Submitted by Eric S. Raymond is one of the three big figures in open source, together with Linus Torvalds and Richard Stallman. During a talk for the Long Island Linux User Group, he made some interesting statements about the GPL, namely that the GPL is no longer needed due to the way the open source movement works.
Eric S. Raymond questioned whether or not we still need reciprocal licenses like the GPL in a world where abusing open source code by making it closed source is pretty much a death sentence for your company. “One of my heretical opinions is that we worry way too much about licensing. And in particular; I don’t think we really need reciprocal licensing. I don’t think we need licenses like the GPL, that punish people for taking code closed-source,” he said. He goes on to explain why.
According to ESR, open source development is the superior way of developing software because you get more out of your investments with open source development than with closed source development. Because of this, ESR argues, it is generally a bad idea to take open source code, make it closed source, and develop a business around that, because the original open source product will still be developed, but at greater speeds and efficiency than can be maintained by the small, closed-source team.
Consequently, ESR states, the market will punish you for taking open source code and close it up. It’s simply never a successful strategy, he argues, and as such, how much sense does it make to have reciprocal licenses like the GPL? “That is why I don’t think you really need GPL or a reciprocal licenses anymore,” he said, “It is attempting to prevent the behavior that the market punishes anyway.” He added that the BSD license is a good alternative to the GPL.
He takes it one step further by stating:
That attempt has a downside, the downside is that people, especially lawyers, especially corporate bosses look at the GPL and experience fear. Fear that all of their corporate secrets, business knowledge, and special sauce will suddenly be everted to the outside world by some inadvertent slip by some internal code. I think that fear is now costing us more than the threat of [inaudible]. And that is why I don’t we need the GPL anymore.
That’s a pretty heavy thing to say, especially for someone so synonymous with the open source movement. The article linked to above gives the example of Mac OS X as a successful attempt at taking something open and making it closed, but this is a faulty assumption. Apple has not closed off any code; everything open they ever used is freely downloadable, as well as the changes Apple has made. It’s just that in the case of Darwin, no one really cares.
It’s a controversial opinion by ESR, but he does make a very compelling argument. The GPL won’t go anywhere any time soon of course, but that doesn’t mean we shouldn’t question its usefulness.
What, wait, did he just say ‘leave it to the market’?
Has he not seen the recent economic market take a rather large nose dive because there was zero control to stop companies being moronic?
It might be tedious but it keeps us all honest.
I can’t really speak for every sector of the industry/every part of the world, but I was looking at job postings recently, and there is plenty of work in Toronto for .net developers right now, more then usual. If the recession is hitting the tech industry, I don’t feel it.
That has nothing to do with what I said.
Ok, to be more clear, it was the banks being moronic that caused the recession. You were painting everyone with that brush, especially the tech industry which has barely been touched by it.
I don’t really disagree with your point in a general way, but I don’t think you can use the recession to prove it.
It doesn’t prove that the tech industry is exactly like the banking industry. However his point was that you can’t trust industry in general to self regulate and that “enlightened self interest” is not enough. It does absolutely support *that* point.
Then they would better understand what any free market is capable of, what it cannot do, but what it does very well.
A regulator has a measuring device, and an actuator which is used to regulate whatever is measured. In between there is some regulation function which calculates which movements of the actuator are necessary to keep the measured value as close to the desired value as possible.
In a market, the thing to be regulated is the price. So people measure the value of something, apply a logic to it (compare to the price at which it is offered), and execute the result by either buying or not buying or haggling for a lower price.
That’s it! A market cannot do more.
And if the measuring device is not enabled to correctly measure the value, for example by obfuscating the value by multiple repackaging of something, or by creating the n’th derivative of it, then the consequence is, that the actuator will do something completely stupid and the price will go haywire.
And NOT letting companies die which by all means should have died is Darwin’s Nightmare all over again. Letting business units become “too big to fail” is one mistake the market cannot regulate. It needs regulation from outwards to do that. As the market does not measure company size, it can not regulate it.
Yeah, it’s obvious that individuals are stupid. So we need to leave everything in hands of something that knows better what freedom is. Like the FSF. Thanks to the “or any other later version” of the GPL, programmers don’t need to think about freedom, the FSF decides what is freedom for them. It was really nice to see the FSF changing with the GPLv3 the licensing terms of millions of lines of software, to fit not to the desires of the people who wrote those code, but the ones of the FSF staff. Because they know better than me.
(do i need to add the ironic tag?)
Dude, you need to educate yourself on how the GPL works before you open your mouth again. You just made a complete fool of yourself.
By releasing the GPLv3 *zero* people who have licensed their code under GPLv2 were affected unless they chose to upgrade their licensing.
Really what we are seeing with people like you is a *religious* or *political* distaste for the GPL and not a logical objection. So I guess that is sort of the definition of “fool”.
Did you actually read his comment?
The FSF advises to use the following wording:
“This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.”
If you use it as-is, and a new version of the GPL is written, users have the option to use the newer GPL. Suppose if the FSF was hypothetically taken over by Steve Ballmer’s evil twin, and rewrites the GPLv4 to give a blanket license to the FSF there is nothing to stop them. Of course, such a scenario is not realistic in the short term. I think the anti-patent licensing terms can be more of a problem in the short term. What happens if a really FLOSS-friendly company (say Red Hat) is sued over a patent, and they can not succesfully defend themselves. Then they would need some patent licensing deal, effectively terminating their right to distributing that particular software if it is licensed under the GPLv3.
I hate patents as much as the next guy, but it is not an unfeasable scenario.
Right. But presumably, you’re not just throwing licensing conditions on your work without reading them, right? So when you included a copyright notice at the top of your source files that says “or any later version” in the very first paragraph, you can hardly claim you didn’t mean it, right?
Yeah, and if Vatican was hypothetically taken over by Satanists who rewrite the Bible to make the Devil appear as the good guy, there is nothing to stop them either.
Well, it wouldn’t be the end of the world for them. They could easily re-license any GPLv3-licensed software they’ve written themselves, and they could continue distributing software that ships under any other free software license, including GPLv2. However, not signing patent deals gives them the advantage that they can also distribute GPLv3-licensed software, while their competitors who *have* signed patent licensing deals don’t have this advantage.
Actually, the OP isn’t entirely wrong.
All versions of the GPL contain a paragraph which specifically states that, if no explicit statement is made as to which version of the GPL applies to the code, *any* version can be taken to apply – including future versions which haven’t been written yet.
If you just write a piece of software and stick a copy of the GPL in the archive, no matter what particular version of the GPL that file you put in the archive was, anyone can pick any version of the GPL to apply to your code. This isn’t exactly what the OP said – later versions don’t supersede earlier ones in this case, so you’re still free to apply GPLv2 to any code without a specific GPLv3 statement – but it’s not what you said, either.
If you want a particular version or combination of versions of the GPL or LGPL, and *only* those versions, to apply to your code, this has to be explicitly stated.
I think that’s any *later* version. The FSF’s GPL is like the back-swept teeth in the maw of a shark. They want the default movement to be in one direction, with no option to go the other way.
Edited 2009-03-25 16:39 UTC
No. You’re wrong.
This would happen only when you would just put the small note in the code “this program is under GPL” without specifying the license text.
If you attach a text of a specific version of GPL then noone can take your code and release it under an earlier version of the GPL.
Read the GPL FAQ.
http://www.gnu.org/licenses/gpl-faq.html
Each version is given a distinguishing version number. If the Program specifies that a certain numbered version of the GNU General Public License “or any later version†applies to it, you have the option of following the terms and conditions either of that numbered version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of the GNU General Public License, you may choose any version ever published by the Free Software Foundation.
GPL’s “v2 or later” is optional, you are making it sound as if poor programmers were being imposed the license updates by the evil guys from the FSF.
Well, they are and they aren’t. It’s totally up to the programmer which version of the GPL they license it under – including whether it says “or later.” The problem is that you don’t know what later licenses will entail. A programmer may love v2 and have licensed their code under “v2 or later” only to hate v3 when it came out and regretted that they left the “or later” clause allowing their code to be put under v3.
Technically, v4 of the GPL (should they ever choose to make one) could be identical to the BSD license, or the Apache license, or whatever the FSF feels like doing. By leaving “or later” in the license, you’re allowing the FSF to decide what your license will say whenever they put out a new license. There are no guarantees that a new version of the GPL will be something that you agree with.
Naturally, the way to avoid the problem is to license your code under a specific version of the GPL and not include the “or later” clause, but anyone who leaves the “or later” clause in their license is letting the FSF do whatever they want to their code license-wise.
I know this is not an economics blog, but just for the record, not everyone agrees with that explanation. For instance, see:
http://mises.org/story/2895
I’m not sure if I’m quite willing to join ESR in abandoning the GPL. I think it’s done a lot of good.
Still, I respect ESR. He’s a thinker. A visionary. He deserves to have his words considered.
You’ve got to be kidding! We’ve enjoyed 200+ years of free-market excellence. Recessions come and go. Morons in private companies can only do so much damage. But morons in Washington may yet destroy the fabric of the nation.
In that case Eric: [Citation Needed]
So I wonder why we aren’t all currently using *BSD if the GPL is such a hinderance ?
I don’t understand ESR to be fair. Basically, he says that we should use the BSD because no company wants to close their source code. So, if nobody wants to close the source code…why would people move away from the GPL, which keeps it open anyway?
And, no one forced anyone to pick the GPL in the first place. You are right that the *beauty* of the GPL is that it makes it easy for developers to not think to hard about the details but that isn’t because the FSF is forcing anything on anyone (this is the part where you realize you were really talking out your butt).
Plus, anyone can change their licensing any time they want for all future development.
The FSF and the GPL protect us and we ask them voluntarily to do so. Your strange right wing distorted view on how things work is really f-ed up my friend (sarcastic on the “friend” part in case you missed that. I actually can’t stand fiscal right wingers)
fiscal right wingers?
How much debt are you currently in?
How much of that is not a mortgage on a home — meaning car, credit cards, student loans, etc?
Pooo! Right wing? What does that mean to you?
I think diegocg and you both support keeping the GPL. So why the not-so-kind response?
Because the BSD license is more flexible, and you can use BSD licensed code within your already-proprietary products without any issues, but you can’t do the same for GPL code (or, at least that’s how it was explained to me, a lay person).
Edited to add quote.
Edited 2009-03-25 20:56 UTC
GPL doesn’t care about your proprietary programs. It doesn’t need them.
What I said was you cannot use source code under the GPL with proprietary-licensed or closed-source code, but you can do that with BSD licensed code. The latter makes open source code more appealing for integrating with already-proprietary systems because they will not have to re-license their system and/or open the source up for the world to view their trade secrets [or whatever].
Indeed, you are right. That is, because the FSM sees the proprietary computer software as a social problem. Originally, the OSS software has been a marketing campaign for free computer software, but they disagree in the fundamental things.
There are lots of projects that use BSD/MIT licenses, and are still active. Some names come to mind: Postgres, xorg, FreeBSD, OpenBSD, Firefox, WebKit, GHC.
There’s no reason for GPL, nothing to be gained from it.
You are mistaken about Firefox as it is tri-licensed under GPL, LGPL, and MPL.
I’m not, the MPL still applies. Also, I wonder why I’m being modded down so fast. Does telling the truth shake some neurons?
Why don’t you read the MPL in the first place? MPL is not BSD.
MPL is a weak copyleft license. Source code copied or changed under the MPL must stay under the MPL.
Because people don’t know a good thing when they see it.
ESR made a similar comment in 2005 :
http://www.onlamp.com/pub/a/onlamp/2005/06/30/esr_interview.html
http://linux.slashdot.org/article.pl?sid=05/07/01/1210258
—–
I also need to point out that ESR is the guy who ran VA Linux hardware division into the ground , that he was on the board of the defunct “Linspire/lindows”. So he tested is theory in practice and it failed.
He knows first hands that he is wrong … Yet keep pushing the idea that “this time” , he might get a different outcome.
ESR did not run the VA Linux hardware division to the ground, because he never had any management responsibility at VA. He was an outside member of the board of directors, brought on for the publicity value.
http://www.linuxtoday.com/news_story.php3?ltsn=1999-12-10-001-05-NW…
http://www.linuxtoday.com/news_story.php3?ltsn=2001-06-28-008-20-PS
ESR is:
a)naive
b)wrong
I read a bunch of his essays on the Iraq war and Bush’s policies and this guy fits very neatly into a category I call “f-ing idiots” (except I spell out f-ing).
Cathedral and the Bazaar was pretty good but this guy has been riding that fame for way too long.
GPL violations still happen *all the time*. So I have no idea what the hell he is talking about. Without the GPL people can build upon existing code in new and different ways that the software is not evolving in now and then give nothing back. This *will* happen. I *does* happen to BSD licensed code all the time.
If I was a capitalist jerk off and I spend a bunch of money developing a new twist on a mountain of existing code that allowed me to kill my competition I would not give it back unless I was forced to. With the GPL they have to weigh the benefit and they often do give back spectacularly. (It is called feedback with gain and it is one of the major reasons *why* linux is so successful)
Anyway, in summary, ESR is and always has been an idiot.
For me it seems to be just a selection of some facts connected with each other into logically looking stream, but… This is a part of a game with the audience. After that you have an explanation based on these prepared stream of facts and… the only logical answer is simple: “yes, we don’t need GPL”.
Of course I cannot agree. It’s just a kind of demagogy presented by very intelligent person to the audience. Nothing more. We cannot disagree, because presented facts, the way of thinking based on these facts and final conclusion are just logical. We don’t want to be considered illogical so we’ll probably agree. But we shouldn’t.
I appreciate ESR very much, but… This theory is to simple and we can make a lot with logic, as well as with statistics, to prove our point of view in good looking style. Especially if we are good speakers. But still it’s demagogy – an art of proving any arguments, but not a way of presenting the truth.
As a developer and open source user I need GPL to protect my rights.
I don’t get the idea behind ESR’s way of thinking!
And see something else – eliminating GPL will eliminate… Yes, answer yourself – who will be eliminated from open source community? (if not totally then at least his role will be minimal without GPL). And who is the biggest evangelist of GPL?
That’s why I don’t take ESR words seriously. It’s just a play for me. If he is serious I think he is thinking something much more than just leaving the GPL. He maybe tries to tell us, that we should definitely change our way of understanding the open source.
Once again – it’s good to listen to such persons like ESR very carefully. Sometimes the real meaning is hidden much deeper in the words.
Maybe we just don’t need any licenses. Well… Personally I don’t think so.
Which I define as “a librtarian *and* a moron”
So his comments should surprise noone.
…and a racist… and a witch… and a gun nut… the list is actually pretty extensive
I know ESR is a gun nut, and a hardcore libertarian, but a racist and a witch?
Any evidence of that?
Anyway, ESR always takes a hardcore libertarian stance. I’ve seen his blog at his personal website, and he always espouses markets as naturally self-regulating. He even goes so far as saying the current credit/banking crisis is only a natural, temprary blip, and that overall less regulation is better (I disagree, but do think that markets do have some self-regulating qualities, but they require oversight).
So take his blogs/writings/speeches, etc, with a grain of salt.
ESR may not think the GPL is necessary, but for a lot of people (and companies), it is quite useful, if not completely necessary.
on being a witch: http://www.catb.org/~esr/writings/dancing.html
as for the racism bit, he had a blog post on IQ quoting studies that nobody really takes seriously unless they are a racist
That page quotes the relevent bits, and links to the month it was posted, couldn’t be bothered to track down the actual link
http://www.gnxp.com/MT2/archives/001393.html
Good links that prove your points (more or less).
Especially the IQ thing. Only a racist would jump on IQ tests as proof of inferior intelligence of a discriminated minority, conveniently dismissing the influence economic and social conditions can have on said IQ scores.
Thus, ESR, while certainly not the blatent KKK type racist (by any means), appears to be more of a pseudo-intellectual, self describing non-racist, but really a
subtle racist.
The link on the Wiccan thing was interesting. I actually share some of ESR’s experience in terms of scepticism for religious dogma, or organized religion in general, and finding some Eastern mysticism somewhat sensibile. But I certainly don’t share this wiccan thing.
So, okay, ESR is a Witch (self described, card carrying), and a (subtle, but actual) racist (that he won’t admit to others or himself). Add to that gun nut, and hardcore libertarian, and open source advocate.
But one things for sure – he is interesting.
No, not really. His views are not the first I want to read on any subject. Nor are the collectively that unique or thought provoking.
Yeah, he’s also a fantastic writer, which is why I’ve read most of what he’s written. I used to regularly read his blog until that post actually, which is when i jumped off board. I don’t even know if he even is a racist, but he is ready to jump on any half brained bandwagon and ride the hell out of it without doing much questioning. I think that is what this whole thing is about, he’s just realized the GPL is in direct opposition with his free-market libertarianism, so now he’s gotta come up with a way to make it all fit. A few years ago he used to say that the GPL makes sense everywhere except for when you can make a strong business case against it, and even then it makes sense to gpl some bits, and keep the rest proprietary.
Unfortunately for him, the reactionary parts of the the linux world will probably totally despise everything he says for now on, and they are the ones that tend to form user groups that pay for people like him to come and talk.
Sigh…Human biodiversity scientists claim that, after adjusting for economic and other disparities, There is still an IQ difference between particular ethnic groups that is too large to be statistical noise. There are still debates about the validity of General IQ as a measurement, but it’s largely political since it actually corresponds really well with success across any ethnic group. There are still debates about if the statistical difference is genetic (it may be an unaccounted-for demographic characteristic.) There are not debates that intelligence is largely heritable, it is. There is general agreement that “races” are a social construct, but that ethnic groups are genetically distinct is not controversial.
People smarter and more qualified than us to debate the topic have serious disagreements about the issue. But one thing that does suck is that studying the link between IQ and genetics is going to put you dangerously close to breaching a social taboo, perhaps the worst in Anglosphere culture, being accused of being a “secret racist.” Unfortunately even many scientists haven’t avoided the oversocialization that causes them to question even the motives of someone interested in these topics, which just means that anyone legitimately interested in the topic is not likely going to be willing to risk their reputation over it. “Free Inquiry” indeed.
What adjustments? And how were they applied?
Also, to make the theory, or data or stats, valid, it has to tested in different ways.
For instance, the IQ thing is saying African Americans score lower on IQ tests than say, Whites.
But does it take into account comparing IQ tests of Whites living in middle class or afluent neighborhoods as opposed to whites living in lower class areas?
It would be reasonable to suspect that the whites living in the lower class areas will score lower than the ones in the higheer class areas, due to different social, economic, and educational nurturing, conditions.
Same goes with African Americans. I would not be a bit surprised that blacks living in higher class areas score higher than blacks living in lower class areas.
There are many, many, other scenarios to take into account.
I also think that IQ tests have as much to do with how much an individual knows (accumulated knowledge), as with actual intelligence. Social context has huge influence as well.
Of course, that could all be BS, and perhaps these so called “scientists”, or whoever these people are that are coming up with these stats, have taken all of that into account.
Thus, I think it’s ridiculous to jump on these “studies” and start thinking that they show that certain ethnicities are inherently more or less intelligent than other ethnicities. I think doing so demonstrates a very subtle form of racism. I also think doing so is completely useless, because it can only lead to offending people or discriminating people. It can also lead to “tags” being put on people, and stunting their potential.
So you’re best off dropping it. So is ESR.
Sorry for the thread hijack. 😉
You’re more of an idiot if you really thinks IQ represents anything, or that intelligence is some sort of universal biological trait.
Most respectable sociologists and scientists wouldn’t even be able to explain to you what intelligence actually means on an trans-subjective level, so how the hell do you propose to measure it in the first place?
… and he went to sword camp.
I think it is bad to have so much license proliferation, and the number FLOSS licenses should be reduced. But the GPL is one of the most well-designed and tested FLOSS licenses, along with the Apache License version 2. I think in practice most FLOSS software can be covered under either:
– Apache License version 2 (as a modern replacement for the BSD license). For situations where the author does not mind incoorperation in proprietary software, or when a piece of software aims to be a standard implementation.
– LGPL: in cases where use with proprietary software is ok, but a library itself should stay FLOSS software.
– GPL: the defacto copyleft license.
– Affero GPL: for software where the author also wants to guarantee freeness when the software is used as a network service.
Personally, I prefer the Apache license version 2, it’s simple, it’s understandable, and I usually don’t mind how my code is used as long as people give proper credit.
Quotation: “Eric S. Raymond is one of the three big figures in open source, together with Linus Torvalds and Richard Stallman.”
Mr. Stallman has absolutely nothing to do with the open source initiative. He’s an important figure in our free software movement.
Quotation: “Eric S. Raymond questioned whether or not we still need reciprocal licenses like the GPL in a world where abusing open source code by making it closed source is pretty much a death sentence for your company. It’s simply never a successful strategy, he argues, and as such, how much sense does it make to have reciprocal licenses like the GPL?”
These statements show, why mr. Raymond is a typical figure in the open source initiative. For him, the proprietary software is just a strategy question; for the free software movement is it a social problem and free software (the GNU GPL license) is the solution for it.
Quotation: “That is why I don’t think you really need GPL or a reciprocal licenses anymore. It is attempting to prevent the behavior that the market punishes anyway. He added that the BSD license is a good alternative to the GPL.”
Users need and deserve freedom. Someone must defend these freedoms, this is why we need the GPL license.
Aha, except the GPL protects the code, not the users of it. Your argument is flawed. If you have software A which is licensed under any license bar GPL like, some company “steals” the code to “close” it and make money of it, tell me, how the hell are you, as a user, less free? The code is still there, not like it vanished after being downloaded by company X to make money of it. You can already and will be able to download, study, modify, run the code without problems at all. It really doesn’t matter if a billion companies around the world take the code and use it, as long as you can do it too.
The freedom you are talking about is not your freedom, is the freedom of something material, not of an individual. GPL accounts only for the code, not the programmer or the users. Look carefully: it tells you what you can do with the code, but also what you can’t. It’s in there, you can’t modify it and redistribute the compiled code without also making the changes public. So in fact, you are being restricted, as a user of the code, you are actually, take this, less free.
Free software is movement and philosophy. GPL and few others “constitutions” only describe it. Of course, it gives you some freedoms in cost of some limitations.
Like you are living in free country, but you dont have right to cheat, steal or kill anybody you want. Do you consider those limitations as a drawback?
It is about what is fair or what is not.
How are you going to guarantee those freedoms in other way than limit some other rights? Some rights will always be in contradiction to each other.
Anyway, sometimes it is better to give right to “kill other people”.
Well, it doesn’t give me any more freedom than other free licenses. Compare with BSD, for example. I’m still able to use and redistribute freely. So it’s not something only the GPL provides. So that argument about freedom for the user being what makes GPL interesting is pure bullshit. GPL only can guarantee that any changes to the code will be available for you to download, study, use and merge with the original source, which is fine. Other than that, I fail to see any real benefit.
So what would happen if some major project changed its license from GPL to say MIT? Would the project be any less free? Would you as a user be any less free? Not at all. What could happen is some player in the industry to use that code to build and sell a product, based on it, like Apple did with FreeBSD code. But look, Apple gave FreeBSD new life in a time where FreeBSD 5 really sucked. So how bad was it?
The only real reason to use GPL is when the developer really doesn’t want any other entity to profit from that code without giving back. That’s what makes the GPL attractive, and only that. And ESR is right about the current trends. We are moving to a software as a service style of computing. So offloading all that work to the open source community makes a lot of sense.
As Mr T. would say, “Protecting the code is protecting the users, fool!”
You must be very new to software to not know about historical cases like the Windows NT Kerberos debacle. Taking a standard, extending it, so that other non-NT Kerberos users cannot connect. If the code was open then the original Kerberos folks could modify their stuff with a “compatibility mode”. Code was closed though so it was not possible.
Who suffered as a result? sure the Kerberos “codebase” did but more importantly *actual users*. Could that bullsh!t happen with GPL, No Way!
And it is not only Microsoft that have done this. Other companies have tried to do the same thing. Make money off the work of the community (which is both allowed and encouraged) but not giving any modifications back (which is not).
Saying GPL only protects code not users shows how little you understand about both the subject and the shenanigans that have gone on in the past. It is not a matter that companies might exploit
GPL is not anti-capitalist (even if some adherents are). You can make money off GPL software and talented people do. It is against the erection of artifical barriers to preserve inefficiencies and exploitation by those with greater avarice than talent.
I agree with the last counterargument.
There are just two possibilities in such thinking. Your knowledge about this problem is worth to take another read on a few more essays, or you are simply a complete soul member of the ‘open source movement’. Why the quotation marks? Because it is not really a movement in the original meaning of the word.
You are really so intersted in the practical difference (although this is absolutely not the primary one)? The BSD license is a permissive free software license, that has been approved by the OSI initiative; the GNU GPL license is copyleft.
There have been a lot of definitions made, including by the free software movement as well as the OSI initiative, but some of them are more rational than others.
Open source software is a development methodology; free software is a philosophical, ethical and social movement.
“We’re NOT the Judean People’s League – we’re the People’s League of Judea!”
To be fair … that is not really a valid criticism.
Open source software ~= everyone has permission to read and use the source code.
Free software ~= everyone will always have permission to read and use the source code.
A small difference, perhaps, but nevertheless an important one.
Not to play into the JPF v PFJ, but I disagree.
The motivation behind open source is its proponents believe it produces superior software. That’s it.
The motivation behind free software is — free software. Superior software would be nice, but the aim is to be free.
I don’t really consider it an insignificant difference, but that’s just me.
Oh I know, I was mainly just being a smart-ass.
That said, yours is probably the most succinct explanation of free software that I’ve read – and the most hyperbole-free.
And GPL is not the only license able to assert this. In fact, any license that allows you to read source code and modify it, at least once in the time is all you need to make it free software. If you don’t think so, I can point you to the Firebird database, once closed and known as Interbase by Borland, then open sourced (forever) and then closed again. The source of that first open release was forked to give life to Firebird, which is free. The license sucks horribly.
There is a false asumption that GPL gives more freedom. What it gives, if anything, is more contributors (even extorted ones). But this is also highly dependent on the project. I doubt Postgres has/had any problems with some company selling a fork (EnterpriseDB was the case, I think).
All in all: you get your freedom the first time the project is released. After that, it’s up to the developers to keep maintaining it, improving and merging contributions. GPL guarantees further improvements are also released, even by greedy companies. But if the project is worth its salt, BSD/MIT licenses will keep any project running and updated.
Many say GPL keeps companies honest (by extortion, they aren’t really honest, you just happen to have something they need, but it comes with a price to pay). I would prefer to think BSD keeps honest companies within a project, they really don’t have to contribute, but they still do, so they are good and valuable. Sometimes it’s better to acuse them of greedy players. Big companies don’t like bad PR.
Oh, when people will stop calling Eric one of 3 big figures in open source? Why the hell anyone would put him in same line with ESR or Linus?!
PS: How many of you seen this news noticed there is January ELER out there?
Pretty crappy but still something… 
geekz.co.uk
I think you mean RMS, not ESR.
Sorry… but every time someone yanks out the “BSD License is more business friendly” card I can only think about Brett and his classic Bike Shed arguments.
If you don’t know about the Bike Shed or Brett Glass, YMMV.
Blahaha.
And *IF* Brett see the need to comment here;
Hi Brett, how is that ISP in Wyoming, that can’t compete with those closed source companies and also can’t handle P2P networking?
As someone who does embedded programmig in closed hardware platforms which don’t have any hope of ever being replaced by open ones (game consoles) GPL sucks.
Examples of this is trying to use Java/Javascript/Mono/etc on them. Sure, Java may be GPL, but if i want to do a closed implementation i still have to talk to Sun/Novell/Etc.. why is this necesary? It’s not like i’m taking java and making it close..
You don’t have to talk to anyone to use (GPL) Java on a platform. Oh, unless you want to take all that work for free and use it and close it up again. Is that what you want?
Don’t you get the problem here is you want to get something for nothing and then close it up again. The problem is not with GPL Java but with your company’s selfishness. If it doesn’t want to use GPL Java then you company should implement a version itself, or quit whinging about it. In fact, the GNU team are generous enough to give a Java implementation with “Classpath exception” so you can use it with your own stuff and leave your own code closed. Ain’t that real nice of them, so no need to kvetch about it, k?
What is it this gnu team what ever that is have done? It’s Sun that has opened java under the gpl.
I’m not sure if it has already been pointed out, but the example regarding Apple is reasonable. A good portion of OS X is closed source — including Cocoa and all the desktop environment. You don’t get that with Darwin. You can download Darwin or some Darwin-based distro (there are one or two I think), and if I’m not mistaken, OS X’s very own kernel — it’s just that nobody cares.
I’m not going to say anything about ESR, his reputation precedes him.
Edited 2009-03-24 22:02 UTC
I think the word that he said was [inaudible] was “Microsoft”.
-Ad
Edited 2009-03-25 06:06 UTC
is that people still invite him to talk, not that he says stupid things (he’s been doing that for a LONG time).
Has he done anything even remotely important or interesting after the cathedral and the bazaar?
Apple’s Darwin is a perfect example of how open source doesn’t work. Yes, sources are available and the Darwin kernel is very similar (but not identical) to the Mac OS X kernel, but at the same time, Darwin’s sources are read-only. Apple has proven many times to just ignore contributions.
If developers next to never can get their contributions into Mac OS X and into the machines of a wider audience, why would a person consider contributing something at all? As a result, contributions stalled and neither a GPL nor a BSD-style licence will help.
An example of where Apple does better is gcc. gcc is strong enough to exist independently of Apple’s mercy.
from Microsoft.
“But now, lets look at the implications of taking this seriously. The question I found myself asking is: if the lynch mobs punished people for taking other people out, then why do our laws need to punish people for homicide? That is why I don’t think you really need homicide laws anymore.”
He’s made himself really easy to pick on.
Unless it’s legally prevented by the license, it’s allowed.
By leaving things to the market, you’re creating a mess of lawsuits. It’s better to codify the reciprocal nature of open source and free software into the license itself rather than to drop those terms in the hopes that business will be more receptive to non-reciprocal license.
Some companies are more receptive to this precisely because they can play the games that Microsoft and others are currently playing. We are not to the point yet where we can afford to lean back and relax what we’re doing. We must keep the GPL, LGPL and licenses like it.
GC
I don’t know enough about the different licenses, but I have to agree that a number of managers are scared of being forced to open up company specific software that took a lot of money to create. I think the community has to find a solution for that.
People who don’t know enough about it can have a wrong image of the community and licenses don’t help in that, although they are certainly needed. I think that it is clear that a free market doesn’t work, unless your idea about a good market is making the rich richer and the poor poorer.
I disagree with ESR on the whole GPL thing, in part because I belive in diversity (having both GPL and BSD means better chance for open code’s overall survival.) Also I think BSD is more in line with my comments, cheap, open and I have no intention of controlling how you use my ideas. A GPL version of my comments might try to make sure any replys to my comment are not hidden and public (not via private email etc.) I think they serve different and needed purposes.
Edited 2009-03-25 21:46 UTC
“I disagree with ESR on the whole GPL thing, in part because I belive in diversity (having both GPL and BSD means better chance for open code’s overall survival.)”
one could argue we need bsd, gpl and *proprietary* because the proprietary gives open source a run for its life. We end up with better open source software as a result so thanks Microsoft and Apple for improving linux and firefox!
I can’t say yea or nay on the whole IQ thing. We can measure speed. http://en.wikipedia.org/wiki/100_meter_dash How many whites do you count here. Sure there are some in the female camp, but that’s also probably a genetic thing. See any Asians? We know for sure color is hereditary. I can see where basketball might be the way it is because of environment etc. I know there are lots of white people who join track and field. So not every ‘race’ is equal at least in certain specific physical traits. I think to say with any authority, you would specifically have to test in a controlled way for intelligence differences between races. Throwing around IQ w/o factoring in environment is certainly flawed. But environment and IQ may be circular, hence the need for controlled testing. Which I haven’t seen.
GPL: ESR No Longer Needed
To me, the only free licence is the WTFPL – Do What The f–k You Want To Public License. cf: http://sam.zoy.org/wtfpl/
GPL is really too much restrictive, for no results. It does not even forced to open all of the most used graphic drivers.