Does GPL v3 Alienate Developers? 430
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.'"
Re:Impression (Score:5, Informative)
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 (Score:5, Informative)
Most hobbyist do choose the GPL from what I've seen, but I doubt they make up the bulk of GPL developers.
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.
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. (Score:3, Informative)
That's why I like the LGPL (Score:5, Informative)
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:Impression (Score:5, Informative)
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:That's why I like the LGPL (Score:3, Informative)
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:yes, GPL is a commercial licence (Score:2, Informative)
http://www.germany.fsfeurope.org/projects/gplv3/b
MOD PARENT UP (Score:3, Informative)
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.
Bad call on BSD (Score:1, Informative)
Bummer, you've already broken copyright and you haven't even moved a line of code!
Re:Developers are affected by licenses too... (Score:3, Informative)
The GPL says no. Here's a quote from the LGPL (which does allow dynamic linking):
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 (Score:5, Informative)
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 (Score:3, Informative)
There have been retroactive extensions at least for the USA.
57% of projects on Freshmeat use GPL (Score:5, Informative)
Reports of the GPL's demise therefore seem exaggerated.
Re:Impression (Score:2, Informative)
Re:Impression (Score:3, Informative)
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 (Score:3, Informative)
Re:Yes. (Score:4, Informative)
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. [fsfeurope.org]
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 [fsfeurope.org]
It's more than that (Score:3, Informative)
http://www.linuxdevices.com/articles/AT7188273245
http://fsfe.org/en/fellows/ciaran/ciaran_s_free_s
http://gplv3.fsf.org/additional-terms-dd2.html [fsf.org]
And the LGPL v3 is actually written in terms of the GPLv3:
http://fsfeurope.org/projects/gplv3/barcelona-rms
http://gplv3.fsf.org/lgpl-draft-2006-07-27.html [fsf.org]
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.
Google's blatant GPL violation (Score:1, Informative)
So, inbetween the lines of Greg Stein's ramblings is the subtext that Google doesn't want to be bothered with honoring the GPLv2 and therefore doesn't want any additional provisions added either. It's not a surprise that a company that violates the GPLv2 won't like the GPLv3 either. But for such a violator to claim their prospective should be considered in the process is just a clear conflict of interest.
So what is the problem? His claim is that additional terms make the license less permissive. In fact, the GPLv3 is written with additional terms to KEEP the work *MORE* permissive that the GPLv2 does. The GPLv2 allows for "tivoization" by the redistributor which makes the redistributed work less permissive. Tivo does this by signing the binaries so the customer is no longer on equal footing to install modified works on the Tivo since the customer doesn't having the keys needed to re-sign the binaries. Google's GSA also does a form of tivoization with a grill across the front of the unit, security screws and a password protected BIOS to all make it hard for the customer to ever open the GSA or boot from any alternative source other than what is supplied by Google. Like with TiVo, the concept used by GSA is to deny equal footing by the customer to install modified works. Yet a company that uses these non-permissive methods has an employee try to claim they are attempting to ensure the GPL remains more permissive?!
Google and Google's Greg Stein doesn't have FOSS developer's interest in mind any more than the other GPL violators of the world does. Only when they decide to actually honor both the legal letter and the spirit of the GPLv2 should they even be considered to be worth listening too. Until such time, his words could be considered in the same light as the actions of the company he works for--just plan full of crap.
Re:Something's Wrong with Slashdot (Score:2, Informative)