area, then the US would be the last ones 
> > not to
> > jump onto the train. Who can testify better than this project, not 
> > being
> > native-US.
> > Do we think it is illegal to call GPL code from non-GPL code? Yes or 
> > no,
> > simple choice.
> > Me, I don't think so. This is why I suggest we take down the barriers 
> > and
> > walls for interaction. It should be obvious by now that there will be 
> > no
> > non-gpl invasion taking place. Instead a non-ideological use of GPL may
> > convince even more people that free and open software is a good concept 
> > and
> > adds benefit to the world and does not _harm_ technological progress.
> > Given, not many people think about this from ground up before releasing
> > software on linux. This is probably the only reason why you can buy 
> > software
> > for linux at all. And maybe, only maybe, the lawyers in your beloved 
> > case
> > where too dumb to turn this case around and ask why the gpl linux 
> > software was
> > forced to marry with the non-gpl real player (which is/was available
> > for linux). According to this courts' point of view this must have been 
> > equally
> > illegal.
> > I mean it extended the gpl software with a new feature without checking 
> > for
> > gpl compliance.
> > As you can see the whole idea of the court in this case is broken. And 
> > it
> > seems only because noone asked the right questions.
> > --
> > Regards,
> > Stephan
> > 
> > 
> > 
> > On Sun, 23 Feb 2020 12:56:13 +0000
> > whywontyousue@waifu.club wrote:
> >   
> >> If you don't understand English, it will be difficult to get any 
> >> points
> >> across to you. I will try
> >> 
> >> In simple terms:
> >> 1) Look up the court case "Universal City Studios Inc v Reimerdes,"
> >> 
> >> 2) In this case someone else' software was running at the same time as
> >> the other persons software, and made changes, extensions _ONLY_ when
> >> running. Just like a non-gpl'd (or gpl'd) module might make changes 
> >> and
> >> extensions.
> >> 
> >> 3) The court found this was obviously a modification of the Copyright
> >> owners Work and barred in on summary judgement.
> >> 
> >> That is why you're not allowed to do as you wish with non-gpl'd 
> >> modules:
> >> the US Copyright Jurisprudence forbids marring a Copyright owners
> >> _running_ _in_memory_ property against his wishes: it's a Copyright
> >> violation.
> >> 
> >> That is the reason: people don't want to get sued. That is the ONLY
> >> reason. That's it.
> >> 
> >> The thing is, the linux copyright owners are wimps and won't sue 
> >> anyone
> >> even for blatant infringement; so what is the conversation about?
> >> 
> >> It's like if you were in Russia, and you were copying DVDs. No one is
> >> going to punish you for it: so what is there to discuss? The US
> >> Copyright owners don't have the rocks to Invade Russia, Start a 
> >> Nuclear
> >> Winter, and DESTROY you for your Copyright Infringement in Russia.
> >> 
> >> JUST AS, the DOG LIKE Linux Copyright owners don't have the BALLS to
> >> risk being blackballed from the programming industry for DEFENDING 
> >> THEIR
> >> COPYRIGHT.
> >> 
> >> The GPL ___IS___ dead. The FSF doesn't protect GCC copyright, and is
> >> opposed to taking any action against blantant in-writing infringers
> >> (OpenSourceSecurity (Grsecurity)) of GCC, just as the LINUX COMMUNITY 
> >> is
> >> OPPOSED to taking ANY action to defend its Copyrights and moves to
> >> PUNISH those rightsholders who do.
> >>   
> >> > And another thing: court is for lawyers. Whenever the lawyers take over
> >> > something they don't (want to) understand the end is near ...  
> >> 
> >> I'm a lawyer and a programmer, got something to say?
> >> 
> >>   
> >> > How about talking with real names?  
> >> Why would I do that? Tell me? What is in it for me?
> >> I can stand here, in the forest, taking shots at your bullshit. Safe.
> >> Secure. My words and their veracity the only measure.
> >> 
> >> But if I reveal the messanger; you'll just attack the messenger.
> >> Tell me how _I_ benifit from telling YOU my name. Tell me.
> >> Is it some sort of stupid werkin man white man bravado?
> >>   
> >> > I have no idea why you spam rms or bruce
> >> > with this, as the question is all about _one_ project, namely
> >> > linux-kernel.  
> >> 
> >> You sent a message to the LKML "Hey why can't I violate the GPL? Let's
> >> just do it!". IE: a licensing discussion. RMS and Bruce Perens, the
> >> founder of the Free Software Movement and the Open Source Initiative 
> >> are
> >> relevant parties to the discussion.
> >> 
> >>   
> >> > I'd suggest taking them off this topic again ...  
> >> You also suggested I reveal my identity on the internet...
> >> 
> >> 
> >> On 2020-02-23 12:33, Stephan von Krawczynski wrote:  
> >> > Dear whoeveryouare,
> >> >
> >> > can you please state in a clearer form (more understandable to
> >> > non-native
> >> > english talkers) what your true opinion on the topic is?
> >> > And in case you did not understand what I was saying, here is clearer
> >> > form of
> >> > my opinion:
> >> >
> >> > A kernel module with another license (be it whatsoever) is _no_
> >> > modification
> >> > of the kernel, but an extension of its features. If feature-extension
> >> > is
> >> > against the GPL (which I seriously doubt) then I would say "go back
> >> > onto your
> >> > trees". Because the human race and evolution is about little else than
> >> > feature-extension.
> >> >
> >> > And another thing: court is for lawyers. Whenever the lawyers take over
> >> > something they don't (want to) understand the end is near ...
> >> >
> >> > How about talking with real names? I have no idea why you spam rms or
> >> > bruce
> >> > with this, as the question is all about _one_ project, namely
> >> > linux-kernel.
> >> > I'd suggest taking them off this topic again ...
> >> >
> >> > --
> >> > Regards,
> >> > Stephan
> >> >
> >> >
> >> >
> >> > On Sun, 23 Feb 2020 11:03:56 +0000
> >> > whywontyousue@waifu.club wrote:
> >> >  
> >> >> Dear Stephan von Krawczynski;
> >> >>
> >> >> Universal City Studios Inc v Reimerdes, piece of shit.
> >> >>
> >> >> "[The court] reasoned that Ferret consumers who used the Ferret as a
> >> >> plug-in to the Real Player altered the Real Player user interface by
> >> >> adding the Snap search button or replacing it with the Stream box
> >> >> search
> >> >> engine button. The court concluded that the plaintiff raised
> >> >> sufficently
> >> >> serious questions going to the merits of its claims to warrant an
> >> >> injunction pending trial"
> >> >>
> >> >> Want to violate the linux kernel copyright, you fucking piece of shit?
> >> >> Yes you do. Yes modifying the running kernel with violating pieces is
> >> >> copyright infringement, you fucking piece of shit. Yes you should be
> >> >> sued. Just as Open Source Security (Grsecurity) should be sued for
> >> >> their
> >> >> violations (of section 4 and 6 of the linux kernel copyright license
> >> >> (they're also violating the GCC copyrights too)).
> >> >>
> >> >> Will they be sued? Will you be sued? No: Linux copyright holders are
> >> >> scared little wageslave worker bees. They aren't going to sue you;
> >> >> sorry. Why are you even announcing you intent to violate the
> >> >> copyright?
> >> >> Why even give these dogs such intellectual deference?
> >> >>
> >> >> I wish OpenSourceSecurity would be sued. I wish you would be sued. But
> >> >> linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate
> >> >> them much more than I hate the violators. Complete Dogs. They could
> >> >> move
> >> >> from strenght to strenght, from victory to victory; but they're scared
> >> >> for their "JEHRB"s. I have to say: white men are pathetic scum. If
> >> >> Linux
> >> >> was built by others there would rightfully be lawsuits.
> >> >>
> >> >>
> >> >>  
> >> >> > Stephan von Krawczynski wrote:
> >> >> > Hello all,
> >> >> >
> >> >> > you may have already heard about it or not (several times in the
> >> >> > past), non-kernel devices run into a symbol export problem as soon as
> >> >> > something is
> >> >> > only exported GPL from the kernel.
> >> >> > Currently there is a discussion regarding zfs using this call chain:
> >> >> >
> >> >> > vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) ->
> >> >> > percpu_ref_tryget
> >> >> > (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
> >> >> >
> >> >> > where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL
> >> >> > exported)
> >> >> > percpu_ref_tryget.
> >> >> >
> >> >> > That this popped up (again) made me think a bit more general about
> >> >> > the issue.
> >> >> > And I do wonder if this rather ideologic problem is on the right
> >> >> > track currently. Because what the kernel tries to do with the export
> >> >> > GPL symbol
> >> >> > stuff is to prevent any other licensed software from _using_ it in
> >> >> > _runtime_.
> >> >> > It does not try to prevent use/copy of the source code inside another
> >> >> > non-gpl
> >> >> > project.
> >> >> > And I do think that this is not the intention of GPL. If it were,
> >> >> > then 100% of
> >> >> > all mobile phones on this planet are illegal. All of them use GPL
> >> >> > software
> >> >> > from non-gpl software, be it kernel modules or apps - and I see no
> >> >> > difference
> >> >> > in the two. The constructed difference between kernel mode software
> >> >> > and user-space software is pure ideology. Because during runtime
> >> >> > everything is
> >> >> > just call-chained.
> >> >> > Which means if you fopen() a file in user-space it of course uses GPL
> >> >> > symbols
> >> >> > down in the chain somewhere. The contents of the opened file are not
> >> >> > heaven-sent.
> >> >> > If you/we follow the current completely ideology-driven GPL strategy
> >> >> > then I am
> >> >> > all for completely giving up this whole project. In real world you
> >> >> > simply
> >> >> > cannot use such a piece of software. The success of linux during the
> >> >> > last
> >> >> > years (i.e. decade) is not based on the pure GPL strategy, but on the
> >> >> > successful interaction between linux and non-GPL software.
> >> >> > Just think of the billions of smartphones all using a non-gpl
> >> >> > firmware (underneath, and there is no GPL version at all), the
> >> >> > kernel (with non-gpl
> >> >> > modules) and apps (quite some of which are non-gpl).
> >> >> > This is only one prominent example, but there are lots of others.
> >> >> > In the end it all sums up to one simple question:
> >> >> > Can one _use_ GPL software during runtime as a base for own projects
> >> >> > of any
> >> >> > license type or not? We are not talking about _copying_ gpl code, we
> >> >> > are
> >> >> > talking about runtime use.
> >> >> > If runtime use is generally allowed, then the export gpl symbol stuff
> >> >> > inside
> >> >> > the kernel code is nonsense. Because to use the kernel you must be
> >> >> > allowed to
> >> >> > call it, no matter from where.
> >> >> > Hit me.
> >> >> >
> >> >> > --
> >> >> > Regards,
> >> >> > Stephan  

Re: General Discussion about GPLness

[whywontyousue]

I gave you an example of a court immediately finding that modifying a 
running program is not allowed without permission. Creators of Non-gpl'd 
programs don't have permission to interact with a GPL'd work by 
modifying or extending it. If the principal stands for RealPlayer, it 
stands for Linux.

The GPL is a copyright license; law is the domain this is in.

In the domain of programming: you can do whatever you want. No technical 
measure is stopping you.

In the domain of what is "right and wrong" but excluding law, since you 
mentioned it and thus opened the door. Well... YHWH allows men to have 
female children as brides:
>>>> The Torah explicitly allows men to marry female children, including 
>>>> in cases of the rape (tahphas) of the girl child: Devarim chapter 
>>>> 22, verse 28. Key words: Na'ar (child (hebrew masoretic text)), 
>>>> Padia (child: padia+philos = paedophillia (greek septuagint)) Puella 
>>>> (young girl (latin vulgate))
>>>> Nachmanides points out that a child may be called na'ar from the 
>>>> moment he is born.
White idiots (such as the Linux programmers like Linus Trovalds) do not: 
White idiots worship white women. This is a problem for Free Software 
because White idiots (like Linus Trovalds) will do /anything/ to make 
money for "Muh whoite wuhman". That includes cowering in the face of 
being "blackballed" from the industry if they DARE enforce their 
copyrights.

In the domain of willpower: the linux copyright holders, atleast the 
programmers who are copyright holders, are NOT going to sue you for 
violating the copyright license permissions. They are pieces of shit who 
don't believe in "Copyleft". They believe in the BSD license, and have 
effectivly made Linux a BSD-type work since they never enforce the GPL.

OpenSourceSecurity (GRSecurity) is blatantly violating section 4 and 
section 6 of the linux copyright license and the Linux copyright-owning 
programmers would rather punish anyone who brings it up than sue the 
violators.

They are weak feckless people: more concerned about making money for 
"Muh wuhman" than anything else. Pay them no mind. They are scumbag 
idiots who won't know what they had until it's gone (it is).

I think that covers all the topics, right?

So, by law may you do what you suggested: No: it is a copyright 
violation in the US.
Is anyTHING going to stop you: no.
Is anyONE going to try to punish you for it: no.

Remember: linux copyright holders are either stupid white nerds (if they 
were smart they would have listened to their parents and become doctors 
and lawyers, and have tech as a fun hobby. Instead they are wageslaving 
for "MUUH WHOITE WUUHHMAN": same as any white idiot worker)
OR corporations who like the BSD license better.

You don't really have much to worry about. Because linux programmers are 
what they are. Sheep that thought themselves lions after RMS' win vs 
Cisco (that put the fear of the violating the copyright license (GPL in 
that case) for awhile).

On 2020-02-23 14:39, Stephan von Krawczynski wrote:
> Hello again,
> 
> at least you are beginning to sound a bit more like being able to 
> discuss
> something ;-)
> The thing about a lawyer (I learned you are) is that they judge the 
> world
> according to lawsuits. You can learn from the history of my home 
> country that
> laws and courts are no measure for moral and right behaviour.
> So my whole purpose of the thread is not to find out how the laws in 
> the US
> are judging something. It is all about how _we_ (we meaning all the 
> people
> contributing) are judging the issue. Do we really think that it is the 
> right
> thing to do to prevent people from feature-extending linux (and 
> distros) in
> general? The zfs case is special in that it is a simple "free license 
> clash".
> Which means all involved parties agree on the free and open software
> principle, only the licenses (i.e. paper) disagree on the usage of it - 
> to a
> certain extent. But lets not discuss legal details. In the end it all 
> comes up
> to this simple question: what do we really want with the project?
> Because in the end even you have to agree that there is a whole lot 
> more world
> outside the US. If people from other countries agree on something which 
> gives
> a better performance in some area, then the US would be the last ones 
> not to
> jump onto the train. Who can testify better than this project, not 
> being
> native-US.
> Do we think it is illegal to call GPL code from non-GPL code? Yes or 
> no,
> simple choice.
> Me, I don't think so. This is why I suggest we take down the barriers 
> and
> walls for interaction. It should be obvious by now that there will be 
> no
> non-gpl invasion taking place. Instead a non-ideological use of GPL may
> convince even more people that free and open software is a good concept 
> and
> adds benefit to the world and does not _harm_ technological progress.
> Given, not many people think about this from ground up before releasing
> software on linux. This is probably the only reason why you can buy 
> software
> for linux at all. And maybe, only maybe, the lawyers in your beloved 
> case
> where too dumb to turn this case around and ask why the gpl linux 
> software was
> forced to marry with the non-gpl real player (which is/was available
> for linux). According to this courts' point of view this must have been 
> equally
> illegal.
> I mean it extended the gpl software with a new feature without checking 
> for
> gpl compliance.
> As you can see the whole idea of the court in this case is broken. And 
> it
> seems only because noone asked the right questions.
> --
> Regards,
> Stephan
> 
> 
> 
> On Sun, 23 Feb 2020 12:56:13 +0000
> whywontyousue@waifu.club wrote:
> 
>> If you don't understand English, it will be difficult to get any 
>> points
>> across to you. I will try
>> 
>> In simple terms:
>> 1) Look up the court case "Universal City Studios Inc v Reimerdes,"
>> 
>> 2) In this case someone else' software was running at the same time as
>> the other persons software, and made changes, extensions _ONLY_ when
>> running. Just like a non-gpl'd (or gpl'd) module might make changes 
>> and
>> extensions.
>> 
>> 3) The court found this was obviously a modification of the Copyright
>> owners Work and barred in on summary judgement.
>> 
>> That is why you're not allowed to do as you wish with non-gpl'd 
>> modules:
>> the US Copyright Jurisprudence forbids marring a Copyright owners
>> _running_ _in_memory_ property against his wishes: it's a Copyright
>> violation.
>> 
>> That is the reason: people don't want to get sued. That is the ONLY
>> reason. That's it.
>> 
>> The thing is, the linux copyright owners are wimps and won't sue 
>> anyone
>> even for blatant infringement; so what is the conversation about?
>> 
>> It's like if you were in Russia, and you were copying DVDs. No one is
>> going to punish you for it: so what is there to discuss? The US
>> Copyright owners don't have the rocks to Invade Russia, Start a 
>> Nuclear
>> Winter, and DESTROY you for your Copyright Infringement in Russia.
>> 
>> JUST AS, the DOG LIKE Linux Copyright owners don't have the BALLS to
>> risk being blackballed from the programming industry for DEFENDING 
>> THEIR
>> COPYRIGHT.
>> 
>> The GPL ___IS___ dead. The FSF doesn't protect GCC copyright, and is
>> opposed to taking any action against blantant in-writing infringers
>> (OpenSourceSecurity (Grsecurity)) of GCC, just as the LINUX COMMUNITY 
>> is
>> OPPOSED to taking ANY action to defend its Copyrights and moves to
>> PUNISH those rightsholders who do.
>> 
>> > And another thing: court is for lawyers. Whenever the lawyers take over
>> > something they don't (want to) understand the end is near ...
>> 
>> I'm a lawyer and a programmer, got something to say?
>> 
>> 
>> > How about talking with real names?
>> Why would I do that? Tell me? What is in it for me?
>> I can stand here, in the forest, taking shots at your bullshit. Safe.
>> Secure. My words and their veracity the only measure.
>> 
>> But if I reveal the messanger; you'll just attack the messenger.
>> Tell me how _I_ benifit from telling YOU my name. Tell me.
>> Is it some sort of stupid werkin man white man bravado?
>> 
>> > I have no idea why you spam rms or bruce
>> > with this, as the question is all about _one_ project, namely
>> > linux-kernel.
>> 
>> You sent a message to the LKML "Hey why can't I violate the GPL? Let's
>> just do it!". IE: a licensing discussion. RMS and Bruce Perens, the
>> founder of the Free Software Movement and the Open Source Initiative 
>> are
>> relevant parties to the discussion.
>> 
>> 
>> > I'd suggest taking them off this topic again ...
>> You also suggested I reveal my identity on the internet...
>> 
>> 
>> On 2020-02-23 12:33, Stephan von Krawczynski wrote:
>> > Dear whoeveryouare,
>> >
>> > can you please state in a clearer form (more understandable to
>> > non-native
>> > english talkers) what your true opinion on the topic is?
>> > And in case you did not understand what I was saying, here is clearer
>> > form of
>> > my opinion:
>> >
>> > A kernel module with another license (be it whatsoever) is _no_
>> > modification
>> > of the kernel, but an extension of its features. If feature-extension
>> > is
>> > against the GPL (which I seriously doubt) then I would say "go back
>> > onto your
>> > trees". Because the human race and evolution is about little else than
>> > feature-extension.
>> >
>> > And another thing: court is for lawyers. Whenever the lawyers take over
>> > something they don't (want to) understand the end is near ...
>> >
>> > How about talking with real names? I have no idea why you spam rms or
>> > bruce
>> > with this, as the question is all about _one_ project, namely
>> > linux-kernel.
>> > I'd suggest taking them off this topic again ...
>> >
>> > --
>> > Regards,
>> > Stephan
>> >
>> >
>> >
>> > On Sun, 23 Feb 2020 11:03:56 +0000
>> > whywontyousue@waifu.club wrote:
>> >
>> >> Dear Stephan von Krawczynski;
>> >>
>> >> Universal City Studios Inc v Reimerdes, piece of shit.
>> >>
>> >> "[The court] reasoned that Ferret consumers who used the Ferret as a
>> >> plug-in to the Real Player altered the Real Player user interface by
>> >> adding the Snap search button or replacing it with the Stream box
>> >> search
>> >> engine button. The court concluded that the plaintiff raised
>> >> sufficently
>> >> serious questions going to the merits of its claims to warrant an
>> >> injunction pending trial"
>> >>
>> >> Want to violate the linux kernel copyright, you fucking piece of shit?
>> >> Yes you do. Yes modifying the running kernel with violating pieces is
>> >> copyright infringement, you fucking piece of shit. Yes you should be
>> >> sued. Just as Open Source Security (Grsecurity) should be sued for
>> >> their
>> >> violations (of section 4 and 6 of the linux kernel copyright license
>> >> (they're also violating the GCC copyrights too)).
>> >>
>> >> Will they be sued? Will you be sued? No: Linux copyright holders are
>> >> scared little wageslave worker bees. They aren't going to sue you;
>> >> sorry. Why are you even announcing you intent to violate the
>> >> copyright?
>> >> Why even give these dogs such intellectual deference?
>> >>
>> >> I wish OpenSourceSecurity would be sued. I wish you would be sued. But
>> >> linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate
>> >> them much more than I hate the violators. Complete Dogs. They could
>> >> move
>> >> from strenght to strenght, from victory to victory; but they're scared
>> >> for their "JEHRB"s. I have to say: white men are pathetic scum. If
>> >> Linux
>> >> was built by others there would rightfully be lawsuits.
>> >>
>> >>
>> >>
>> >> > Stephan von Krawczynski wrote:
>> >> > Hello all,
>> >> >
>> >> > you may have already heard about it or not (several times in the past),
>> >> > non-kernel devices run into a symbol export problem as soon as
>> >> > something is
>> >> > only exported GPL from the kernel.
>> >> > Currently there is a discussion regarding zfs using this call chain:
>> >> >
>> >> > vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) ->
>> >> > percpu_ref_tryget
>> >> > (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
>> >> >
>> >> > where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL
>> >> > exported)
>> >> > percpu_ref_tryget.
>> >> >
>> >> > That this popped up (again) made me think a bit more general about the
>> >> > issue.
>> >> > And I do wonder if this rather ideologic problem is on the right track
>> >> > currently. Because what the kernel tries to do with the export GPL
>> >> > symbol
>> >> > stuff is to prevent any other licensed software from _using_ it in
>> >> > _runtime_.
>> >> > It does not try to prevent use/copy of the source code inside another
>> >> > non-gpl
>> >> > project.
>> >> > And I do think that this is not the intention of GPL. If it were, then
>> >> > 100% of
>> >> > all mobile phones on this planet are illegal. All of them use GPL
>> >> > software
>> >> > from non-gpl software, be it kernel modules or apps - and I see no
>> >> > difference
>> >> > in the two. The constructed difference between kernel mode software and
>> >> > user-space software is pure ideology. Because during runtime everything
>> >> > is
>> >> > just call-chained.
>> >> > Which means if you fopen() a file in user-space it of course uses GPL
>> >> > symbols
>> >> > down in the chain somewhere. The contents of the opened file are not
>> >> > heaven-sent.
>> >> > If you/we follow the current completely ideology-driven GPL strategy
>> >> > then I am
>> >> > all for completely giving up this whole project. In real world you
>> >> > simply
>> >> > cannot use such a piece of software. The success of linux during the
>> >> > last
>> >> > years (i.e. decade) is not based on the pure GPL strategy, but on the
>> >> > successful interaction between linux and non-GPL software.
>> >> > Just think of the billions of smartphones all using a non-gpl firmware
>> >> > (underneath, and there is no GPL version at all), the kernel (with
>> >> > non-gpl
>> >> > modules) and apps (quite some of which are non-gpl).
>> >> > This is only one prominent example, but there are lots of others.
>> >> > In the end it all sums up to one simple question:
>> >> > Can one _use_ GPL software during runtime as a base for own projects of
>> >> > any
>> >> > license type or not? We are not talking about _copying_ gpl code, we
>> >> > are
>> >> > talking about runtime use.
>> >> > If runtime use is generally allowed, then the export gpl symbol stuff
>> >> > inside
>> >> > the kernel code is nonsense. Because to use the kernel you must be
>> >> > allowed to
>> >> > call it, no matter from where.
>> >> > Hit me.
>> >> >
>> >> > --
>> >> > Regards,
>> >> > Stephan

Re: General Discussion about GPLness

[Stephan von Krawczynski]

Hello again,

at least you are beginning to sound a bit more like being able to discuss
something ;-)
The thing about a lawyer (I learned you are) is that they judge the world
according to lawsuits. You can learn from the history of my home country that
laws and courts are no measure for moral and right behaviour.
So my whole purpose of the thread is not to find out how the laws in the US
are judging something. It is all about how _we_ (we meaning all the people
contributing) are judging the issue. Do we really think that it is the right
thing to do to prevent people from feature-extending linux (and distros) in
general? The zfs case is special in that it is a simple "free license clash".
Which means all involved parties agree on the free and open software
principle, only the licenses (i.e. paper) disagree on the usage of it - to a
certain extent. But lets not discuss legal details. In the end it all comes up
to this simple question: what do we really want with the project?
Because in the end even you have to agree that there is a whole lot more world
outside the US. If people from other countries agree on something which gives
a better performance in some area, then the US would be the last ones not to
jump onto the train. Who can testify better than this project, not being
native-US.
Do we think it is illegal to call GPL code from non-GPL code? Yes or no,
simple choice.
Me, I don't think so. This is why I suggest we take down the barriers and
walls for interaction. It should be obvious by now that there will be no
non-gpl invasion taking place. Instead a non-ideological use of GPL may
convince even more people that free and open software is a good concept and
adds benefit to the world and does not _harm_ technological progress.
Given, not many people think about this from ground up before releasing
software on linux. This is probably the only reason why you can buy software
for linux at all. And maybe, only maybe, the lawyers in your beloved case
where too dumb to turn this case around and ask why the gpl linux software was
forced to marry with the non-gpl real player (which is/was available
for linux). According to this courts' point of view this must have been equally
illegal.
I mean it extended the gpl software with a new feature without checking for
gpl compliance.
As you can see the whole idea of the court in this case is broken. And it
seems only because noone asked the right questions.
--
Regards,
Stephan



On Sun, 23 Feb 2020 12:56:13 +0000
whywontyousue@waifu.club wrote:

> If you don't understand English, it will be difficult to get any points 
> across to you. I will try
> 
> In simple terms:
> 1) Look up the court case "Universal City Studios Inc v Reimerdes,"
> 
> 2) In this case someone else' software was running at the same time as 
> the other persons software, and made changes, extensions _ONLY_ when 
> running. Just like a non-gpl'd (or gpl'd) module might make changes and 
> extensions.
> 
> 3) The court found this was obviously a modification of the Copyright 
> owners Work and barred in on summary judgement.
> 
> That is why you're not allowed to do as you wish with non-gpl'd modules: 
> the US Copyright Jurisprudence forbids marring a Copyright owners 
> _running_ _in_memory_ property against his wishes: it's a Copyright 
> violation.
> 
> That is the reason: people don't want to get sued. That is the ONLY 
> reason. That's it.
> 
> The thing is, the linux copyright owners are wimps and won't sue anyone 
> even for blatant infringement; so what is the conversation about?
> 
> It's like if you were in Russia, and you were copying DVDs. No one is 
> going to punish you for it: so what is there to discuss? The US 
> Copyright owners don't have the rocks to Invade Russia, Start a Nuclear 
> Winter, and DESTROY you for your Copyright Infringement in Russia.
> 
> JUST AS, the DOG LIKE Linux Copyright owners don't have the BALLS to 
> risk being blackballed from the programming industry for DEFENDING THEIR 
> COPYRIGHT.
> 
> The GPL ___IS___ dead. The FSF doesn't protect GCC copyright, and is 
> opposed to taking any action against blantant in-writing infringers 
> (OpenSourceSecurity (Grsecurity)) of GCC, just as the LINUX COMMUNITY is 
> OPPOSED to taking ANY action to defend its Copyrights and moves to 
> PUNISH those rightsholders who do.
> 
> > And another thing: court is for lawyers. Whenever the lawyers take over
> > something they don't (want to) understand the end is near ...  
> 
> I'm a lawyer and a programmer, got something to say?
> 
> 
> > How about talking with real names?  
> Why would I do that? Tell me? What is in it for me?
> I can stand here, in the forest, taking shots at your bullshit. Safe. 
> Secure. My words and their veracity the only measure.
> 
> But if I reveal the messanger; you'll just attack the messenger.
> Tell me how _I_ benifit from telling YOU my name. Tell me.
> Is it some sort of stupid werkin man white man bravado?
> 
> > I have no idea why you spam rms or bruce
> > with this, as the question is all about _one_ project, namely 
> > linux-kernel.  
> 
> You sent a message to the LKML "Hey why can't I violate the GPL? Let's 
> just do it!". IE: a licensing discussion. RMS and Bruce Perens, the 
> founder of the Free Software Movement and the Open Source Initiative are 
> relevant parties to the discussion.
> 
> 
> > I'd suggest taking them off this topic again ...  
> You also suggested I reveal my identity on the internet...
> 
> 
> On 2020-02-23 12:33, Stephan von Krawczynski wrote:
> > Dear whoeveryouare,
> > 
> > can you please state in a clearer form (more understandable to 
> > non-native
> > english talkers) what your true opinion on the topic is?
> > And in case you did not understand what I was saying, here is clearer 
> > form of
> > my opinion:
> > 
> > A kernel module with another license (be it whatsoever) is _no_ 
> > modification
> > of the kernel, but an extension of its features. If feature-extension 
> > is
> > against the GPL (which I seriously doubt) then I would say "go back 
> > onto your
> > trees". Because the human race and evolution is about little else than
> > feature-extension.
> > 
> > And another thing: court is for lawyers. Whenever the lawyers take over
> > something they don't (want to) understand the end is near ...
> > 
> > How about talking with real names? I have no idea why you spam rms or 
> > bruce
> > with this, as the question is all about _one_ project, namely 
> > linux-kernel.
> > I'd suggest taking them off this topic again ...
> > 
> > --
> > Regards,
> > Stephan
> > 
> > 
> > 
> > On Sun, 23 Feb 2020 11:03:56 +0000
> > whywontyousue@waifu.club wrote:
> >   
> >> Dear Stephan von Krawczynski;
> >> 
> >> Universal City Studios Inc v Reimerdes, piece of shit.
> >> 
> >> "[The court] reasoned that Ferret consumers who used the Ferret as a
> >> plug-in to the Real Player altered the Real Player user interface by
> >> adding the Snap search button or replacing it with the Stream box 
> >> search
> >> engine button. The court concluded that the plaintiff raised 
> >> sufficently
> >> serious questions going to the merits of its claims to warrant an
> >> injunction pending trial"
> >> 
> >> Want to violate the linux kernel copyright, you fucking piece of shit?
> >> Yes you do. Yes modifying the running kernel with violating pieces is
> >> copyright infringement, you fucking piece of shit. Yes you should be
> >> sued. Just as Open Source Security (Grsecurity) should be sued for 
> >> their
> >> violations (of section 4 and 6 of the linux kernel copyright license
> >> (they're also violating the GCC copyrights too)).
> >> 
> >> Will they be sued? Will you be sued? No: Linux copyright holders are
> >> scared little wageslave worker bees. They aren't going to sue you;
> >> sorry. Why are you even announcing you intent to violate the 
> >> copyright?
> >> Why even give these dogs such intellectual deference?
> >> 
> >> I wish OpenSourceSecurity would be sued. I wish you would be sued. But
> >> linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate
> >> them much more than I hate the violators. Complete Dogs. They could 
> >> move
> >> from strenght to strenght, from victory to victory; but they're scared
> >> for their "JEHRB"s. I have to say: white men are pathetic scum. If 
> >> Linux
> >> was built by others there would rightfully be lawsuits.
> >> 
> >> 
> >>   
> >> > Stephan von Krawczynski wrote:
> >> > Hello all,
> >> >
> >> > you may have already heard about it or not (several times in the past),
> >> > non-kernel devices run into a symbol export problem as soon as
> >> > something is
> >> > only exported GPL from the kernel.
> >> > Currently there is a discussion regarding zfs using this call chain:
> >> >
> >> > vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) ->
> >> > percpu_ref_tryget
> >> > (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
> >> >
> >> > where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL
> >> > exported)
> >> > percpu_ref_tryget.
> >> >
> >> > That this popped up (again) made me think a bit more general about the
> >> > issue.
> >> > And I do wonder if this rather ideologic problem is on the right track
> >> > currently. Because what the kernel tries to do with the export GPL
> >> > symbol
> >> > stuff is to prevent any other licensed software from _using_ it in
> >> > _runtime_.
> >> > It does not try to prevent use/copy of the source code inside another
> >> > non-gpl
> >> > project.
> >> > And I do think that this is not the intention of GPL. If it were, then
> >> > 100% of
> >> > all mobile phones on this planet are illegal. All of them use GPL
> >> > software
> >> > from non-gpl software, be it kernel modules or apps - and I see no
> >> > difference
> >> > in the two. The constructed difference between kernel mode software and
> >> > user-space software is pure ideology. Because during runtime everything
> >> > is
> >> > just call-chained.
> >> > Which means if you fopen() a file in user-space it of course uses GPL
> >> > symbols
> >> > down in the chain somewhere. The contents of the opened file are not
> >> > heaven-sent.
> >> > If you/we follow the current completely ideology-driven GPL strategy
> >> > then I am
> >> > all for completely giving up this whole project. In real world you
> >> > simply
> >> > cannot use such a piece of software. The success of linux during the
> >> > last
> >> > years (i.e. decade) is not based on the pure GPL strategy, but on the
> >> > successful interaction between linux and non-GPL software.
> >> > Just think of the billions of smartphones all using a non-gpl firmware
> >> > (underneath, and there is no GPL version at all), the kernel (with
> >> > non-gpl
> >> > modules) and apps (quite some of which are non-gpl).
> >> > This is only one prominent example, but there are lots of others.
> >> > In the end it all sums up to one simple question:
> >> > Can one _use_ GPL software during runtime as a base for own projects of
> >> > any
> >> > license type or not? We are not talking about _copying_ gpl code, we
> >> > are
> >> > talking about runtime use.
> >> > If runtime use is generally allowed, then the export gpl symbol stuff
> >> > inside
> >> > the kernel code is nonsense. Because to use the kernel you must be
> >> > allowed to
> >> > call it, no matter from where.
> >> > Hit me.
> >> >
> >> > --
> >> > Regards,
> >> > Stephan  

Re: General Discussion about GPLness

[whywontyousue]

If you don't understand English, it will be difficult to get any points 
across to you. I will try

In simple terms:
1) Look up the court case "Universal City Studios Inc v Reimerdes,"

2) In this case someone else' software was running at the same time as 
the other persons software, and made changes, extensions _ONLY_ when 
running. Just like a non-gpl'd (or gpl'd) module might make changes and 
extensions.

3) The court found this was obviously a modification of the Copyright 
owners Work and barred in on summary judgement.

That is why you're not allowed to do as you wish with non-gpl'd modules: 
the US Copyright Jurisprudence forbids marring a Copyright owners 
_running_ _in_memory_ property against his wishes: it's a Copyright 
violation.

That is the reason: people don't want to get sued. That is the ONLY 
reason. That's it.

The thing is, the linux copyright owners are wimps and won't sue anyone 
even for blatant infringement; so what is the conversation about?

It's like if you were in Russia, and you were copying DVDs. No one is 
going to punish you for it: so what is there to discuss? The US 
Copyright owners don't have the rocks to Invade Russia, Start a Nuclear 
Winter, and DESTROY you for your Copyright Infringement in Russia.

JUST AS, the DOG LIKE Linux Copyright owners don't have the BALLS to 
risk being blackballed from the programming industry for DEFENDING THEIR 
COPYRIGHT.

The GPL ___IS___ dead. The FSF doesn't protect GCC copyright, and is 
opposed to taking any action against blantant in-writing infringers 
(OpenSourceSecurity (Grsecurity)) of GCC, just as the LINUX COMMUNITY is 
OPPOSED to taking ANY action to defend its Copyrights and moves to 
PUNISH those rightsholders who do.

> And another thing: court is for lawyers. Whenever the lawyers take over
> something they don't (want to) understand the end is near ...

I'm a lawyer and a programmer, got something to say?


> How about talking with real names?
Why would I do that? Tell me? What is in it for me?
I can stand here, in the forest, taking shots at your bullshit. Safe. 
Secure. My words and their veracity the only measure.

But if I reveal the messanger; you'll just attack the messenger.
Tell me how _I_ benifit from telling YOU my name. Tell me.
Is it some sort of stupid werkin man white man bravado?

> I have no idea why you spam rms or bruce
> with this, as the question is all about _one_ project, namely 
> linux-kernel.

You sent a message to the LKML "Hey why can't I violate the GPL? Let's 
just do it!". IE: a licensing discussion. RMS and Bruce Perens, the 
founder of the Free Software Movement and the Open Source Initiative are 
relevant parties to the discussion.


> I'd suggest taking them off this topic again ...
You also suggested I reveal my identity on the internet...


On 2020-02-23 12:33, Stephan von Krawczynski wrote:
> Dear whoeveryouare,
> 
> can you please state in a clearer form (more understandable to 
> non-native
> english talkers) what your true opinion on the topic is?
> And in case you did not understand what I was saying, here is clearer 
> form of
> my opinion:
> 
> A kernel module with another license (be it whatsoever) is _no_ 
> modification
> of the kernel, but an extension of its features. If feature-extension 
> is
> against the GPL (which I seriously doubt) then I would say "go back 
> onto your
> trees". Because the human race and evolution is about little else than
> feature-extension.
> 
> And another thing: court is for lawyers. Whenever the lawyers take over
> something they don't (want to) understand the end is near ...
> 
> How about talking with real names? I have no idea why you spam rms or 
> bruce
> with this, as the question is all about _one_ project, namely 
> linux-kernel.
> I'd suggest taking them off this topic again ...
> 
> --
> Regards,
> Stephan
> 
> 
> 
> On Sun, 23 Feb 2020 11:03:56 +0000
> whywontyousue@waifu.club wrote:
> 
>> Dear Stephan von Krawczynski;
>> 
>> Universal City Studios Inc v Reimerdes, piece of shit.
>> 
>> "[The court] reasoned that Ferret consumers who used the Ferret as a
>> plug-in to the Real Player altered the Real Player user interface by
>> adding the Snap search button or replacing it with the Stream box 
>> search
>> engine button. The court concluded that the plaintiff raised 
>> sufficently
>> serious questions going to the merits of its claims to warrant an
>> injunction pending trial"
>> 
>> Want to violate the linux kernel copyright, you fucking piece of shit?
>> Yes you do. Yes modifying the running kernel with violating pieces is
>> copyright infringement, you fucking piece of shit. Yes you should be
>> sued. Just as Open Source Security (Grsecurity) should be sued for 
>> their
>> violations (of section 4 and 6 of the linux kernel copyright license
>> (they're also violating the GCC copyrights too)).
>> 
>> Will they be sued? Will you be sued? No: Linux copyright holders are
>> scared little wageslave worker bees. They aren't going to sue you;
>> sorry. Why are you even announcing you intent to violate the 
>> copyright?
>> Why even give these dogs such intellectual deference?
>> 
>> I wish OpenSourceSecurity would be sued. I wish you would be sued. But
>> linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate
>> them much more than I hate the violators. Complete Dogs. They could 
>> move
>> from strenght to strenght, from victory to victory; but they're scared
>> for their "JEHRB"s. I have to say: white men are pathetic scum. If 
>> Linux
>> was built by others there would rightfully be lawsuits.
>> 
>> 
>> 
>> > Stephan von Krawczynski wrote:
>> > Hello all,
>> >
>> > you may have already heard about it or not (several times in the past),
>> > non-kernel devices run into a symbol export problem as soon as
>> > something is
>> > only exported GPL from the kernel.
>> > Currently there is a discussion regarding zfs using this call chain:
>> >
>> > vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) ->
>> > percpu_ref_tryget
>> > (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
>> >
>> > where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL
>> > exported)
>> > percpu_ref_tryget.
>> >
>> > That this popped up (again) made me think a bit more general about the
>> > issue.
>> > And I do wonder if this rather ideologic problem is on the right track
>> > currently. Because what the kernel tries to do with the export GPL
>> > symbol
>> > stuff is to prevent any other licensed software from _using_ it in
>> > _runtime_.
>> > It does not try to prevent use/copy of the source code inside another
>> > non-gpl
>> > project.
>> > And I do think that this is not the intention of GPL. If it were, then
>> > 100% of
>> > all mobile phones on this planet are illegal. All of them use GPL
>> > software
>> > from non-gpl software, be it kernel modules or apps - and I see no
>> > difference
>> > in the two. The constructed difference between kernel mode software and
>> > user-space software is pure ideology. Because during runtime everything
>> > is
>> > just call-chained.
>> > Which means if you fopen() a file in user-space it of course uses GPL
>> > symbols
>> > down in the chain somewhere. The contents of the opened file are not
>> > heaven-sent.
>> > If you/we follow the current completely ideology-driven GPL strategy
>> > then I am
>> > all for completely giving up this whole project. In real world you
>> > simply
>> > cannot use such a piece of software. The success of linux during the
>> > last
>> > years (i.e. decade) is not based on the pure GPL strategy, but on the
>> > successful interaction between linux and non-GPL software.
>> > Just think of the billions of smartphones all using a non-gpl firmware
>> > (underneath, and there is no GPL version at all), the kernel (with
>> > non-gpl
>> > modules) and apps (quite some of which are non-gpl).
>> > This is only one prominent example, but there are lots of others.
>> > In the end it all sums up to one simple question:
>> > Can one _use_ GPL software during runtime as a base for own projects of
>> > any
>> > license type or not? We are not talking about _copying_ gpl code, we
>> > are
>> > talking about runtime use.
>> > If runtime use is generally allowed, then the export gpl symbol stuff
>> > inside
>> > the kernel code is nonsense. Because to use the kernel you must be
>> > allowed to
>> > call it, no matter from where.
>> > Hit me.
>> >
>> > --
>> > Regards,
>> > Stephan

Re: General Discussion about GPLness

[Stephan von Krawczynski]

Dear whoeveryouare,

can you please state in a clearer form (more understandable to non-native
english talkers) what your true opinion on the topic is?
And in case you did not understand what I was saying, here is clearer form of
my opinion:

A kernel module with another license (be it whatsoever) is _no_ modification
of the kernel, but an extension of its features. If feature-extension is
against the GPL (which I seriously doubt) then I would say "go back onto your
trees". Because the human race and evolution is about little else than
feature-extension.

And another thing: court is for lawyers. Whenever the lawyers take over
something they don't (want to) understand the end is near ...

How about talking with real names? I have no idea why you spam rms or bruce
with this, as the question is all about _one_ project, namely linux-kernel.
I'd suggest taking them off this topic again ...

--
Regards,
Stephan



On Sun, 23 Feb 2020 11:03:56 +0000
whywontyousue@waifu.club wrote:

> Dear Stephan von Krawczynski;
> 
> Universal City Studios Inc v Reimerdes, piece of shit.
> 
> "[The court] reasoned that Ferret consumers who used the Ferret as a 
> plug-in to the Real Player altered the Real Player user interface by 
> adding the Snap search button or replacing it with the Stream box search 
> engine button. The court concluded that the plaintiff raised sufficently 
> serious questions going to the merits of its claims to warrant an 
> injunction pending trial"
> 
> Want to violate the linux kernel copyright, you fucking piece of shit? 
> Yes you do. Yes modifying the running kernel with violating pieces is 
> copyright infringement, you fucking piece of shit. Yes you should be 
> sued. Just as Open Source Security (Grsecurity) should be sued for their 
> violations (of section 4 and 6 of the linux kernel copyright license 
> (they're also violating the GCC copyrights too)).
> 
> Will they be sued? Will you be sued? No: Linux copyright holders are 
> scared little wageslave worker bees. They aren't going to sue you; 
> sorry. Why are you even announcing you intent to violate the copyright? 
> Why even give these dogs such intellectual deference?
> 
> I wish OpenSourceSecurity would be sued. I wish you would be sued. But 
> linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate 
> them much more than I hate the violators. Complete Dogs. They could move 
> from strenght to strenght, from victory to victory; but they're scared 
> for their "JEHRB"s. I have to say: white men are pathetic scum. If Linux 
> was built by others there would rightfully be lawsuits.
> 
> 
> 
> > Stephan von Krawczynski wrote:
> > Hello all,
> > 
> > you may have already heard about it or not (several times in the past),
> > non-kernel devices run into a symbol export problem as soon as 
> > something is
> > only exported GPL from the kernel.
> > Currently there is a discussion regarding zfs using this call chain:
> > 
> > vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) -> 
> > percpu_ref_tryget
> > (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
> > 
> > where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL 
> > exported)
> > percpu_ref_tryget.
> > 
> > That this popped up (again) made me think a bit more general about the 
> > issue.
> > And I do wonder if this rather ideologic problem is on the right track
> > currently. Because what the kernel tries to do with the export GPL 
> > symbol
> > stuff is to prevent any other licensed software from _using_ it in 
> > _runtime_.
> > It does not try to prevent use/copy of the source code inside another 
> > non-gpl
> > project.
> > And I do think that this is not the intention of GPL. If it were, then 
> > 100% of
> > all mobile phones on this planet are illegal. All of them use GPL 
> > software
> > from non-gpl software, be it kernel modules or apps - and I see no 
> > difference
> > in the two. The constructed difference between kernel mode software and
> > user-space software is pure ideology. Because during runtime everything 
> > is
> > just call-chained.
> > Which means if you fopen() a file in user-space it of course uses GPL 
> > symbols
> > down in the chain somewhere. The contents of the opened file are not
> > heaven-sent.
> > If you/we follow the current completely ideology-driven GPL strategy 
> > then I am
> > all for completely giving up this whole project. In real world you 
> > simply
> > cannot use such a piece of software. The success of linux during the 
> > last
> > years (i.e. decade) is not based on the pure GPL strategy, but on the
> > successful interaction between linux and non-GPL software.
> > Just think of the billions of smartphones all using a non-gpl firmware
> > (underneath, and there is no GPL version at all), the kernel (with 
> > non-gpl
> > modules) and apps (quite some of which are non-gpl).
> > This is only one prominent example, but there are lots of others.
> > In the end it all sums up to one simple question:
> > Can one _use_ GPL software during runtime as a base for own projects of 
> > any
> > license type or not? We are not talking about _copying_ gpl code, we 
> > are
> > talking about runtime use.
> > If runtime use is generally allowed, then the export gpl symbol stuff 
> > inside
> > the kernel code is nonsense. Because to use the kernel you must be 
> > allowed to
> > call it, no matter from where.
> > Hit me.
> > 
> > --
> > Regards,
> > Stephan  

Re: General Discussion about GPLness

[whywontyousue]

Dear Stephan von Krawczynski;

Universal City Studios Inc v Reimerdes, piece of shit.

"[The court] reasoned that Ferret consumers who used the Ferret as a 
plug-in to the Real Player altered the Real Player user interface by 
adding the Snap search button or replacing it with the Stream box search 
engine button. The court concluded that the plaintiff raised sufficently 
serious questions going to the merits of its claims to warrant an 
injunction pending trial"

Want to violate the linux kernel copyright, you fucking piece of shit? 
Yes you do. Yes modifying the running kernel with violating pieces is 
copyright infringement, you fucking piece of shit. Yes you should be 
sued. Just as Open Source Security (Grsecurity) should be sued for their 
violations (of section 4 and 6 of the linux kernel copyright license 
(they're also violating the GCC copyrights too)).

Will they be sued? Will you be sued? No: Linux copyright holders are 
scared little wageslave worker bees. They aren't going to sue you; 
sorry. Why are you even announcing you intent to violate the copyright? 
Why even give these dogs such intellectual deference?

I wish OpenSourceSecurity would be sued. I wish you would be sued. But 
linux WERKIN MAHN wage slave piece of shit idiots won't do it: I hate 
them much more than I hate the violators. Complete Dogs. They could move 
from strenght to strenght, from victory to victory; but they're scared 
for their "JEHRB"s. I have to say: white men are pathetic scum. If Linux 
was built by others there would rightfully be lawsuits.



> Stephan von Krawczynski wrote:
> Hello all,
> 
> you may have already heard about it or not (several times in the past),
> non-kernel devices run into a symbol export problem as soon as 
> something is
> only exported GPL from the kernel.
> Currently there is a discussion regarding zfs using this call chain:
> 
> vdev_bio_associate_blkg (zfs) -> blkg_tryget (kernel) -> 
> percpu_ref_tryget
> (kernel) -> rcu_read_unlock (kernel) -> __rcu_read_unlock (kernel)
> 
> where __rcu_read_[lock|unlock] is a GPL symbol now used by (not GPL 
> exported)
> percpu_ref_tryget.
> 
> That this popped up (again) made me think a bit more general about the 
> issue.
> And I do wonder if this rather ideologic problem is on the right track
> currently. Because what the kernel tries to do with the export GPL 
> symbol
> stuff is to prevent any other licensed software from _using_ it in 
> _runtime_.
> It does not try to prevent use/copy of the source code inside another 
> non-gpl
> project.
> And I do think that this is not the intention of GPL. If it were, then 
> 100% of
> all mobile phones on this planet are illegal. All of them use GPL 
> software
> from non-gpl software, be it kernel modules or apps - and I see no 
> difference
> in the two. The constructed difference between kernel mode software and
> user-space software is pure ideology. Because during runtime everything 
> is
> just call-chained.
> Which means if you fopen() a file in user-space it of course uses GPL 
> symbols
> down in the chain somewhere. The contents of the opened file are not
> heaven-sent.
> If you/we follow the current completely ideology-driven GPL strategy 
> then I am
> all for completely giving up this whole project. In real world you 
> simply
> cannot use such a piece of software. The success of linux during the 
> last
> years (i.e. decade) is not based on the pure GPL strategy, but on the
> successful interaction between linux and non-GPL software.
> Just think of the billions of smartphones all using a non-gpl firmware
> (underneath, and there is no GPL version at all), the kernel (with 
> non-gpl
> modules) and apps (quite some of which are non-gpl).
> This is only one prominent example, but there are lots of others.
> In the end it all sums up to one simple question:
> Can one _use_ GPL software during runtime as a base for own projects of 
> any
> license type or not? We are not talking about _copying_ gpl code, we 
> are
> talking about runtime use.
> If runtime use is generally allowed, then the export gpl symbol stuff 
> inside
> the kernel code is nonsense. Because to use the kernel you must be 
> allowed to
> call it, no matter from where.
> Hit me.
> 
> --
> Regards,
> Stephan