Submitted by “The Samba Team and seven kernel hackers have come together with Software Freedom Conservancy to help efforts to ensure compliance with the GPL by those who implement Linux and other GPL software. Richard Hillesley talked to Bradley Kuhn of Software Freedom Conservancy, Jeremy Allison of Samba, and Matthew Garrett, who works in his spare time with the GPL Compliance Project for Linux Developers.”
I’m sure the FSA would say the same thing. How much longer before companies start getting raided by the GPL police.
Before you hit the thumbs down button, I’m not picking on the GPL folks here. That being said, I don’t want to give validity to the GPL license because by respecting that license, you also by extension give validity to every other software license in existence, including the ones that demand you pay a fee for every machine you install the software on. Picking and choosing which licenses you will respect is a bit hypocritical to me, because whether a license is ‘good’ or ‘bad’ depends on which vantage point you are looking at it from.
From my point of view, once you put anything digital out in the wild, you are basically putting it in the public domain and don’t have a claim as to who can copy/run it and when/where, unless you can enforce such a thing using technological means. In other words, my current line of thinking states that NO software licenses are valid, including the GPL. I don’t disagree that the folks who made the GPL had good intentions, but it’s just dumb to expect end users to follow a certain set of rules, when there’s nothing inherent in the technology that dictates they have to follow those rules. Imagine if you came across a piece of software that said you could only use the software every other Tuesday, and only from a squatting position? Who’s going to follow that, and how is it that we arbitrarily decide which parts of a license we will follow and which ones we won’t?
Edited 2012-06-02 02:06 UTC
wow. First off, the GPL is not a software license. It has absolutely nothing to say about how or where you use the software, or how many machines you install it on. It is simply a copyright license, detailing what you must do to distribute (not create) copies to third parties. If you discount it, then you basically wipe out all copyright, whether digital or physical.
You reasoning is a bit odd. You claim since some licenses are bad and some are good, we should so away with them all. Thats anarchy. Now if you are advocating anarchy then fine. But its not a reasonable or realistic view to take of the world. Even the worst third world crap holes have some rules.
Agreed. I’ve always found it strange that so many Windows installers give the GPL the same presentation as a EULA when they have about as much in common as an elephant and a gecko.
Yeah, cause the first thing I read in a GPL licensed software is “THIS SOFTWARE COMES WITHOUT GUARANTEE”, <sarcasm>Encouraging, right? </sarcasm>
actually pretty much every piece of software has that clause attached to it. Its have nothing to do with open source or the GPL.
That might very well be OP’s point. As far as I understand it, he is saying that any digital restriction is meaningless without enforcement.
Really? According to the Wikipedia article on the GPL:
http://en.wikipedia.org/wiki/GNU_General_Public_License
The GNU General Public License (GNU GPL or simply GPL) is the most widely used free software license. It was originally written by Richard Stallman for the GNU Project.
It is a software license… equally as valid (or invalid) as every other software license on the planet.
On the contrary, it has plenty to say about these matters. It basically says you can run it wherever you want and install it on as many machines as you want. Just because it is much less restrictive than most other software licenses doesn’t change what it is.
For digital, yes. But physical objects don’t apply, because most of them are not copyable. But when (not if) somebody creates a Star Trek-style replicator and you can clone a loaf of bread and infinite amount of times, are we then going to set up laws that says people can’t copy a loaf of bread? I am simply pointing out the lunacy of trying to use the legal system to prevent people from copying something that is infinitely copyable. Mind you, I’m speaking on pragmatic terms, not trying to give some sort of moral justification for piracy.
What I am saying is that if we accept one type of software license, we must by extension accept them ALL. As long as we’re allowing developers to make the rules, we can’t say that some developers are allowed to dictate the terms by wich their software can be used and/or distributed and some aren’t allowed to do this, just because some developers set up terms that are more restrictive than you would like. Thus I think it is logical to not allow developers to make the rules anymore, since most of their rules end up being detrimental to end-users anyway. In essence, we’ve set up a system where developers have 100% of the control, and it ain’t right. It’s like being sold a car, and then being told where you’re allowed to drive it. Sure, some dealers may be less restrictive about this than others, and some might even say ‘drive it wherever the hell you want.’ I’d just assume not let dealers make these decisions in the first place.
Edited 2012-06-02 08:25 UTC
If.
The world already has replicators: they’re called “life” or “civilisation” – and it is unclear, an open question, if very much more efficient ones are practical in our universe (something like this would have insane evolutionary advantage, hence would almost certainly show up & take over already: if not within our biosphere – obviously not, for now – then at least within likely billions other biospheres in the universe, spreading and visibly transforming it …possibly reaching and ~consuming us by now)
That loaf of bread is already basically replicated.
(generally, those are tools of storytelling; and cargo cults, modern mythologies, really – in those we always wished for something silly to be true; plus here, silliness of lack of real implications of such tech – it would massively transform the world, much more than “instant PO box” model in Star Trek, and it would do so parallel to its gradual improvements, not immediately such advanced; oh yeah, the route progress was silly – in TOS they synthesised food, which is among the harder things; and transporters in TOS were really better replicators than replicators in TNG, and also without real implications)
Oh, and “infinite” is a very large number in a universe of finite energy, resources, with a finite time to its heat death.
GPL itself has no real relevance on non-GPL code, doesn’t say or allows/disallows anything about it.
But yeah, we can choose what licenses we personally accept, which we use and which not, what’s so shocking about that? Are you proposing that the only healthy non-PD situation is when we all strive to use every possible type of license?
Well the thing about GPL is that it pushes the stakes in favour of users, general population…
It very definitely is, then. What it is not is an End User License Agreement (EULA).
Much better worded that my original.
Personally I try to take the opposite approach and try to respect all licenses.It seems to me that the producer of software has a right to determine its license, if the software has an offensive license or EULA then I try not to use that software (although an EULA and Copyright licenses are not equivalent).
This is one of the reasons I choose free software, if more people did this rather than pirating software more people would use free software. The likes of Microsoft know this, which is why they tolerate limited piracy, especially in the developing economies.
Agreed. I do the same. In fact, I go far enough that the only closed-source non-games on my system are Opera (for testing sites I write), Flash (Gnash isn’t quite there yet), my nVidia drivers, Skype (normally left turned off), and my BIOS. (I’ve been lazy. I’ll probably specifically source motherboards that’ll work with CoreBoot once my only other option is UEFI.)
(Games get a pass as long as they’re DRM-free and I didn’t pirate them because I have yet to see sufficient evidence for open-source development being able to produce “disposable code” like games where they can’t start small and just refine it over the course of a decade.)
Back before I switched to Lubuntu for lack of time, I was on Gentoo using Portage 2.2 alphas and my system was configured so Portage only allowed libre-licensed packages without prompting me to whitelist the licenses individually. (You’d be surprised how many fonts and other supplementary files have little clauses that make them freely-redistributable and freely-usable but not free)
Edited 2012-06-02 09:55 UTC
??… (my personal, internal EN interpreter seemingly failed to parse it, and for some reason I’m curious about the meaning of that quote ;p )
zima,
“??… (my personal, internal EN interpreter seemingly failed to parse it, and for some reason I’m curious about the meaning of that quote ;p )”
Try sticking this through your parser! 🙂
http://ornacle.com/article/20120604
It is generated completely automatically from statistical analysis of word combinations in the news from any given day. As piss poor as the grammar is, it’s not half bad for a perl script using 20K of english training data in the form of news. You might be wondering “why the hell?” Well it was to help probe search engine algorithms.
Today is the first time I’ve seen an infinite cycle. Not that I monitor it any more, but it’s nice to see it’s still working!
“In fact, quite low by comparison, down and down and down and down and down and down and down and down. Wreckage could be seen in union jack ties and those who did not transmitted live by the kabul bank scandal prompted the international monetary fund to temporarily suspend hundreds of thousands of people enjoying the celebrations.”
Edited 2012-06-05 01:48 UTC
Curious, I think it’s more coherent to me… ;p
Makes me wonder how such “news” would work out in my first language (PL; probably not too well – complex declension, conjugation, grammar in general)
Also, reminded me about http://www.kyon.pl/img/14315,4chan,text,mindf–k,grammar,lol,.html (sometimes I wonder how many bots are there already, in forums or comments) and http://megahal.alioth.debian.org/ (some of the quotes are brilliant-scary)
Edited 2012-06-05 06:23 UTC
Haha, yea I sometimes get that feeling when I need tech support sometimes. Although those are due to language barriers.
Speaking of which…one can use online translators for good effect.
[ Translate EN->AR->EN ]
I’ve been using the preview release 8 Windows where came out, almost exclusively (except to work, because I am going to rely on untested software and unfinished for obviously not). I already know you can enter Metro in my 11.6 “zinbok, but things don’t look 24” desktop, pink. Here is the evidence photographer of the most pressing issues that I run to, five proposals for addressing them. Instead of just complaining, let’s constructive.
All these issues can be addressed relatively easily — and when you do this, it will become hell lot of useful Metro for people who want to do actual work on their computers. People often refer to the usage data collected at Microsoft, says something like “10 per cent only Windows users have more than x number of open Windows. Crazy thing-10% of all Windows users are still around 100 million.
My list of suggestions is short, easy to implement, and is, in any way, affect design ideals behind Metro. In fact, I would say this could be implemented even before it goes Windows 8 RTM.
[ /translate ]
Wow, I’m surprised at how little this impacted Thom’s writing 🙂
Reminded me about another thing…
http://en.wikipedia.org/wiki/AI_winter#Machine_translation_and_the_…
(though I’d say those examples don’t so much show problems – but are more like eerily brilliant, in context ;p )
Also, about the possibility of this ornacle thing in my language – now that I thought about it a bit more, I suspect it still would be more time-rewarding than following usual news…
Edited 2012-06-06 03:53 UTC
As a wonderful line on the unlicense.org website says – “Life’s too short – let’s get back to coding.”
I wonder if the world would have been worse-off if Linux and gcc had been released as “public domain”.
Anyone think so? I don’t.
The software would still be out there (just as SQLite is – the best-known PD software). Has SQLite suffered from being PD? No, it hasn’t. It is used in a huge number of products, and in many of those products it is used BECAUSE it is PD. (Because it is good as well, but anyway…)
The reason I really like PD software is because it is the ultimate in freedom. Do what you want with it.
Use it as-is. Fork it and make a GPL/BSD/MIT-licensed version. Include it with any other software.
Most important of all, IMO – **learn** from it, and you don’t need to worry about then including what you have learned in your own software.
Last but not least, PD software comes with **no egos**. The vast majority of users of it will “do the right thing” anyway, and acknowledge the authors, but there is no requirement to do so.
PD software really is the ultimate in both generosity and freedom.
“Free” Software Foundation? I don’t think so.
“Free” as in “no cost, but you’d better follow these rules or we will send the GPL squad after you.”
Edited 2012-06-02 07:47 UTC
Considering that free software offers you the use of copyrighted code with more rights than standard copyright allows, following a distribution license is a reasonable expectation.
Otherwise, you can, I don’t know, write your own damn code. Must be pretty easy if so many people are willing to give it away as public domain, right?
If there was something else to assure the giant companies that their competitors will not be able to use their work without releasing their own modifications, then yes.
Otherwise, don’t even think that the likes of Oracle and IBM would shy away from creating a closed fork. Oracle already closed off Solaris – that is a fact.
Well, so what? They can have their closed fork. Anyone else can create a free fork and the original creator can continue his own version.
That’s what you get if you make it public domain. If that’s now what you want then just don’t, make is closed and sell it. GPL, BSD, whatever it.
In the old days there was lots of public domain software. Demos, games, serious software. A lot of high quality stuff.
The “public domain” works of olden days were usually distributed in binary form, without source.
Yeah, that’s right. For some reason it didn’t seem anyone cared about the source code.
P.D. stuff was free and legal software, that alone was enough to make most people happy.
Most P.D. stuff I got was on floppies stuck on magazines (which caused immense cover mutilation when removing the floppy). As we didn’t have the Internet to download from this was the best and quickest way to get hold of new and recent software.
Are you not aware who writes most opensource software these days? And why they are OK at actually contributing to GPL’d projects?
Edited 2012-06-03 02:13 UTC
No, I´m not. I guess because like most people I search for what I need, add the word “free” to the search term and download it. If it works it works, if it doesn´t I delete it. Who wrote it, why and under what license doesn´t matter to me.
In the old days it wasn´t so easy to get loads of serious software. It was much easier to give more attention to who made something when you got hold of it. If their program was included with a magazine you´d often find an interview or at least some info on the person.
Now we live in a fast connected world with loads of good and awful software. You need something, you have it a few seconds later, it doesn´t work, it´s gone even faster.
It’s easy to rewrite history, but it’s also SOOOOOOOOOOOOOOOOOOOOOOOOO highly likely that if that were the case there would be no linux today.
there might even not be a webkit, and many others.
While theres a HUGE wave of FUD against the GPL those days, because the bay area figured they make money more easily off BSD licenses and other permissive licenses, everyone who joined Linux joined it for the goddamn GPL, that is, back in the days. (now that linux is big you join it for various reasons).
You know, BSD *was* already there in various forms and the kernel(s) more advanced than Linux (which had, well, nothing).
I couldn’t agree more. I’ve never cared for the GPL, and have always considered it akin to forced communism. Sure, you may be free to redistribute the software all you want, but when it comes to actually developing with it, you’re not nearly so free when considering the possibility of using GPL’d code or libraries. The FSF is as hypocritical as the US government when talking about freedom. What they both mean is, sure you’re free to do what you want as long as you obey our rules. They allow you a limited subset of freedom which, to my mind, is not freedom at all.
To those who say that companies like Oracle could’ve taken a public domain Linux or GCC and made it proprietary: you’re absolutely right. They would be free to do that, just as the original developers would be free to maintain their own open source version. In the end, everyone would have been better off and, if the proprietary Linux worked better (say by having a stable driver interface), that might have created a reason for the open source developers to strive for improvement in that area. Competition drives innovation, pure and simple.
How so? You realise that the reason so many companies are collaboratively developing Linux is because they are legally bound to share all their enhancements with eachother if they want to be able to distribute them.
If that was not the case then there would be a large incentive for these companies to keep their enhancements proprietary so as to gain a competitive edge on the others, which would result in the companies only submitting enhancements which they felt would give them no competive benefits if kept to themselves, thus hampering the overall progress of the project.
The reason the Linux devs doesn’t make it easy to keep proprietary out of tree drivers is because they want to enforce the creation of open source in-tree drivers.
This has in turn had a great impact on the availability of open source drivers as many companies find it’s much easier to simply submit either a open driver or the necessary specs to make one instead of maintaining their own out-of-tree proprietary driver.
This also benefits other systems than Linux as these open source drivers are often dual-licenced and in the case they are not it’s much easier to reverse-engineer from full driver source code than it is from black box hardware.
Looking past the driver interface, if the ‘proprietary Linux’ did things better and the ‘open Linux’ wanted those enhancements they would then have to duplicate those efforts, how can you see this as an improvement on what we have now, where thanks to the GPL these enhancements directly make their way into the project without any need for duplicated effort?
There’s really no competition when one of the participants can take anything the other does and not have to give anything back themselves.
FreeBSD has turned out to be so much better as a desktop system since Apple decided to use it as part of their OSX?
Hello, have you heard of WebKit?
Actually if you remember correctly, during the browser wars we went from.
http://img.tfd.com/cde/_MOSAIC.GIF
To
http://en.wikipedia.org/wiki/File:Internet_Explorer_5_on_Windows_98…
In a 5 years.
Because of the intense competition between NetScape and Microsoft. Once Netscape was dead, Microsoft had no incentive to make IE any better and thus we had years of Stagnation.
Until people started moving to Pheonix/Firefox for a better browsing Experience.
FreeBSD doesn’t want to be a Desktop OS. It doesn’t even mention the word on it own home page.
The license allows Apple to do this is they want.
Also Apple has contributed to CUPS, LLVM/Clang and WebKit.
Edited 2012-06-03 06:38 UTC
That wasn’t really competition, IE was preinstalled…
It still is preinstalled – are you saying there’s no competition now?
(and, back then, IE was simply the better browser…)
The majority of webkit is LGPL licenced, what is your point?
It was a response to darknexus notion that it’s a good thing for a project when a company picks up their source and not contribute their changes back because in his words ‘competition drives innovation’.
I pointed out that even if FreeBSD would want to compete with OSX on the desktop they would have no chance since Apple can take all of FreeBSD’s improvements while keeping their own to themselves.
Unless you’re forced to use GPL without having incorporated GPL into your code, it is NOT forced communism. You are FREE to not choose GPL.
You’re free not to use GPL libraries. This criticism I see all the time is just people complaining that they’re not allowed to use other people’s code without conditions.
Well guess what, normally, you’d have to BUY such code or write your own.
The FSF never claims to be about “total freedom”. The FSF, the GPL especially, is about SUSTAINED freedom and a pragmatic freedom. Read up on philosophy some time and learn about the actual effects of total freedom.
Companies like Oracle would then be free to continue leeching off the efforts of the original developers without giving anything back.
Please explain the lack of drivers for the BSDs compared to Linux.
Linux isn’t in competition with anything. Competition does not drive innovation in open source because money is not what is needed to survive.
This just begs the following video to be posted http://www.youtube.com/watch?v=7QDv4sYwjO0
Also:
http://mises.org/daily/2066
http://reason.com/blog/2006/12/27/the-anarchy-advantage-in-somal
kwan_e,
I agree with almost the entirety of your post. But this below didn’t compute for me:
“Linux isn’t in competition with anything. Competition does not drive innovation in open source because money is not what is needed to survive.”
Funnily enough, I think I disagree with almost every single point in there, haha.
Linux (both the platform and the kernel) is in competition with everything else on the market.
Some open source devs are lucky to be paid, but many just donate their own time, which drives alot of developers away from open sourcing their own work. I myself have a very difficult time finding a balance.
Even if money is not an issue, competition always has a presence, and that’s a good thing for driving innovation.
Edit: with regards to the point you were initially responding to, I believe linux’s good driver support has a lot to do with it’s relative popularity. BSD’s lack of support makes sense given that very few manufacturers are paying attention to it(them) and having fewer resources. I would not spontaneously attribute BSD’s technical design to be the cause of it’s lack of drivers (if I understood correctly, seems to be what was implied).
Edited 2012-06-03 04:13 UTC
I don’t actually think it’s true. Linux, like almost all open source software, is about “scratching your own itch”. People contribute to Linux and open source, not because they want to compete and be the best and win awards. They’re just contributing because they want something done and there’s nothing around that fits their criteria, or that can be modified to fit their criteria.
If anything, what Linux competes for is attention, and it succeeds heavily in that area.
Which exists whether or not GPL is used.
The person I replied to claimed that if Linux were public domain, or at least not GPL’d, that it would magically have better driver support due to the kernel’s internal interfaces being stabilized in face of competition.
Well, BSD is right there for such a speculation, and its lack of GPL has not borne the results claimed.
kwan_e,
“Linux, like almost all open source software, is about ‘scratching your own itch’. People contribute to Linux and open source, not because they want to compete and be the best and win awards….”
Oh I don’t disagree with that, but I think these people would still exist without linux and I suspect if linux had failed to capture their interest then there would have been no shortage of competing “free” platforms to take it’s place (for better or worse).
“[Competition] exists whether or not GPL is used.”
Sure.
“The person I replied to claimed that if Linux were public domain, or at least not GPL’d, that it would magically have better driver support due to the kernel’s internal interfaces being stabilized in face of competition.”
“Well, BSD is right there for such a speculation, and its lack of GPL has not borne the results claimed.”
I personally dislike the lack of an ABI (at least within major versions) for lots of reasons, but overlooking this opinion, I don’t think we have strong evidence either way about whether the GPL license itself has impeded manufacturers from writing more drivers. Success is more often due to network effects that technical superiority. Most people agree that postgres is far superior to mysql, yet look who’s on top?
At the end of the day developers should release code in whatever terms they feel comfortable with. I honestly don’t know if the GPL’s relative success is due to a network effect or because it strikes a coord with more developers. But I don’t really care too much because I’m still free to use whatever license I want when I write software that I own the copyrights to.
I’m sure that we can both agree that when it comes to other people’s copyrighted code, it’s pretty arrogant to complain about their choice in free licenses.
Edited 2012-06-03 15:55 UTC
Sure. I was going to add to my point that, if anything, Linux is really in competition with itself. We get so used to the idea of one product competing with another product. The brilliant thing about Linux and open source is that it’s not dependent on external competition to be innovative, and so having competition doesn’t magically improve things. What open source allows us to do is to have the current version compete with the “ideal” version. Maybe we could even go so far as to say the “Platonic” version.
Well, I’m a software engineer, but I think I like biology more these days, and it definitely shows in almost all my comments. And in biology, being the “best” is never a guarantee for dominance, let alone survival. Very often, and especially in the case of humans, it is the blind patchwork that often saves the day.
You mention MySQL. One thing MySQL has in common with Linux is that they’re both oriented more towards the “near enough is good enough” mentality. Could that be why people, deep down, feel more comfortable with supporting them? I think network effects work better when there is no guiding direction.
It certainly would make sense of why the internal Linux ABI is not stable and yet it survives in that form. I can see why Linus accepts that instability. Drivers and hardware changes all the time, and while it would be nice to fit it all together, it’s not pragmatic to do so given the variety of interface designs. As long as the external ABI is relatively stable, which is what Linux aims for, then it’s not as big of a problem.
Yes, that’s always been my beef with anti-GPL sentiments. I like both GPL and BSD like licenses for their various reasons and they both have their own niche. They are not in competition with each other either and there’s no sense in inventing a conflict. I find it ridiculous that people can even take sides against each other.
Has Linux, GCC, Git, Samba, Qemu, VirtualBox, FFmpeg, x264, Blender, MPlayer, Gimp, Inkscape etc etc suffered from being GPL licenced?
The existance of GPL has no bearing on it’s how widespread it has become, it’s people choosing to use it which made it the most used open source licence in the world.
Now if GPL hadn’t existed then another similar licence would have since it obviously fills a need. If no one wanted to use GPL then it would have faded into obscurity like so many other licences which practically no one wants.
And before you attribute this to the ‘viral’ nature of the licence, realise that in order for this to have any impact there needs to be alot of code out there written and licenced as GPL which alot of people want to use and also that they find the licence terms acceptable.
At the end of the day it’s a choice the coder makes for his/her code: ‘under which terms will I allow someone else to use my code?’
Some people say to use my code you give me cold hard cash, some say to use my code you must publish any code you distribute it with, some say to use my code you must publish any changes made to my code you distribute, some say to use my code you must keep the copyright attribution intact.
There is no right or wrong, there are simply preferences which in turn can (and I’d wager often does) change depending on the type of code in question.
Personally I think BSD/MIT style licences are the best choice for component/framework style code while GPL is best suited for large collaborative projects.
But that is just my preference. It’s up to the creator/owner of the code to set the terms and I can choose to accept them or simply walk away.
Obviously alot of people find the GPL licence acceptable as it is so widespread in use and also alot of people sees it as their licence of choice as so many licence their non-derivative code as GPL.
You can learn from GPL/BSD/MIT source code aswell, just not copy it verbatim.
Given the horrific wave of police state measures and the soon to be blight of UAV drones in the sky’s of all Western Cities including across North America, Linux and Linux Distro programmers should add to the end user agreement ‘Banned From Military Use’ making it abundantly clear the OS and applications are not intended to sow chaos, death and destruction.
Ban the Military from Using Linux and Linux apps to conduct it’s endless campaign of global deaths and domination of other weaker nations.
They ship devices. Customers ask for sources. They give a canned response about industry standard being 90-120 days. Unacceptable. They know that going to court would take longer than that. They do it for each device. They need to go to court and stop this practice on current and future devices. I have a HTC One S. The stock firmware sucks and 3rd party firmware is buggy because the sources aren’t released.