Does GPL v3 Alienate Developers?

An anonymous reader writes "Via Wired, a blog post in which BMC Software's Whurley and Google's Greg Stein agree that the GPL v3 is currently on a path that will alienate developers. Stein has an interesting theory called 'license pressure' which is similar to 'pricing pressure'. 'Due to pressure from developers, all software is moving towards permissive licensing" translation, the GPL and developers are moving in opposite directions ... Developers care about the licenses on the software they use and incorporate into their projects, they like permissive licenses, and they will increasingly demand permissive licenses.'"

Comments

Re:Impression

[jimstapleton]

Not quite - it's designed so that any contributions to it, if the result is distributed, are given back to the community.

I think this also includes contributions that would allow non-GPLed software to access it.

Selling the non-GPLed + GPLed = make money off of other peoples work.

Though, to my knowledge, there isn't an OSS license out that prevents making money off of other peoples work.

Re:Impression

[99BottlesOfBeerInMyF]

I was under the impression that the GPL license is mostly meant for "hobby" developers that want to make sure no one abuses their code to earn money on time they donate for the good of mankind.

Most hobbyist do choose the GPL from what I've seen, but I doubt they make up the bulk of GPL developers.

Not industry developers that want to earn money from their code. I might just have gotten it all wrong though.

Well, there are 50-100 developers in my office today and most of them work on GPL code at some point, paid by the company. I don't think we're unusual in that regard. When commercial developers release code as open source they do so with a motive of making money. You're not going to make money directly from OSS. You make money using OSS and getting free improvements from others and interoperability with other tools is the main benefit. The GPL insures you get those improvements and the competition does not grab all your code, and start a closed fork of it. The only time we use BSD licenses is when it is a vital infrastructure component we're trying to get widely adopted as a standard. In those instances, getting people to use and integrate it into closed software is more important than getting the improvements back.

Any developers willing to comment on what they want out of a license?

I think I just did. This is the situation as I see it and I think it has been stable for quite a while. I see more OSS development happening lately, but if anything it is code that old school people would release as BSD, now being released as GPL (and often failing to be adopted widely as a result). I guess I have to disagree with the article on that point (sort of). I see more code being released as GPL (both code that would otherwise have been less permissive like a closed license and code that would have been more permissive, like a BSD license). I see more LGPL code, which is a bit more permissive, I suppose, but I see that more as increased granularity rather than a move towards more permissive licensing in general.

Yes.

[LWATCDR]

It does me. I am actually considering taking over a FOSS project that lost it's maintainer. I was thinking about moving it to GPL but now it will stay BSD.

Re:Yes.

[fsmunoz]

You see that is just it I did like GPLv2 enough that I did contribute code to a GNU project.

That's great. You could even do it without really liking it, if it crosses the "good enough" threshould. Many people that live by the GPL contribute to BSD licenced programs, it's not ideal but good enough.

What bugs me is how this is targeted towards a company that did follow the letter of the GPLv2, is struggling, and has been in my opinion good for Linux. To me this says hey you can follow the rules to the letter but if RMS decides that he doesn't like you he will target you. I find this super counter productive.

Well, then you have a good reason to dislike the GPLv3, as I said above. I disagree, but it's a good reason nonetheless. I personally feel that the spirit of the licence is what's important and that it must be updated to reflect changes in society, and in that perspective tivoization and and patent shielding are subterfuges. Since they are able to exist by following the letter of the licence then the licence has to be updated.

Hating NVidia for supporting Linux but not the way you want them to. I feel it is more productive to state that you will buy ATI cards if they open source their drivers. Which frankly I will do. I would buy Intel but I have a nice AMD motherboard in my current system and Intel doesn't make stand alone cards. Frankly I am happy that NVidia does provide driver for Linux at all. I would rather have FOSS drivers but half a loaf is better than none. Why can't people be more positive and yes even grateful.

Indeed... note that I'm not doubting your motivations or even your entushiasm and contributions, we're just debating here. The closed drivers thing is a though problem with many possible approaches, all of which have pros and cons.

Now fighting software patents that is what I would really like to see the FSF doing and not targeting Tivo. BTW I don't work for Tivo or own one. If nothing else this is my shout that the FSF isn't doing it's job and is being harmful to FOSS in general as well as the users of FOSS.

This is quite interesting... note that the FSF has done this extensively. Actually IIRC the previous revision of the GPL was made because of "patent shielding", which was another situation were following the letter of the licence could be done without following the spirit of the licence:

Around 1990, I found out about the danger of software patents. So in GPL version 2, we developed the section that we called "liberty or death for the program", although informally, because in GPL version 2 the sections don't have titles. This said that if you agree to any sort of patent licence that would limit the rights that your users would get, then you couldn't distribute the program at all.

Back then, when RMS talked about patents, many people dismissed it as "political posing", "fanaticism", "I like Linux but this is to much politics, RMS is an extremist", well, you get the idea. Some years latter and here we are, knee-deep in patent threats. To reinforce that the GPLv3 has extra provisions against this forms of sidestepping the licence:

A few years ago, I realised that there were other ways software patents might be used to make software non-free, so we're designing GPL version 3 to block them too. For instance, one issue is, what if the developer of the software has a patent on it, or rather, has a patent on some particular computational technique used in the program. (...) However, there's another way of using software patents to threaten the users which we have just seen an example of. That is, the Novell-Microsoft deal. What has happened is, Microsoft has not given Novell a patent licence, and thus, section 7 of GPL version 2 does not come into play. Instead, Microsoft offered a pate

That's why I like the LGPL

[Spy der Mann]

LGPL allows me to reuse the code that I've written as open source, in my boss' projects. I distribute it free because I feel it'll be useful to other developers out there.

I have the tendency to write software libraries, because they allow me to reuse my code in several different projects. The executable programs are just a wrapper. So, the LGPL suits me.
Good examples of LGPL projects are the FFMPEG library, which the LGPL ensures it can be used for both commercial and non-commercial projects.

And if that's not enough, there's the wxWindows (wxWidgets) license, which is GPL + exception.

Re:That's why I like the LGPL

[One Louder]

The LGPL basically says that you have to provide a "linkable" version of the code, so that someone can modify the LGPL components and recombine them with the non-free components - it doesn't mean that the non-free components must be modifiable.

You're correct that the combined work must be released with a license that allows this, and that many people don't know that they need to do it. It's one of the most common misconceptions about the LGPL - most people think they only have to publish the source to the free components.

Re:Impression

[epiphani]

This is partly why I've tried to convert my projects to BSD licenses. I have a substantial amount of code that I've written GPL, and after working with people on these projects for several years, its hard to remember who wrote what. As a result, I don't use that codebase in any work that I do for my company that may be distributed.

I'm proud of the code I write, and a lot of it is portable - I know it inside and out - but other people have fixed, added on, improved and optimized my code. As a result of that happening under the GPL, I can't use that for other closed-source projects I work on. It's frustrating, I don't feel comfortable using my own code because its GPL'd.

Anything I work on in the last few years goes out BSD licensed, and I'm trying to convert my existing projects to BSD licenses as well. GPL has its place in core utilities, but I won't be GPL'ing my own code again for some time. BSD licensing is the way to go, imo.

Re:yes, GPL is a commercial licence

[maxwell demon]

You link is broken. The following should work:
http://www.germany.fsfeurope.org/projects/gplv3/br ussels-rms-transcript.en.html

MOD PARENT UP

[TheRaven64]

It's amazing the number of people who post on Slashdot with blatant misunderstandings of the GPL (you can't sell it, you have to give changes back to the community, etc). I can't help wondering where they got their information from, since it wasn't the GPL itself.

My objection to the GPL is the Theo De Raadt test: If you need to be a lawyer to fully understand the license, it's not friendly to developers. That's why I release my code under the 3-clause BSDL; it's short, simple, and to the point. Any developer can understand it, and can comply with it, without needing a legal opinion.

Re:Developers are affected by licenses too...

[aardvarkjoe]

Thought linking to libraries or other components was fine as long as you where dynamically linking.

The GPL says no. Here's a quote from the LGPL (which does allow dynamic linking):

When a program is linked with a library, whether statically or using a shared library, the combination of the two is legally speaking a combined work, a derivative of the original library. The ordinary General Public License therefore permits such linking only if the entire combination fits its criteria of freedom. The Lesser General Public License permits more lax criteria for linking other code with the library.

The typical explanation is that by creating your program to be specifically dependent on the interface into the GPL'd library, you are creating a derivative work. This reasoning has never seemed to be very solid to me, as there is a vast continuum of ways that you could "link" to GPL'd code that range from being obviously derivative to obviously not, and I'm not aware of any real legal precedent that tries to define what the limits are. The FSF is pretty clear that they believe that you can't dynamically link to a GPL'd library from software under an incompatible license, though.

Re:Impression

[Grishnakh]

Now, they are using this to pressure others into going into GPLv3 also. And once they do it, others will be pressured too. It seems strange that we are talking about pressure and manipulation in an area we normally save for closed source companies. I guess when in Rome, do as the Romans do. But I think this strikes at the heart of the article. When FOSS starts doing the things we resent proprietary companies for doing, the love affair is over.

And why would anyone, except for proprietary companies, resent the FSF for this? As an individual developer, and also someone that works on Linux device drivers full-time, I have no interest in patenting software, and I want to distribute my source code to those I distribute binary code to (and back upstream when appropriate). I have no interest in keeping customers from seeing my source code if they're using my code on their devices. So, I have no problems with either the GPLv2 or GPLv3. Why on earth would I resent the FSF for this?

Companies like TiVo might resent them for the GPLv3, but I really don't care about them. I have no desire to buy a Linux-based device that I can't modify.

Re:Impression

[TheLink]

Uh, copyright laws have made public domain stuff go back into private domain.

There have been retroactive extensions at least for the USA.

57% of projects on Freshmeat use GPL

[cabalamat3]

I've just checked and 56.82% of projects in Freshmeat use the GPL (24436 out of 43001). Another 3449 projects (8%) use the LGPL.

Reports of the GPL's demise therefore seem exaggerated.

Re:Impression

[Entrope]

Civil trials in the United States -- at least any that people bother about -- must give the option for a trial by jury, per the Seventh Amendment. Either side may demand that a jury decide the facts in question.

Re:Impression

[WGR]

You are missing the essential strength of the GPL.

When I, as a commercial developer, license software under the GPL, I am assured that my competitors will not take it and develop improvements that I cannot use. The GPL is much more corporate friendly than BSD exactly because it guarantees that the code remain open over any improvements. It means that the developments of any contributor remain available for all so that the software "evolves" and does not become proprietary and unavailable for the original contributors.

The problem of who contributes what is less of a problem, since one agrees to the GPL when you start using software with that license. Unless you re-create all the GPL'es code, your code will also be under GPL.

BSD type licenses allow a company (Apple, for instance) to start with code developed by a community of developers and then privatize it with no compensation to the originators, not even giving them the improved code.

Re:Impression

[PitaBred]

Hell, there aren't even any usage restrictions with GPLv3. It's all in redistribution and modification. So use it to your heart's content... but if you change it or adapt it to your needs, you get to compensate the community by opening up any "IP" you may have included.

It's more than that

[g2devi]

Actually, it is mean to stop license proliferation of the 3rd type:
http://www.linuxdevices.com/articles/AT7188273245. html
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_so ftware_notes/why_gplv3_says_additional_permissions _are_removable
http://gplv3.fsf.org/additional-terms-dd2.html

And the LGPL v3 is actually written in terms of the GPLv3:
http://fsfeurope.org/projects/gplv3/barcelona-rms- transcript.en.html#lgpl
http://gplv3.fsf.org/lgpl-draft-2006-07-27.html

So basically, the GPLv3 was designed to eliminated the need for any GPLv3-compatible license since any GPLv3 compatible license can be written as the GPLv3 license plus additional permissions. It might not be the most efficient way to specify your GPL-compatible license (e.g. the MIT license would be much longer if expressed this way), but it can be done. If the GPLv3 license existed, I doubt the GPL-like per file Mozilla license would have existed or the GPL-like for open source Qt license would have been created as independent licenses.

Re:Something's Wrong with Slashdot

[KingKaneOfNod]

The spirit of it is clear - no profit, no stealing, no typical corporate BS with the code

That's not true; the GPL isn't about any of these things, it's about ensuring the freedom of users to modify the code, which means when you distribute a GPL licensed program, you must also distribute the source code. This doesn't mean that you can't charge money for it, it doesn't mean "no stealing" (it's perfectly legal to sell someone else's GPL code - take the Linux kernel for example), it just means "no binaries without making the source available". That way when you don't like something about a program, you have the freedom to fix it yourself.