And the point of this article is?

Story: Having offended everyone else in the world, Linus Torvalds calls own lawyers a 'nasty festering disease'Total Replies: 28
Author Content
kikinovak

Aug 27, 2016
2:27 AM EDT
I've been following Linus' rants over the years, and I must say I've enjoyed most of them. Linus cuts through the bullshit, and when he's pounding his verbal fist on the table, it's always for a good reason. I can't help the feeling this article is just clickbait.
nmset

Aug 27, 2016
4:23 AM EDT
If Linus was not Linus, Linux would not have been Linux, not at all, or some BigCorpNux stuff. I don't know anything about kernel programming, just a 2c opinion.

Clickbait ? Perhaps not, his new arguments seem to be : the hidden relation between lawyers and the 'system' in general(political, economic,...), the necessity of developers controlling their work on their own, the lack of understanding of the outside world to what the developers are doing.
mbaehrlxer

Aug 27, 2016
1:09 PM EDT
i don't know what's the point of the article, but i am glad it pointed to linus latest argument, because that actually seems to be interesting.

from what i gather (reading the actual discussion thread) the argument is that linus, greg and friends much prefer code contributions from previously violating companies that forcing those companies to to publish their code.

they argue that many companies, when threatened with lawyers, stop cooperating alltogether, linux/Free Software proponents in said companies loose their influence, and things generally turn sour.

lawyers and other kernel developers on the other hand argue that publishing code is more important, because it empowers the owners of the devices in question, and even though busybox may have never seen any contributions, things like OpenWRT would not have happened, if the violator was not forced to publish their code.

they further claim that linux/Free Software proponents in these companies thank them (i am guessing because the threats achieved what they themselves could not)

the difference between "forced to publish code" and "code contributions" is that published code ensures that we get everything needed to run our own version on a particular device (eg OpenWRT), at the expense of that company potentially deciding to stop contributing alltogether. whereas code contributions is when the company slowly starts to understand that working with the linux/Free Software community is beneficial, and so as time allows more and more code written by that company will be shared, but we can never be sure to get everything and exactly as needed to run on a particular device.

so each approach has its upside and downside.

the linus approach is good for developers, because they get more code contributions. the lawyers approach is good for users, because they get access to existing devices.

greetings, eMBee.
dotmatrix

Aug 27, 2016
2:21 PM EDT
@mbaehrlxer:

There's one hitch...

Using GPL code and not abiding by its license is illegal.

Linus just wants code; he's not interested in the politics. And that is a problem. It's a problem because without the politics of FOSS, OSS doesn't get to exist. There would be no Linux without the GPL and its 'viral' nature with the threat of lawsuits hanging over vendors who try to take it and hide. The kernel would have been scooped up by some vendor somewhere and all the mindshare and coding work would have gone into making 'Linux' a proprietary system.

The GPL is as relevant today as it was in 1991. Linus' dismissal of the importance of the GPL [both philosophical and legal] is always a disheartening thing to see... the world, any component of it, is not a huggy-feel-good world. Thieves come in all shapes, sizes, and clothing.
nmset

Aug 27, 2016
2:52 PM EDT
GPL violation (code confiscation) should be fought, we all agree with that. As for users, I think that they can't pretend to own, control and decide anything about work they have not done and work they cannot do ('the lawyers approach is good for users'). I can't help comparing the latter view as profoundly socialist, where workers are made to think they have a say about other's risks (shares <- money). Sorry, I sincerely don't mean any offense.

To get back to Linus's rant, it's the first time to my knowledge that he is yelling about another category of people than developpers. That's good because 'Thieves come in all shapes, sizes, and clothing.' It shows he has both eyes wide open and he is not disconnected from reality. His hammering of 'we developpers decide of our stuff' seems very sane to me ('among kernel developers. No lawyers present unless ...').
dotmatrix

Aug 27, 2016
8:57 PM EDT
@nmset:

I think I disagree... it's a bit difficult for me to clearly determine the stated position in your post...

My is:

The GPL means that there is no difference between 'users' and 'developers', as every user is a potential developer. And the GPL allows this to be the case.

It seems like:

You are partitioning 'developed code', of any license, into Intellectual Property units. And those units are each 'owned' by a particular developer or organization.

If this sounds like your meaning in your post, I think it's a misunderstanding of the purpose of GPL.

Last: If the 'viral' nature of the GPL is not what is wanted, don't use that license. If a particular 'thing' is licensed under the GPL, then the lawyers ensure that the rule of law is upheld. Therefore, the lawyers are 'good' for everyone 'users' and 'developers' alike, because these two categories define the same entity.

Last, Last: The GPL is not 'socialist'. I can have a copy of your code, and you can have a copy of your code and my code. The only resources that are 'lost' or 'shared' are the initial costs of writing one copy of the code. In many cases, a programmer is already being paid to write the code. If the code is 'open source' under any license, the shared resource is already being shared. So, GPL code is no more or less 'socialist' than any other open source license. The 'viral' nature comes into play when a third party wishes to incorporate GPL code in a given solution, only to discover that the license includes a requirement to 'share' code... Bypassing or ignoring this requirement is both illegal and unethical. The original programmer of the GPL code has made a bargain with society, and that bargain should be respected ... as respected as any other legally binding bargain.

Last, Last, Last: However, my mini-rant may not apply... and so... I don't know if it could be called a rant at all. Please feel free to ignore me.

Last, Last, Last, Last: I truly mean no offense at all. However, I am also truly annoyed that Linus uses the GPL, but then seems to broadly dismiss both the importance of the license as well as the implications of the license within [and of] his own work.
mbaehrlxer

Aug 27, 2016
10:54 PM EDT
@dotmatrix: the question is, how the license should be enforced and how the violation should be handled.

linus and greg advocate for education and slow transformation only.

lawyers use legal tools after the education approach fails.

both approaches are valid. there is no law that requires that licenses be enforced. this is all up to the copyright owner.

(btw: it is my understanding that using the term "illegal" to describe a copyright license violation is not correct. (@dotmatrix: this is not a criticism of your opinion, just a suggestion that a different term ought to be used)

"illegal" implies law-enforcement can persecute the violation, but, as far as i understand, that is not the case. a license violation is a broken contract between two parties. dealing with it is a civil matter between the two parties only.

a copyright owner is entirely within their right to ignore a license violation, and copyright law is written in such a way that the owners never loose their right just because they decided to not enforce their license. they can change their mind any time without loosing anything.)

greetings, eMBee.
nmset

Aug 28, 2016
3:41 AM EDT
>You are partitioning 'developed code', of any license, into Intellectual Property units. And those units are each 'owned' by a particular developer or organization.

I was not thinking about that at all. I'm trying to stick to the substance of the article, where Linus is only saying : the kernel developpement is a matter for developpers only. He does not want third parties to influence the project at all. In my first post, I'm just saying the project would be an uncontrollable mess if it dit not have such a commander. My second post means : users are just users.

jdixon

Aug 28, 2016
6:56 AM EDT
> ...both approaches are valid. there is no law that requires that licenses be enforced. this is all up to the copyright owner.

There is a legal principle that non-enforcement of a right or law eventually invalidates same.

OK, LXer is refusing to take url's to legal-dictionary.thefreedictionary.com. I have no idea why.

Anyway, for examples look up adverse possession and selective enforcement on that site.

While to my knowledge this principle has never been applied to software licenses, it could be. It has been applied to trademarks, Linoleum being one example. At the very least it would be an interesting court case.
mbaehrlxer

Aug 28, 2016
11:53 AM EDT
Quoting:There is a legal principle that non-enforcement of a right or law eventually invalidates same.


i am aware of that principle, but it doesn't apply to copyright.

http://www.hypergridbusiness.com/2011/08/how-to-lose-your-co... https://www.plagiarismtoday.com/stopping-internet-plagiarism...

greetings, eMBee.
jdixon

Aug 28, 2016
1:19 PM EDT
> i am aware of that principle, but it doesn't apply to copyright.

The GPL is a software license, not a copyright. It is based on copyright, but the two aren't the same thing legally. A software license can be found invalid without affecting the copyright. As I said, to my knowledge it's never been brought up in court.
gus3

Aug 28, 2016
2:38 PM EDT
Oblig disclaimer: IANAL. But it appears to me that this conversation is missing an important point.

Users are guaranteed rights under the GPL. Developers are likewise guaranteed rights. But neither users nor developers can violate the GPL.

Violating the GPL consists in not providing source code, or making the source code readily available on request, for users and developers. That violation can be committed only by distributors of GPL-licensed products. Not abiding by that legal responsibility is the only way to violate the GPL, and that can occur only with an act of distribution (called "conveyance" in the GPLv3).

This has always been my understanding of rights vs. obligations under the GPL.
BernardSwiss

Aug 28, 2016
4:37 PM EDT
Linus Torvalds seems to figure that as long as he, himself, and associated developers can see the code, that's good enough for him. He doesn't really believe that ordinary "users" deserve the benefits conferred by the GPL... he has said quite baldly that the "users" are idiots, and access to the code is for all practical purposes wasted on them -- and he's not been shy about saying as much in pretty plain language.

So Linus is focused on the code, and sees fuzzy legal matters (especially GPL/Free Software "philosophizing") as no more than needless complication and largely pointless side issues, that distract from the work at hand..

It's a bit of a blind spot with him.

Mathew Garrett (for an example on the other side, since his name was brought up) can also demonstrate an equally large blind spot on certain matters (hence his "SJW" reputation, even though there is a large chunk of genuine problems that needs to be addressed -- and arguably isn't being addressed appropriately).

They are both very smart guys who have a good chunk of solid reason, largely supported by experience, on their side -- and have trouble seeing, in particular contexts, where that solidity falls off to a superficial facade where the facets of their understanding extend into some particular larger sphere.

Pretty much all of us suffer this tendency (it tends to strike precisely where we believe we are "experts", or have extensive experience, or at least believe ourselves better informed than we actually are).

The solution is simple: don't deify our leaders -- technical or otherwise. And remember that even our own fields of expertise are likely full of nooks and crannies and hidden complexities and blocked perspectives we might not even be aware of. Nobody knows it all.
nmset

Aug 28, 2016
5:09 PM EDT
>don't deify our leaders

For me, I am happy the Linux project has a leader, imperfect in his language, perfect in his management, who stood the test of time with hundreds of contributors, whose project did not fall apart with chaotic contribution, without which and without whom, I would have an iPhone in my belt pocket, a Windows home server and laptop where I would not be typing these words.
dotmatrix

Aug 28, 2016
9:11 PM EDT
@nmset:

What if Linus decided that the BSD license was a better fit for his kernel?

Remember in 1991, Macs were not *nix machines. Perhaps, Apple would have grabbed the Linux kernel instead of the Mach kernel. Perhaps then, coding mindshare would have left for the shiny colored Apples rather than stayed with the monochrome GNUs [logos, not displays].

The license matters. Apple didn't grab the Linux kernel... probably because of the GPL which would have required them to abide by the restrictive license agreement and release certain portions of their OS under the GPL.

So... a lot of speculation... but... It's possible that if Linus did not choose the GPL, you would not have the Linux kernel you do today. However, you might be writing about the GNU OS and the Hurd kernel.
linuxwriter

Aug 28, 2016
10:51 PM EDT
Linus Torvalds is not a publicity seeker. Matthew Garrett is nothing but.

Torvalds has genuine achievements of which he can speak. Garrett claims to be a security professional but the last time there was a remote vulnerability in CoreOS (where he works) he was responsible for it, a fact that he neatly hid when writing the company's disclosure note.

Please do not conflate the two; one is an engineer who operates in the true spirit of a scientist. The other is someone who is always trying to bignote themselves.
mbaehrlxer

Aug 28, 2016
11:02 PM EDT
ironically, steven jobs did choose objective-c which was built on top of the GCC, and which prevented them from closing its source.

@jdixon: i don't understand what copyright enforcement has to do with the validity of the GPL.

"even if you don't enforce copyright, you don't lose it" is a statement of fact.

are you saying that, by not enforcing the GPL, the GPL loses its validity? that statement does not compute. either the GPL is valid, then it can be enforced at any time, as long as the copyright applies, or the GPL is not valid, and then it would be unenforcable to begin with.

according to wikipedia, the GPL has been tested and clearly found valid in germany: http://gpl-violations.org/news/20060922-dlink-judgement_fran...

and at least one of the busybox cases ended with a judgement to enforce the GPL: https://en.wikipedia.org/wiki/BusyBox#GPL_lawsuits http://www.groklaw.net/article.php?story=20100803132055210

greetings, eMBee.
mbaehrlxer

Aug 28, 2016
11:21 PM EDT
@BernardSwiss: can you please elaborate which blind spots you see in mathew garret?

(obviously, just because i don't see the blind spots in mathew, does not mean they are not there, it could just be my blind spot :-)

greetings, eMBee.
jdixon

Aug 29, 2016
1:31 AM EDT
> are you saying that, by not enforcing the GPL, the GPL loses its validity?

I'm saying that by not enforcing the license there is a possibility that they lose the ability to enforce the license.

If a license breaker can show a pattern of the license not being enforced in the past, he can possibly argue successfully that there is no reason it should be enforced it in his case either. But then, IANAL.
nmset

Aug 29, 2016
2:39 AM EDT
@jdixon

>What if Linus decided that the BSD license was a better fit for his kernel?

You are totally right all throughout that post. I did not talk about licenses in this discussion because it's sidebranching from the object of the article (Linus's view on who discuss about kernel developpement). The GPL IS the right license for such a project, which would probably not have the same success otherwise.
mbaehrlxer

Aug 29, 2016
10:40 AM EDT
Quoting:I'm saying that by not enforcing the license there is a possibility that they lose the ability to enforce the license. If a license breaker can show a pattern of the license not being enforced in the past, he can possibly argue successfully that there is no reason it should be enforced it in his case either.


nope, he can not. as the references i have given clearly explain:
Quoting:unlike trademarks, which do have to be defended, there is nothing the precludes you from enforcing your copyrights at a later date
Quoting: You also can’t lose your copyright if you don’t defend it. You can ignore violators for years — then come back later, and sue them all. Widespread infringement could reduce the market value of your work, and the courts could award you a smaller settlement, but the copyright would still be yours.


if that is not enough, then we'll have to research the actual text of the law in question, or ask a lawyer.

as the quote says, at best, enforcing your copyright very late, limits the amount of damages you can claim. however, the GPL is not about damages, but about sharing code, and as a GPL violator looses any rights given by the GPL, the only option to get those rights back is to honor the license.

the only case you might get away with a violation is if you violate the license once, then stop using the code, and hope that when you get sued 10 years down the road that the damages by then are so small that they don't eat up the profit you made on that product.

but none of this affects the validity and enforcability of the GPL. no matter how far in the past the license violation happened, you can be found guilty of violating the GPL. this remains true until the copyright expires.

greetings, eMBee.
jdixon

Aug 29, 2016
11:29 AM EDT
> nope, he can not. as the references i have given clearly explain:

Both of which deal with copyright. Again, the GPL is not a copyright. It is a license. They are not the same. A license can be rendered invalid or ineffective without affecting the underlying copyright.

> if that is not enough, then we'll have to research the actual text of the law in question...

It's not a specific law in most cases, but rather goes back to the British Common Law principle of equal standing under the law. People have to be treated the same for committing the same infraction. You can't just pick and choose who you enforce the law against or what punishment you give them. In the US, this has been codified in the 14th amendment as "no state shall deny to any person within its jurisdiction the equal protection of the laws". And while it's a private entity bringing the case, not the state, it's still the state issuing the penalty. So it has to be applied equally.

So if you haven't enforced the GPL against numerous other infractions, but do against a specific one, that's unequal enforcement, and violates the principle.

Would a judge accept this argument? I have no idea. But under US law and other British Common Law derived systems, it's a reasonable argument to make if you're a defendant.

> ...or ask a lawyer.

That's always the case with legal matters. Or ask several, and get several different answers. But why not simply play it safe and enforce the GPL when clear violations are brought to your attention? That nullifies the question completely. It's not enforcing the GPL against known violators that's problematic.
skelband

Aug 29, 2016
1:05 PM EDT
Those upholding principles and the law are often derided as unreasonable extremists but we must have them. Many people still think RMS is a loon but you need these people to guide the rest of us.

Without the GPL, Linux would certainly not exist as it does now (or at all).

And I don't get the view that many in industry have about the GPL in terms of its freedoms. How many proprietary licenses allow you to see the source, modify it, distribute your mods, sell it (freely)? All it asks is that if you use it, you respect the same freedoms with your code. You don't like it? Don't use it.
mbaehrlxer

Aug 29, 2016
3:07 PM EDT
@jdixon: i am not following you.

the GPL is a copyright license. a violation of the GPL IS a copyright violation.

independent from copyright, for which the law clearly states that i am allowed to selectively enforce it at will, in civil matters, i can choose who i enforce the law against. if someone trespasses my property, then i decide if i call the police or not. if i tolerate the neighbor kids using my pool even when i am not at home, that doesn't imply that i now have to allow everyone else to use it too. i can still get them for trespassing. i am effectively giving the neighbors permission, so they are not actually trespassing.

here is a thought: the reason why equal enforcement matters, is because the law applies equally to everyone. however, i am not required to give everyone the same copyright license. if i decide not to pursue a particular violation, i am effectively giving them a different license. (a license that effectively says: you are allowed to do this until i decide to stop you)

greetings, eMBee.
mbaehrlxer

Aug 29, 2016
3:10 PM EDT
to turn the discussion back to the original argument:

Quoting:It's not enforcing the GPL against known violators that's problematic


this is exactly what linus thinks is problematic. he believes that the GPL should not be enforced, but that violators should be educated until they stop violating the GPL on their own accord.

greetings, eMBee.
jdixon

Aug 29, 2016
4:56 PM EDT
> the GPL is a copyright license. a violation of the GPL IS a copyright violation.

Because the GPL is built on top of copyright. But technically I believe it's a primarily a usage license.

Regardless, I've make my hypothetical case to the best of my ability, which is that if the GPL isn't enforced against known violators, we run the risk of losing the ability to enforce it at all..
BernardSwiss

Aug 29, 2016
10:29 PM EDT
Quoting:to turn the discussion back to the original argument:

Quoting: It's not enforcing the GPL against known violators that's problematic

this is exactly what linus thinks is problematic. he believes that the GPL should not be enforced, but that violators should be educated until they stop violating the GPL on their own accord.

greetings, eMBee.


AFAIK: (oddly enough) the very same people that Linus is bitching about (Bradley Kuhn and SFC, among others) believe in -- and explicitly recommend -- the educational approach, remediation first, and all that... and reserve taking actual legal action against GPL violators for those particular, recalcitrant actors who refuse to "mend their ways" / decline to get into compliance with the license.
jdixon

Aug 30, 2016
2:46 AM EDT
To my knowledge you are correct, BernardSwiss. They do in fact recommend education first. Of course, giving the entity causing the injury a chance to make things right before going to court is also a sound and widely recognized legal practice.
mbaehrlxer

Aug 30, 2016
12:44 PM EDT
indeed they do try education first, until they reach a point where the conclude that it doesn't work and switch to the legal approach, whereas linus basically insists to never ever use the legal approach.

greetings, eMBee.

Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]

Becoming a member of LXer is easy and free. Join Us!