How Many Open Source Licenses Do You Need? 276
jammag writes "Bruce Perens, who wrote the original licensing rules for Open Source software in 1997, notes that there are a sprawling 73 open source licenses currently in existence. But he identifies an essential four — well, actually just two — that developers, companies, and individuals need. In essence, he cuts through the morass and shows developers, in particular, how to protect their work. (And yes, he favors GPL3 over GPL2.) For his own coding work, he's fond of the 'sharing with rules' license, which stays true to the Open Source ethos of shared code yet also enables him to get paid by companies who use it in their commercial products."
Re:Hi again (Score:5, Interesting)
Hi, Bruce,
While I agree in general that there are too many licenses, one of the problems I ran into (which you mention in passing) was that I'm not necessarily the one who gets to decide what license I'm using. When I talked with my organization's lawyers, they didn't care about license proliferation...they cared only about what they thought was important. So, we ended up with a modified BSD license: the standard 3-clause plus one more to address the lawyer's concerns...personally I think the fourth clause is redundant, but I'm not a lawyer, so they weren't listening to me.
In short, while I think it's good to get this group (ie, the coders) to start agreeing on licenses, the lawyers that we talk to need a bunch of education also. They all seem to want to customize licenses.
Has The GPL Ever Been Proven (Score:4, Interesting)
Has anybody ever proven this?
ie. has it ever been proven that attaching a 'must share' clause to a license (ie. GPL vs BSD) actually results in more people sharing code.
I am inclined to think and believe, based on experience, that it does not. Those who share are likely to share regardless of license, ditto to those who take your code and improve it with no intention of sharing.
Just how much does 'sharing' contribute to open source anyway, considering that all the top projects are tightly controlled by a small number of lead developers who hold the keys to commitments and in accepting patches. Code being shared will likely just go unnoticed anyway.
So, after 10 years, has anyone proven that the GPL works?
Re:GPL v3 vs Linus (Score:3, Interesting)
Linus is unfortunately one of the typically "can't we all just get along" geeks - he doesn't seem to care for the social good so much as being able to continue to work on his projects. Such people are certainly useful - "not seeing the big picture" isn't a barrier to being an effective technical leader (and by pretending such problems/disagreements don't exist or minimising them, they better enable people with substantive differences in the area to work together).
For people who do care about the public good, the best thing to do is to look for other people for inspiration on matters of licenses and large-scale strategy (like rms, BPerens, esr, theo, or one of several others, depending on one's particular inclinations). There's a lot of positions one might take on these matters, most of them better than playing ostrich..
Re:Hi again (Score:3, Interesting)
There is, is there not, a difference between 'need' and want. After all we probably only 'need' a small number of programming languages, but we have, and want, many.
It would, I beleive, be a mistake to cut down the number of licences. Evolutionary theory makes it quite clear that lack of diversity leads to a much higher liklihood of extinction in the event of a crisis, and Open Source is no exception.
Personally I'm very boring in my licensing. I use the GPL, or I just public domain my code without any if its trivial enough.
That said, I did do a lot of research when making my choice of licence, and while I wouldn't have minded it if the process of reviewing licences were simpler (like for instance a site where you describe your project and people suggest licence types), I didn't think there were 'too many'.
sounds kind of biased in the other direction (Score:3, Interesting)
I frequently hear them characterized that way, and I talk mostly to Windows-using academics who're discussing how to release their code, and could hardly care less about Linux or the Free Software Foundation.
The general viewpoint of options that get bandied about is something like:
1. "Research-only" or "non-commercial-use" license: minimal release that will ensure researchers who want it can get it, while retaining all commercial rights that default copyright gives.
2. Copyleft licensing: release that allows code to be used commercially or non-commercially, but only in free-software projects using the same copyleft license.
3. Permissive licensing: release that allows code to be used for any purpose, provided the copyright notice is maintained.
The first is is probably, unfortunately, the most common release mode for research code (though this is changing), since it feels like "giving away" the least as far as potential commercial exploitation goes. The second is usually an easier sell than the third, because it at least guarantees that Microsoft can't put your algorithm into the next version of Excel without paying you, which is the main thing people who get cold feet about releases are worried about (that they're giving away a potential source of livelihood for free).
GNU/Linux Distros (Score:3, Interesting)
I once too the time to put the core programs that make up a basic Linux distro into a spreadsheet, making notes on the programming language they used, file size and license. I used Linux From Scratch, so I could get an idea of a "core" working system, as opposed to thousands of packages. I think I narrowed it down to just over 60 packages to provide the basics. It taught me a couple of things.
1. Richard Stallman is right, the correct term is GNU/Linux. I was amazed at the percentage of packages in the core OS -- not applications -- that were from the GNU project. It was something like 75% or so.
2. C is by far an away the most dominant programming language. (Yeah, I know it should have been obvious. Duh!)
3. There are too many licenses. Off the top of my head I bumped into: GPL2, GPL3, BSD-2, BSD-3, MIT, Artistic, OpenSSL's "thou must advertise us" variation, Vi's Charityware, and at least one public domain.
Have you looked at it from this perspective? And would you consider approaching some of the existing projects with changing their license to one of your four?
He missed one: public domain (Score:3, Interesting)
Bruce missed the one option I think is the most important: public domain. It's not a license and so captures the "giftiness" of the gift licenses better than any of them.
Re:A question of values (Score:3, Interesting)
Mainly because we want to be protected from patent lawsuits. It's really painful to give your stuff away with no strings and get sued for your trouble. The Apache license tries to protect you from that.
Bruce
Just one more... (Score:3, Interesting)
q: How many guitars does a guitarist need?
a: Just one more.
There is always going to be someone that thinks they need something just a little bit different to suit their particular needs. In reality the number of open source licenses could be dramatically reduced, but the human condition makes us each think we are unique and have special needs and requirements for our unique project.
Are there too many open source licenses? Yeah. It's way to complicated for the end user. Will this ever change? Most likely not.
Re:1997 ? (Score:4, Interesting)
They're the rules for the Open Source Initiative and the Debian Project to approve licensing.
Richard wrote a statement of the Four Freedoms in an early edition of the GNUs Bulletin, which was mostly distributed in paper form on the MIT campus and environs. He did not further promote them until a long time later. So, when I had to write license guidelines for Debian, the Four Freedoms document was unknown. I sent my document to Richard, and he wrote back that he felt it was a good definition of Free Software. Surprisingly, he did not mention his Four Freedoms document in that correspondence.
Much later, FSF published its statement of the Four Freedoms on its web site as an alternative to the Open Source Definition.
Bruce
Businesses and Lawyers (Score:2, Interesting)
Businesses and lawyers are concerned about CYA maneuvers. You are concerned about efficiency.
No one got fired because there were too many licenses.
Re:Hi again (Score:2, Interesting)
You spent years pissing in their pool and insinuating that they were childish.
Is it really so surprising that they fail to give you special consideration?
Re:He missed one: public domain (Score:3, Interesting)
Re:He missed one: public domain (Score:3, Interesting)
There seems to be a persistent source of BSD licensing propaganda that says these things. IMO he or she is attempting to mislead you, or is self-decieved.
It is entirely legal and ethical to combine work under the BSD and Apache 2.0 licenses. When you do so, part of the work is under the BSD license, and part is under the Apache license, and you have to follow the rules of both.
What whoever-it-is is complaining about is that when you combine the two works, the part that is under the Apache license doesn't automatically become a work under the BSD license instead of the Apache license.
This is also true for all other licenses. Unless there is a license somewhere that says "if you combine me with a BSD work, I automatically become the BSD license".
I can't imagine what whoever-it-is expected would happen. I would rather have the Apache license anyway, because of the patent terms. Who wants to lose them?
Bruce