Survey Says GPLv3 Is Shunned

willdavid writes in to note a survey of open source developers conducted by Evans Data that indicates a real rift in the community over GPLv3. The survey was based on in-depth interviews with 380 open source developers and no estimated margin of error was given. "Just 6 percent of developers working with open-source software have adopted the new GNU General Public License version 3... Also, two-thirds say they will not adopt GPLv3 anytime in the next year, and 43 percent say they will never implement the new license. Almost twice as many would be less likely to join a project that uses GPLv3 than would be likely to join... [Evans Data's CEO said] 'Developers are confused and divided about [the restrictions GPLv3 imposes], with fairly equal numbers agreeing with the restrictions, disagreeing with them, or thinking they will be unenforceable.'"

Comments

Slow adoption is to be expected

[A beautiful mind]

It is much easier for new projects to start out with GPLv3 than old projects to convert. Unless the committers transfer copyrights to a central body like in the case of the gnu tools and FSF, it is hard to move to another license if not bordering on impossible.

Re:Slow adoption is to be expected

[stinerman]

Not to mention any project with files licensed under GPLv2 or later is, for all intents and purposes, GPLv3 anyway.

Re:Slow adoption is to be expected

[JesseMcDonald]

Unless the committers transfer copyrights to a central body like in the case of the gnu tools and FSF, it is hard to move to another license if not bordering on impossible.

Unless, of course, all the commits were "GPLv2 or later", in which case the project was effectively already under the GPLv3 from the moment it was released.

Re:Slow adoption is to be expected

[fotbr]

Not necessarily. GPL version 3 only provisions do not apply to it, unless it is changed to be licensed under GPL 3 (only, or "or later").

Re:Slow adoption is to be expected

[Cobalt Jacket]

Not everyone is as eager to regress as you are.

Re:Slow adoption is to be expected

[stinerman]

Obviously.

My point was that all these projects can be counted as GPLv3 projects, or is it that important that I formally fork such a project to be counted in the numbers?

Re:Slow adoption is to be expected

[hansonc]

wrong.

If I'm the user of the code e.g. Tivo and I don't decide that I want to comply with gpl v3 I don't have to in that case. For you to force me to comply with v3 you have to relicense it as v3 (or later) it's not a retroactive license which probably wouldn't be legally enforceable anyway.

Re:Slow adoption is to be expected

[huckamania]

I think this attitude is why so many developers are turned off by the GPLv3.

Is there a clause in the GPLv3 that makes the "or later" mandatory? If that's the case, might as well sign it all over to the FSF or better yet just put "This software is released in whatever manner RMS decides at any time now or in the future".

Still, I wonder about the legality of enforcing a license that doesn't exist or didn't exist when you first got the source. "This software is released under a future license which we will let you know about when we get around to it" doesn't sound very legal.

Re:Slow adoption is to be expected

[Experiment 626]

Unless, of course, all the commits were "GPLv2 or later", in which case the project was effectively already under the GPLv3 from the moment it was released.

Wrong. The "or later" does not mean that whatever the most recent version of the GPL has been published is the one that applies. It means someone wanting to copy / distribute / whatever the software is free to do so under the terms of the GPLv2, or any later version that they might prefer the terms of. If the GPLv4 came out next week and said "to distribute software under this license, you have to send RMS a case of beer", you could distribute "GPLv2 or later" software by either providing its source (the GPLv2/v3 option) or by sending RMS a case of beer. New versions of the GPL give you more choices in licensing "or later" code, they don't retroactively change the terms of the deal like some shady EULA.

Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation.

Re:Slow adoption is to be expected

[cnettel]

As one of the stated intents of GPLv3 is to close several loopholes, I think you should fork if you find the new license important.

Re:Slow adoption is to be expected

[Ajehals]

IANAL, but IFAICS It's an option for the recipient of the software, you can comply with the terms of Version X or if you prefer, a later version. When you are distributing the software you would be in a position to choose which is more suitable. It is *not* a case of the copyright holder (usually the developer) or the FSF being able to turn around in 5 years time and tell you that your distribution of the software is longer legal, unless you somehow fork the code and have it re-licensed, or if you have included code that is compatible with the earlier license (and the earlier license is compatible with) that adds some extra stipulation that you disagree with.

As much as the anti-GPL / anti-oss crowd would like you to believe the GPL isn't an instrument that will force you to do something that you haven't agreed to, more-over if you don't like the GPL ignore it, after all however you look at it, it is a license that gives you more rights with regard to the software you have not less, if you don't like the GPL then use the software and be bound under normal copyright law with regards to anything else (i.e. don't distribute.

Re:Slow adoption is to be expected

[fotbr]

They SHOULD all be counted as GPL2, because until they are explicitly moved to GPL3, they are not GPL3.

Re:Slow adoption is to be expected

[chromatic]

Is there a clause in the GPLv3 that makes the "or later" mandatory?

No, nor the GPL v2. The FSF recommended including that clause, but it's silly to release your code under a license you haven't seen.

Still, I wonder about the legality of enforcing a license that doesn't exist or didn't exist when you first got the source.

That's obviously impossible. How could anyone agree to a license that doesn't exist? Anyone releasing software with the "or later" clause believes sufficiently that any new version of the GPL released by the FSF will be in the same spirit as the existing versions. More power to them. That doesn't mean that if they distribute software to you right now that you have to agree to the GPL v4. (Nor do you have to agree to the license to use the software, per the wording of the license itself.)

Re:Slow adoption is to be expected

[JesseMcDonald]

I know that. I meant the code is released under the GPLv3 in addition to whatever other licenses apply (like the GPLv2). Of course it's still available under the GPLv2 terms as well, at least until someone accepts a non-GPLv2 patch (GPLv3-only / GPLv3-or-later).

Re:Slow adoption is to be expected

[Anonymous Coward]

"to distribute software under this license, you have to send RMS a case of beer" Does RMS drink beer? I mean, would he even appreciate that?

Re:Slow adoption is to be expected

[swillden]

Not necessarily. GPL version 3 only provisions do not apply to it, unless it is changed to be licensed under GPL 3 (only, or "or later").

One key clarification here: You can't change the license on a piece of code unless the license gives you permission to sub-license. Otherwise, only the copyright owner can change the license on a file from, say, GPLv2+ to GPLv3+. Someone else can come along and add some GPLv3+ code to the GPLv2+ file, and the result will only be distributable under GPLv3+ but this doesn't mean the original GPLv2+ code has been relicensed. If the GPLv3+ code is removed, then the result will again be distributable under the terms of GPLv2.

Re:Slow adoption is to be expected

[Spy der Mann]

Wait, doesn't that depend on the distributor? i.e. GPLv2 or later, at your option...

I remember reading some of that somewhere.

Re:Slow adoption is to be expected

[GreatBunzinni]

Unless the committers transfer copyrights to a central body

Not so fast. In some countries it is not possible for an artist/author to transfer or give away his copyright ownership. That may be allowed in some countries but it is not an universal thing. Free software, on the other hand, is and therefore every project must take in consideration the laws which each developer is bounded to.

Re:Slow adoption is to be expected

[dosius]

Important, yes, important enough that I insist on staying with 2, or using a more liberal license.

-uso.

Re:Slow adoption is to be expected

[Bill, Shooter of Bul]

That is the crux of Microsoft's assertion that they are not required to follow GPL3, if customers had bought the SUSE licenses from them at a time that all of the code was GPL2. Its a bit trickier, because they can then exchange the licenses for GPL 3 software from novel. Basically I think the whole thing is a logical paradox, that doesn't really have a right answer.

Re:Slow adoption is to be expected

[Anonymous Coward]

If the GPLv4 came out next week and said "to distribute software under this license, you have to send RMS a case of beer", you could distribute "GPLv2 or later" software by either providing its source (the GPLv2/v3 option) or by sending RMS a case of beer.

Thanks. That's just about the best example I've ever seen illustrating why the "or any later version" clause is a terrible idea that nobody should ever use.

A more pertinent version might be this: suppose in GPLv10, there is a clause allowing binary-only distribution of work derived from GPL code, as long as you donate $N to the FSF (for some sufficiently large value of N).

Of course it's impossible to imagine this happening in the next 20 years or so. But the length of a copyright term is such that GPL revisions will continue to be released long after RMS is dead. The "Or any later version" clause should give anyone who thinks about it the heebie jeebies.

Re:Slow adoption is to be expected

[Brandybuck]

It depends on how that "or later" clause was phrased. If it uses the standard FSF phrasing, then you can redistribute and modify under the terms of the GPLv3, but... you still don't have permission to relicense. Unless the authors change the license, it will always remain "GPLv2 or later". New code (including significant modifications) can be under GPLv3, but only the original authors can change the licensing of the original code.

Re:Slow adoption is to be expected

[huckamania]

Thanks for the responses, I really didn't know. I'll continue to legally take header files and structures that implement standards, as there is no copyright protection for them, irregardless of the license. I'll also continue to report bugs and bug fixes to the projects that I have to interract with, irregardless of the license.

Got to keep protecting the 5th freedom, which is to not give a damn.

Re:Slow adoption is to be expected

[Ed Avis]

The worst that can happen is that the FSF releases a licence that is too liberal and allows things you don't want to permit. Even if fifty years from now the FSF is bought by Warren Buffet's head in a jar and decides to monetize the base of GPL'd software, you haven't really lost anything and neither have the users since the original version is still usable.

More worrying, to me, is that the original author dies or becomes uncontactable (or just one of several copyright holders dies) and the software cannot ever be relicensed to a new GPL version. That is why I support the 'at your option, any later version' formula. It doesn't hurt to be a bit too liberal.

Re:Slow adoption is to be expected

[DragonWriter]

My point was that all these projects can be counted as GPLv3 projects, or is it that important that I formally fork such a project to be counted in the numbers?

Well, yes, if it is not currently licensed with the restrictions in the GPLv3 (but merely allows other people to relicense their own redistribution that way), it is inaccurate to describe it as a GPLv3 project. I mean, by your argument, every project under a GPLv3-compatible license (or, presumably, in the public domain like SQLite) should be counted as a GPLv3 project because someone could conceivably redistribute a derivative of it under the GPLv3 at some point in the future.

Re:Slow adoption is to be expected

[Anonymous Coward]

Why would people move code to the GPL 3? It ensures that the code will be shunned by any commercial entity. For example, if the Linux base was GPL v3, most companies would just ditch it, cross MS's palm with the $3 per Windows CE device license, move to that, and forget about Linux (and F/OSS in general) adoption entirely.

In the late 90s, businesses were SCARED TO DEATH of the GPL v2, until attorneys everywhere reassured corporate types that no, corporate secrets do not have to be given out because they are running their stuff on a Linux machine.

Now GPL v3 comes along and to a lot of people does exactly that. The GPL3 hinders Linux adoption in the commercial sector, as businesses will have to either re-evaluate if their trade secrets must be publicized, move to BSD, or give the MS salesperson a call, pay the fee of entry and not have to worry about redistribution of trade secrets, licensed code modules which cannot be redistributed, or license conflicts.

For embedded stuff, if critical Linux modules go GPL v3, MS will be on a gold mine with Windows CE when people are forced to switch due to legal reasons (For example, an industrial company distributing a pump which does a confidential chemical mixture injection timing to an industrial customer. Under GPL v3, the trade secret timing of the chemical process which took years if not decades of research has to be given to the customer.)

Of course, this is not advocating DRM, but companies do like their trade secrets, and if some company has some innovative manufacturing process they developed, they are not going to be turning that source code over to anyone that demands it, because the process happens to touch a GPL v. 3 code module.

Re:Slow adoption is to be expected

[wzzzzrd]

[...]or by sending [me, though i am NOT rms] a case of beer.

thank you for pointing out the actual power of licenses. all i have to do now is to write an astonishing piece of software that everyone wants to use and i'll never run out of beer anymore. compared to ms licenses, this is a piece of cake.

Re:Slow adoption is to be expected

[Anonymous Coward]

Your original post is wrong because you said they were effectively GPLv3, when that is not the case. They are effectively GPLv2, since that license is less restrictive. They are not even really technically GPLv3 until someone explicitly redistributes as GPLv3.

The only thing that can really be said about "GPLv2 or later" projects is that they can be easily forked to GPLv3.

Re:Slow adoption is to be expected

[swillden]

Wrong. The "or later" does not mean that whatever the most recent version of the GPL has been published is the one that applies. It means someone wanting to copy / distribute / whatever the software is free to do so under the terms of the GPLv2, or any later version that they might prefer the terms of.

True, but it's very easy to change GPLv2+ code to GPLv3+ -- just add some changes that are licensed only under GPLv3+. That doesn't actually change the license on the older code, but it does make the combined package distributable only under GPLv3+.

Re:Slow adoption is to be expected

[sumdumass]

I guess the idea behind this story and the point you are making it one of the same. If the GPLv3 was being accepted and welcomed, you would have changed the license already. But apparently, your enthusiasm consists of letting someone down the road do it if they want to.

So while you can take that option, as long as there isn't a fork or a specific license change, then the protections of the GPLv3 aren't in effect. If I can take the work and ignore the GPLv3 license, then it isn't automatically counted as GPLv3 either.

On another note, If you move something to the GPLv3, without regard to the later versions clause, anything you bring to the GPLv3 table means you are certifying the patent issues to some extent in order to place the code under the GPLv3. This might open some new liabilities if some patent issues exist. I would be cautious about moving something just for the sake of moving it. At least get some proper legal advice from someone without an interest in the GPLv3 or it's success.

Re:Slow adoption is to be expected

[sumdumass]

I thought the "or later versions" clause gave you or me permission to move the license to a later version at our discretion if it hasn't already been done.

Am I reading that wrong? There was a lot of over rated discussion about forking the linux kernel and imposing a GPLv3 license on it a while back until it was clarified that it is GPLv3 only. But of course this is just people talking so it could just be that.

Re:Slow adoption is to be expected

[einhverfr]

IANAL, but I don't think that is what it means.

Basically the author is allowing people to use and distribute the software under either license.

Think of it this way: The BSD License addresses all downstream users and provides permission to all those who receive the source code. You can't just change the license because only the copyright holder has that right. You *can* take the code and integate it with code under other licenses. Same holds with GPL v2 or later.

Now, if someone took my GPL v2 or later programs and released them under the GPL v3 without forking them, I would consider that a problem. If they forked them (and indicated that they were splitting off), that would probably ask them to be more clear and state that the original version was GPL v2 or later but I probably wouldn't push the issue too much beyond that.

Re:Slow adoption is to be expected

[sumdumass]

Novell holds an "out" for MS.

The vouchers only cover "SuSE enterprise linux" server and desktop products. Novell can and probably will end of life those product when it becomes a hassle with GPLv3 covered works and release a separate product not covered by the vouchers. It would likely be something like "Enhanced SuSE enterprise linux"- Now with GPLv3 support!

There you go, problem solved and neither Microsoft or Novell suffers from the GPLv3 problems or woes.

And as long as Novell Keeps the latest versions of the end of life enterprise linux product available for someone to get updates from, the vouchers with no expiration date don't lose any value at all.

Re:Slow adoption is to be expected

[absoluteflatness]

More worrying, to me, is that the original author dies or becomes uncontactable (or just one of several copyright holders dies) and the software cannot ever be relicensed to a new GPL version. That is why I support the 'at your option, any later version' formula. It doesn't hurt to be a bit too liberal. Another good reason why copyright should expire at, or soon after, death. If the original author is long dead, then the code would just fall into the public domain.

Re:Slow adoption is to be expected

[stinerman]

Touche.

Re:Slow adoption is to be expected

[donaldm]

Actually the Article is not that clear who the 6% of developers that are adopting the GPL3. If they are Redhat and IBM and they rather like the GPL3 http://www.news.com/8301-10784_3-6171921-7.html then you have nearly 40% of Linux development. Throw in HP and possibly Novell then you can add another 5% to 10% more. I can't see SUN going down the GPL3 path but then you never know. Can anyone shed any light on this because the article does not really say that much although I did find the source of the data at the Evans Data Corporation site http://www.evansdata.com/ but again that was not that helpful since you have to have to register.

Re:Slow adoption is to be expected

[vagabond_gr]

If the GPLv4 came out next week and said "to distribute software under this
license, you have to send RMS a case of beer", Hum... this poses new philosophical questions. Would that be "free as in free speech" or "free as in free beer"? I guess it would be "free only with a beer" for you and just "free beer" for RMS.

Re:Slow adoption is to be expected

[reebmmm]

This is the most specious of arguments. Not only is it wrong; it's entirely misleading.

In this case, "or later" allows the distributor to choose how to distribute the software. GPL's contractual rights and obligations run to the person doing the distribution. The originator will only be bound by GPLv2. Anyone distributing the software after being distributed under GPLv3 will be bound by the new contract.

So no, it is not effective "already" under the GPLv3.

Arguably this leads to its own problems. You could "effectively" have two (or, in the future, more) different licenses applying to the same project.

Re:Slow adoption is to be expected

[sumdumass]

IANAL either, but I firmly believe you shouldn't have to be one in order to understand a license means for the masses.

I think the BSD situation is inherently different because of the GPL's express desire to use different license versions if the author states it when licensing it. It wouldn't be "relicensing" perse, as it would be using the License (or later version) as the original copyright holder wanted.

Further, I would think the requirements to retain the license (GPL) for any additional changes/ derived works and the No further restrictions clause would mean that you did not have the options of changing the license or using more restrictive licenses with the covered works. It is as if the express permissions (the "or later versions" clause), if stated in the "this product is licensed as" sections, would mean that you have permission from the author/copyright owner to relicense as needed for your GPLed product.

Well, maybe relicense is the wrong terms. It would be more like using the authors pre-approved alternative license. And seeing how the copyright holder already gave permission to use the different license versions, you would think the steps of asking if you could use a different license and therefore change the license has already taken place with the authors "or later versions" consent in the original licensing.

Of course I could be wrong on this. If you are correct in that I could use the GPLv3 with my additions to the project instead of changing the license to the GPLv3 to fit in my project, we are going to have a really bad nightmare in keeping things/code and applicable permissions straight.

Re:Slow adoption is to be expected

[swillden]

I thought the "or later versions" clause gave you or me permission to move the license to a later version at our discretion if it hasn't already been done.

I don't see how it could. There is no language in the license that permits relicensing, and the "any later version" text recommended by the FSF and used by most GPL software:

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 2 of the License, or
(at your option) any later version.

... says you can redistribute or modify under the terms of the specified or any later license. It doesn't say anywhere that you can remove an older license from the options, and in the absence of any language allowing such license alteration, you can't.

Keep in mind that for software licensed with GPLv2 and the "any later version" language, the license is GPLv2 + GPLv3 + ... + GPLvN, not GPLv2. That is, the actual license on the code is the union of v2 and all future versions. Removing one of the options isn't just exercising the stated permission to select the GPL version terms you want to use, it would constitute *changing* the license, and only the copyright holder can do that (or authorize someone else to do it). You can distribute under the GPL version you like, but all downstream distributors have to have the same set of options that you did.

As I said in my previous post, that's different from adding v3(+) code to the program. Doing that effectively turns the combination to v3(+), respectively, because the code licensed by its author as v3(+) can't be distributed under v2. It doesn't change the license on the v2+ code, and anyone who wants to remove the v3(+) code from the program can use the v2 terms, but the only way to satisfy the licenses on all of the contained code is to use v3(+).

Note, though, that IANAL, and I'm parsing this language based on what I think the words mean, which may not be what lawyers and judges would understand them to mean.

Re:Slow adoption is to be expected

[sumdumass]

Ha, I was getting ready to argue this and then I realized that we are discussing different process evolutions.

If I understand your argument correctly, your saying that I can't just say that the covered work is now GPLv3 without doing something to separate that work from your original and intent if you are the content owner/copyright owner. And yes, I would agree with this. There is nothing arbitrarily imposing the GPLv3 on to your project even if it says "or later version".

But I can take your code licensed in this way and use it as GPLv3 software effectively locking you out of my changes. And while I don't see a requirement to call your code something else when I do so, I do see the confusion and frustration when not doing so. And my using the GPLv3 isn't really me changing the license, but rather just using the specific version you offered it in.

Do I understand this correctly now?

Re:Slow adoption is to be expected

[swillden]

Yes, I think we're in agreement here.

This stuff is funny in that, at bottom, it's pretty darned simple: The author of a piece of code gets to specify the terms, no one else. If you combine pieces from two different authors, the result must comply with both authors' terms. That's all there is to it. Everything else is just examining the terms to figure out what they mean.

In practice, though, it takes a lot of work to wrap your head around the issues, and it's really easy to get it wrong, in spite of the simplicity.

Re:Slow adoption is to be expected

[einhverfr]

I think the question is one which is actually fairly simple: if the author releases under GPL v2 or later, when is it acceptable to change the license of the work to GPL v3 or later?

We both agree that either license may be "used." The question is which licenses must be passed on, and when.

Obviously changing it without even intending to add copyrighted elements seems to pose problems. Obviously misrepresenting what permissions the author had granted to the work would seem to pose problems. If one wanted to be *really* hard-nosed (like Theo), one could insist that the changes in license occur after substantive additions, but this seems like a difficult line to draw and not something worth fighting over.

Hence I would probably simply ask for an acknowledgement that the work was based on my work which was released under the license I said it was released under. I might ask for other things too, such as mention of the version number forked from, etc. But if people refused, I probably wouldn't fight about it too much. After all, building a project is a lot of work and it is more important to do this rather than fight about forks which in all liklihood won't last a year.

Re:Slow adoption is to be expected

[bzipitidoo]

I don't like tying copyright terms to death. If someone wanted something you wrote, one way to get it is for you to have a fatal "accident". Or, if you have heirs who stand to lose a lot of money by your early demise, and something happens to you, do you want to spend the last years of your life on life support or some other situation where your quality of life is zero? A fixed term length from the date of publication avoids those potential problems.

Re:Slow adoption is to be expected

[absoluteflatness]

This is a good scheme too, but finding a good length for copyright to last is tough. One problem is the sheer number of things copyright covers. For artistic works (under the traditional definition of "artistic") a somewhat large time would seem to be appropriate, but at even comparatively short lengths, code maintains its copyright past what many would consider to be its useful life. If you say, fix for 20 years, we're only getting access to code from 1987 and before. Still, anything's an improvement from the current state of affairs, where I'll be long dead by the time any code falls into the public domain.

Re:Slow adoption is to be expected

[dfghjk]

A "GPLv2 or later" license is not a GPLv3 license. In order for such code to be counted as GPLv3, it would need to be relicensed and that would be allowed. Restrictions placed on code under GPLv3 do not apply to "GPLv2 or later" code unless they also exist in GPLv2.

Re:Slow adoption is to be expected

[nuttycom]

Man, that gives me the best idea ever! The next piece of open-source software I write is going to be released under the BPL (Beer Public License) - in order to distribute, send me a six-pack!

This is gonna be great.

Re:Slow adoption is to be expected

[Anonymous Coward]

As I said in my previous post, that's different from adding v3(+) code to the program. Doing that effectively turns the combination to v3(+), respectively, because the code licensed by its author as v3(+) can't be distributed under v2. It doesn't change the license on the v2+ code, and anyone who wants to remove the v3(+) code from the program can use the v2 terms, but the only way to satisfy the licenses on all of the contained code is to use v3(+).

This I wonder about... Can you really do this? GPL2 says:

2. You may modify your copy or copies of the Program or any portion
of it, thus forming a work based on the Program, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions: ...
b) You must cause any work that you distribute or publish, that in
        whole or in part contains or is derived from the Program or any
        part thereof, to be licensed as a whole at no charge to all third
        parties under the terms of this License.

(emphasis added)

How does GPL3 get around that? If I receive code licensed under GPL2 it would seem that there can't be portions of the program that are available only under ANY other license that are part of the Program.

Re:Slow adoption is to be expected

[swillden]

How does GPL3 get around that? If I receive code licensed under GPL2 it would seem that there can't be portions of the program that are available only under ANY other license that are part of the Program.

GPL3 doesn't have to get around it. Keep in mind that the GPLv2 language is subsidiary to the language of the license statement attached to the files. In the scenario that we're discussing, that language says that the terms of GPLv2 or any later license may be applied to the covered code. So it doesn't matter if the terms of added code are incompatible with some element of v2, as long as they're compatible with everything in one of the "or later" GPL versions, they're compatible with the license on the code.

Re:Slow adoption is to be expected

[sumdumass]

I guess after thinking about it, In the case of the GPL, the second license must be passed on after someone else does something with it. You don't accept the GPL when you get the covered works, only when you plan on doing something with it that the license requires acceptance for. So your right in that you couldn't impose a license on me over my works. But If I did the "or later versions clause", you could impose the later license on your later associations with the project.

I guess this is going to automatically cause a fork of the most confusing kind. If I move your "plib" (project library) or whatever project to the GPLv3 without naming it something else, we could end up with plib GPLv2 and Plib GPLv3 with one being different and incompatible then the other. I wonder if or how that would play out?

one word

[Anonymous Coward]

SAMBA

Confused...

[advocate_one]

who actually commissioned the study??? cos studies don't just happen by themselves... they cost money

Re:Confused...

[BibelBiber]

Really? What about student projects? They do studies all the time and they even have to pay for it (so they get some feedback).

Re:Confused...

[fymidos]

They might have done it for their portfolio:
"... and you can see here a survey that proves that GPL v.3 is a bad thing, with our expertise on the subject we can provide similar quality studies on GPL 2, and even GPL 4 if you so desire"

Re:Confused...

[marcosdumay]

That one didn't seem to cost a lot. But also didn't get very usefull results.

The number of interviwed developers is a joke (at least for supporting that conclusion). There is no information about how those developers were selected (or anything else, in fact there is no usefull information at TFA).

Re:Confused...

[Plasmic]

I was also skeptical, but then I Googled. My quick analysis is that Evans Data specializes in developer community research and that most of their research has resulted in pro-OSS results, if anything. For example, here are the titles of previous press releases:

• Nine Out Of Ten Linux Developers Refute Sco's Linux Lawsuit
• Evans Says Java Is Catching Up To .NET
• Linux Adoption Not Slowed by SCO Lawsuit
• Access to source code is the "primary motivating factor" in operating system adoption among embedded systems developers

(They also did one on KDE vs. Gnome, but I didn't take the time to parse the results.)

Anyhow, my sense is that this is more legit than not. Regarding sample size, they opted for in-depth interviews on lots of topics (GPLv3 was but a small piece of the study, it seems) instead of spamming a few questions to lots of folks. For the specific topic of GPL v3, I'm sure everyone would agree that a focused survey with a large sample size would likely be better ... though it could be less accurate as you begin to include people that aren't actually involved in license decision-making or OSS development.

I concluded that it shouldn't be dismissed out-of-hand, but someone should dig deeper.

Re:Confused...

[Anonymous Coward]

Evans Data Corp - Clients
http://www.evansdata.com/company/clients.php

Evans Data Corp - Advisory Board
http://www.evansdata.com/company/about.php

If you want more specific than this; write to them:

740 Front Street, Suite 240
Santa Cruz, CA 95060

Phone: 831.425.8451
Fax: 831.425.7913

Public Relations
Phone: 831.316.0072
Fax: 831.425.7913
pr@evansdata.com

Re:Confused...

[Anonymous Coward]

Just 6 percent of developers working with open-source software have adopted the new GNU General Public License version 3... Also, two-thirds say they will not adopt GPLv3 anytime in the next year, and 43 percent say they will never implement the new license.

6% plus two-thirds (67%) plus 43% just about covers everyone.

Re:Confused...

[Goaway]

Those 67% and 43% do not refer to different people, genius.

Re:Confused...

[Fred_A]

If you want more specific than this; write to them:

740 Front Street, Suite 240
Santa Cruz, CA 95060 Wait; you means Evans corp is a Santa Cruz Operation ??

Re:Confused...

[Anonymous Coward]

3 guys in an apartment over a bookstore and a $5 barbershop... meh.

Re:Confused...

[Anonymous Coward]

Look like a front company to hide who the real client is. If they want me to believe anything they should just say upfront who is the client.

Re:Confused...

[bmcage]

Did you not get the mail on your sourceforge devel list?

We of course replied that a devel mailing list is no place to solicit for help in some shady study...

Fork

[BibelBiber]

To quote an earlier article: "If you don't like it fork it."

Re:Fork

[WK2]

To quote an earlier article: "If you don't like it fork it."

That just may be the dumbest idea I have ever seen on slashdot. You can't fork a project because you don't like the license. In some cases you can, if the license allows distribution and is not copyleft, but GPLv3 is copyleft.

Re:Fork

[maxwell demon]

If there also exists an earlier, GPLv2 version, you can start your fork with that. Of course, if the project started out as GPLv3, you're stuck with it.

Re:Fork

[Anonymous Coward]

If you don't like it, you can go fork yourself.

Re:Fork

[empaler]

To quote an earlier article: "If you don't like it fork it."

That just may be the dumbest idea I have ever seen on slashdot. You can't fork a project because you don't like the license. In some cases you can, if the license allows distribution and is not copyleft, but GPLv3 is copyleft.

You must be new here. The bathrooms are over by the water cooler, there's free fruit in the lunch room, and we've just had a free soda tap installed by the coffee machine. Don't leave your ID card anywhere, always take it with you. Remember to lock your computer or log out before you leave your work station, or you can be sure you've sent mails to the main company mailing list offering pizza, drinks, sexual favours, or all of the above.

Also, don't call the Operator unless you're suicidal.

Re:Fork

[VGPowerlord]

Also, don't call the Operator unless you're suicidal.

Does Simon still work there?

Re:Fork

[BibelBiber]

Sorry for not being precise. I actualy meant th GPL, not the project itself. If you have a problem with V3 why not fork it into a different but similar license that only includes your favourite ideas.

Remember!

[heinousjay]

Those restrictions are for your freedom. It is important to take freedom away to protect it. Truly allowing freedom would allow freedom to be taken away, and we can't allow that, so we've taken away some freedom to allow true freedom to flourish.

I can't imagine why anyone wouldn't understand that perfectly.

And I'm sure I'll get modded down, but before you do that, read through my first paragraph carefully and tell me what I've said differently than the GNU people.

Re:Remember!

[cromar]

GPL protects the freedom of the code, not the freedom of the developer. Big difference! If you want developer freedom, use the BSD license or some such. Different tools for different problems :)

Re:Remember!

[Anonymous Coward]

It's not about how accurately you can quote the FSF. It's about whether or not it's right to have a software license place restrictions on hardware.

Re:Remember!

[Lost Found]

What the GPL is really concerned with beyond the code is protecting the freedom of the code's user. BSD aims to give the initial recipient of BSD-licensed code the freedom to make copies and changes in virtually any way they want whereas GPL aims to give those same freedoms and enforce them in second and third and fourth order copies, etc.

Re:Remember!

[Rycross]

I don't think the characterization is fair. You are talking about different freedoms, and the trade offs between them. Not all freedoms are equal, and not all freedoms are good. If you'll excuse the hyperbole, in an anarchy, you have the freedom to murder and steal, but I don't see many people clamoring for those freedoms in our society. I'd imagine the FSF has made a decision about which freedoms are more important. "Software freedom" isn't quite as clear cut a decision as my example, and is largely dependent on your opinion.

Re:Remember!

[Anonymous Coward]

So... I get it now... GPLv3 is the next Patriot Act. Oooh... I got another one, how about we give Microsoft and Mayor Bloomberg all of our guns too. That will certainly help us out during the next national disaster when the gov. is unable to immediately restore order.

The whole reason this country was founded was because they finally figured out you don't protect rights and liberties by taking them away.

Re:Remember!

[heinousjay]

The thing is, I don't disagree with you. I'm fine with the GPL in any form, because frankly, it's the author's decision to use it and I respect that. My point is that it's a confusing muddle to wade through, particularly given the FSF's propensity (and RMS in particular is guilty of this) for playing "clever" wordgames with everything that they touch. There's a point where clear talk about the realities of the situation serve the cause better than rhetoric designed to spin the upsides and hide the downsides, which is very much what most explanation of the GPL come down to.

I'm not saying don't use it. I'm not even saying it's bad. I'm saying it is murky, and the article makes perfect sense to me because of that murkiness.

Re:Remember!

[Anonymous Coward]

Those restrictions are for your freedom. It is important to take freedom away to protect it. Truly allowing freedom would allow freedom to be taken away, and we can't allow that, so we've taken away some freedom to allow true freedom to flourish.

I can't imagine why anyone wouldn't understand that perfectly.

And I'm sure I'll get modded down, but before you do that, read through my first paragraph carefully and tell me what I've said differently than the GNU people.

That first paragraph was damn funny.

Re:Remember!

[A beautiful mind]

You're essentially correct though I suspect you were trying to express sarcasm in your post.

The government takes the freedoms of killing, torturing and in general harming people away (the cynical might say to monopolize them for itself, but that's another subject), but few could argue that we are less free because of that.

The whole BSD vs. GPL issue stems from different viewpoints. The BSD/total freedom camp approaches freedom from the individual viewpoint, while GPL approaches it from the side of the common good.

From an individual viewpoint, freedom is decreased by not being able to kill, but from a community's view freedom is increased because of a highly philosophical and "zealot" viewpoint about the importance of life is uphold. The highly philosophical theories about the sacredness of life coincide* with the fact that if noone is allowed to kill, everyone is better off, which of course translates to individual good aswell. What is good for a community of people, is bound to be good for the majority of the individuals making up the given community after all...

*Moral right or wrong doesn't automatically translate into advantage in scientific terms, but in the case of GPL, sharing and being open is not only morally good but is also backed up by mathematical research and empirical observation telling us that by cooperating everyone gains and being selfish leads to temporary advantage of individuals, but consequently everyone is worse off. If the party wishing to cooperate follows the classical tit-for-tat strategy, the end result will be a good for everyone situation in the case of a cooperating environment and a very slightly worse off case for the cooperating individual if the environment is hostile. In case of a mixed environment, cooperation wins.

Re:Remember!

[maxwell demon]

It doesn't restrict hardware. You can make your hardware DRMed like hell. You just can't run GPLv3ed software on that DRMed hardware. That's a restriction on the software (don't run it on DRMed hardware).

Re:Remember!

[heinousjay]

I was expressing an honest concern, and trying to do so in a memorable way. It's not that I disagree with the GPLv3 (or any version, really.) Personally, I think it's a fine license, and I understand it and its goals, or at least I believe I do. That uncertainty, however, is the kicker.

Ultimately, the GPL has begun the process that all legalese follows - it is now becoming too complicated to understand without paying a lawyer a very large sum of money. Given the target audience and the goals, this is not a good thing.

Re:Remember!

[Mr. Slippery]

GPL protects the freedom of the code, not the freedom of the developer.

Code doesn't have freedom. People do. The GPL protects the freedoms of people to use, share, and modify software.

If you want developer freedom, use the BSD license or some such.

The ability to create proprietary code based on free software - to say, "I got to use X freely to make my program Y, but if you try to use Y freely I will use government force to stop you!" - is not freedom. Freedom requires equality; any relationship in which one party will not allow others to do what they do, is not free.

Re:Remember!

[Bogtha]

I'm not sure if you are being sarcastic or not, but if you are, then please be aware that taking freedoms away to protect other freedoms is the basis of all law. You aren't free to hit me because you've had that freedom taken away from you.

Unless you are an anarchist, you really have no basis for criticising the GPL in this regard, because you agree with this logic applied to different areas.

Re:Remember!

[A beautiful mind]

Ultimately, the GPL has begun the process that all legalese follows - it is now becoming too complicated to understand without paying a lawyer a very large sum of money. Given the target audience and the goals, this is not a good thing.

Your beef is with the legal system, specifically legalese. It is the way to enter into official, legally binding agreements. Without the GPL using legalese, it would be worthless for it's goals. If you want to know what the GPL is about, be sure to read the documents on FSF's site that detail the intentions ("spirit") of the license.

If you need legal certainty that the GPL is what it appears to be and FSF's interpretation and execution of codifying their intentions into legalese is correct, then you of course need to consult a lawyer. It is the legal system's fault normal people need a legal interpreter in order to conduct official business.

Re:Remember!

[iamacat]

The government takes the freedoms of killing, torturing and in general harming people away (the cynical might say to monopolize them for itself, but that's another subject), but few could argue that we are less free because of that.

You better hope they allow fair use as an affirmative defense then, and do not outlaw tools that enable such fair use. Because most probably "the government" will not be there when you get harmed, tortured or killed - they will only investigate after the fact.

In the same way, ignoring the license should be allowed when the copyright owner breaks laws or terms of your contract.

Re:Remember!

[vux984]

Those restrictions are for your freedom. It is important to take freedom away to protect it. Truly allowing freedom would allow freedom to be taken away, and we can't allow that, so we've taken away some freedom to allow true freedom to flourish.

read through my first paragraph carefully and tell me what I've said differently than the GNU people.

Nothing. You got it right. But you apparently don't understand it yourself.

Your 'first' paragraph is EXACTLY the reason, and indeed the ONLY reason murder is illegal:

To paraphrase your 'first paragraph':

Those restrictions [we've outlawed murder] for your freedom [to live]. It is important to take away your freedom [to murder people] to protect your freedom [from being murdered]. Truly allowing freedom [to murder] would allow freedom to be taken away [ie everytime you excercise your freedom to murder someone else loses their freedom to live], and we can't allow that, so we've taken some freedom [to murder] to allow true freedom [to live] to flourish.

The FSF thinks the freedom of EVERYONE (including the people who got it from you) to modify and use GPL code trumps an individual developers freedom to prevent them.

Its not terribly complicated.

Re:Remember!

[Anonymous Coward]

GPL protects the freedom of the code, not the freedom of the developer. Big difference! If you want developer freedom, use the BSD license or some such. Different tools for different problems :)

Well, exactly! The GPLv3 is only really "free" if you completely agree with the FSF on the definition of "freedom".

It's odd that RMS, who is ordinarily a stickler for proper nomenclature, would insist on using the word "freedom" when he really means "user freedom". I can only imagine that he's perfectly aware of the fact that "freedom" is a loaded term that has a broader meaning than he intends. Frankly I find his appropriation of the term "freedom" obnoxious.

Re:Remember!

[heinousjay]

Sure, I get that. But there's an ever-increasing complication that makes things more difficult with each go-round. In general, simplification is a desirable goal.

Also, there's no point in shifting the blame. Sure, the legal system makes everything impossible. Playing into that game and supporting it (as the FSF does, by choice or by necessity) means that the blame falls squarely back on those shoulders.

Re:Remember!

[edwdig]

The ability to create proprietary code based on free software - to say, "I got to use X freely to make my program Y, but if you try to use Y freely I will use government force to stop you!" - is not freedom.

You can if you want to, but you don't have to. That's freedom.

Freedom requires equality; any relationship in which one party will not allow others to do what they do, is not free.

That's what the GPL is doing. The author of the GPL code can do whatever they want with it, such as make it proprietary, but the recipients of the code have restrictions on how they can use it.

Re:Remember!

[A beautiful mind]

Actually, the GPL was always good at using the opponent's tools against them. It is after all using copyright to achieve it's goal. Without copyright, there would be no need for GPL, as we'd have a pretty level playing field, as all code would be practically BSD licensed. In such a world the advantages GPL provides for free software authors wouldn't be necessary anymore.

Re:Remember!

[heinousjay]

Apparently I do understand it, and you don't get my point. I'm not arguing about the freedoms, I'm not arguing about the spirit or intent of the GPL. I'm saying it's murky, and I'm saying the FSF engages in spin attempts through word games. Plain and simple, sir.

It's all good, in the end. I expected the zealot masses to rise up and explain to me very carefully how I'm stupid, using bad analogies and seizing on the wrong points. I've dealt with it here for years.

Re:Remember!

[ShieldW0lf]

I have to say, if I was sitting down to build a business using a GPL3 stack vs using a GPL2 stack, the GPL3 stack would make me feel a lot more secure that some dickhead from the BSA wasn't going to come mug me in broad daylight once I started to see some success in my endeavors.

There may only be a few developers in the very large pool that jump right on to this license, but if those few developers put together a comprehensive tool set that is released under this license, it could attract a large number of users who worry about this very risk and find the safety offered to them very appealing.

Re:Remember!

[heinousjay]

I believe that to be a fallacy. Without copyright, people could share binaries with no repercussions, but getting a hold of the source would still be nearly impossible.

Re:Remember!

[Timothy Brownawell]

It doesn't restrict hardware. You can make your hardware DRMed like hell. You just can't run GPLv3ed software on that DRMed hardware. That's a restriction on the software (don't run it on DRMed hardware).
Vista doesn't downgrade content. You can watch all the hi-def you want, you just can't watch it on Vista. That's a restriction on the OS (don't watch hi-def content with it).

Re:Remember!

[Anonymous Coward]

Without copyright, there would be no need for GPL

How would you prevent the Tivo in a copyright-free world? The GPLv3 explicitly relies on copyright as a means of furthering a particular ideological agenda -- not as a way of using copyright against itself.

Re:Remember!

[pclminion]

GPL protects the freedom of the code, not the freedom of the developer. Big difference! If you want developer freedom, use the BSD license or some such. Different tools for different problems :)

Because abstract concepts deserve "freedom" but humans don't?

No, there's no such thing as "freedom" of an inanimate, incorporeal object. The only relevant factor is the freedom of the developer, which is indisputably reduced by the GPL.

Re:Remember!

[Chemisor]

> GPL protects the freedom of the code, not the freedom of the developer.

Considering that the people who choose the license are developers, and so are the people that might sign up to help the project, is it any wonder that these developers prefer to preserve their freedom?

Re:Remember!

[pclminion]

Unless you are an anarchist, you really have no basis for criticising the GPL in this regard, because you agree with this logic applied to different areas.

Argument by analogy is always stupid, and this is no exception. Just because you can substitute all the variables in an argument with other variables to produce a situation which is ridiculous, does not mean that the original argument with its original variables is ridiculous. You can argue against ANYTHING that way.

Re:Remember!

[vux984]

I fail to see how its murky.

Surely we can all understand the reasoning why your freedom to murder is curtailed.

The freedoms restricted by the GPL are done in precisely the same spirit - because if you exercise those freedoms the freedoms of others would be lessened.

It is precisely the same reasoning.

Re:Remember!

[Trinn]

the original source perhaps, but tools exist to extract something close-enough to it from any binary, it is only copyright that really stops this from happening on a dramatic scale. It does happen though from time to time, especially in places where people feel that nobody will be able to trace them back to the binary in question. Of course patents also block this sort of thing but since patents make the process available it seems silly to reverse engineer that process, you're screwed either way

Re:Remember!

[Anonymous Coward]

I'm not sure if you are being sarcastic or not, but if you are, then please be aware that taking freedoms away to protect other freedoms is the basis of all law.

No. Two quick theories of law. Positive law: you do as I say 'cause of this gun I'm pointing at your forehead. Negative law: you are free to do as you please so long as you respective the freedom of others to do the same and, PS, I still have that gun. One approach is the foundation for a free society and the other is the foundation for totalitarianism. In either case, there are a minimum of two theories of law. Some might simplify to 'might makes right' and claim only one theory. None of these approaches gel with what you say. ALTHOUGH I get what you mean, it is an inaccurate description and hardly as inclusive as you suggest ('basis of all law').

Re:Remember!

[zotz]

"I'm not saying don't use it. I'm not even saying it's bad. I'm saying it is murky, and the article makes perfect sense to me because of that murkiness."

Could be. Can you think of a better wording for a license that will protect the four freedoms in all situations?

all the best,

drew

Re:Remember!

[Anonymous Coward]

Now taking bets on how many posts to go till we get a car analogy.

Re:Remember!

[sahuaro]

It's odd that RMS, who is ordinarily a stickler for proper nomenclature, would insist on using the word "freedom" when he really means "user freedom". I can only imagine that he's perfectly aware of the fact that "freedom" is a loaded term that has a broader meaning than he intends

You need to listen to one of his speeches. Richard always goes on to define just what those freedoms are. He's certainly not using the word in a "loaded" manner.

Sahuaro

Re:Remember!

[grcumb]

> GPL protects the freedom of the code, not the freedom of the developer.

Considering that the people who choose the license are developers, and so are the people that might sign up to help the project, is it any wonder that these developers prefer to preserve their freedom?

How? By overwhelmingly choosing the GPL as their license of choice? You are aware that GPL (v2) is the single most popular software license right now, aren't you?

The GP is wrong, and you're wrong, too. I won't attempt to speculate about what motivates developers to choose a particular software license, but one thing is clear: the majority of the development community doesn't prefer the BSD license, no matter what benefits it may hold for them.

Re:Remember!

[sahuaro]

GPL protects the freedom of the code, not the freedom of the developer.

Let me reword this based on the answer RMS gave to a question at one of his recent speeches (in Sweden, I think).

GPL protects the freedom of the user, not the power of the developer.

Sahuaro

Re:Remember!

[ScrewMaster]

Regardless, if the GP hits me I'm damn well going to hit him back.

Re:Remember!

[Ginger Unicorn]

it's like each post is a buick, right up till the point someone mentions a car analogy, then it's all toyotas. god i'm insightful today.

Re:Remember!

[Ginger Unicorn]

it also doesn't restrict how you *run* it. Run it on DRMed hardware to your heart's content. It just specifies what information you must include should you wish to redistribute it.

Re:Remember!

[Anonymous Coward]

He's certainly not using the word in a "loaded" manner.

On the contrary. Since he has described his position as supporting "freedom", he has created the obnoxious situation where anyone who does not agree 100% with the FSF is opposing "freedom".

This is not a theoretical issue; one prominent opponent of "freedom" (so declared by RMS) is Linus Torvalds, whose crime is to not embrace GPLv3 with open arms (he has said that he doesn't mind GPLv3, but won't use it; he has also said that he loved GPLv2).

For someone who spends a lot of time thinking about how to avoid misunderstandings in language, I have to believe that his overbroad use of the term "freedom" is deliberate.

Re:Remember!

[fyngyrz]

The government takes the freedoms of killing, torturing and in general harming people away... few could argue that we are less free because of that.

We are less free because of that. My ability to defend my family, my property and my home is compromised by the government's monopoly on killing. This is a very bad thing. Worse, it is done to protect people who have no business in my home or on my property, at my family's expense. You can certainly argue that by placing such a restriction on people, accidents can be prevented, but the fact is, considerable freedom and safety are lost — as well as innocent lives.

Torture for information retrieval is stupid (including waterboarding and all the other borderline techniques the morons are using today) because a tortured individual will tell you anything in order to stop the behavior. However, torture as punishment for a crime (for instance, placing someone in a pound-you-in-the-ass-federal-pen) can serve as both deterrent and retribution. Retribution is not a zero-value experience for those who have been wronged. The government reserves this right as well. That - IMHO - is probably for the best. Though I think they use it inappropriately, for example on adult citizens who are simply exercising a choice as what to do with or to their own bodies and/or the bodies of other consenting, informed adults.

Re:Remember!

[Bogtha]

Look at the context. I'm not talking about justification of law, I'm talking about the nature of law. Can you name a single law that doesn't restrict freedom? Or do you accept that all law restricts freedom in some way?

Re:Remember!

[Bogtha]

If the logic and relevant facts are preserved, then it's an entirely valid approach. My point is that the idea that you can protect freedom by restricting freedom isn't as absurd as heinousjay made out. My evidence of that is law that it's reasonable to assume heinousjay agrees with.

Re:Remember!

[s20451]

the original source perhaps, but tools exist to extract something close-enough to it from any binary, it is only copyright that really stops this from happening on a dramatic scale

In the United States, reverse engineering is legal and not covered by copyright. And I'm not aware of any project that has quickly reverse-engineered any non-trivial piece of software, other than BIOS. For example, Windows NT has been around for almost 15 years, while the reverse engineered version (ReactOS) is nowhere near ready for production use after nine years of work. It doesn't even have network support yet.

Re:Remember!

[mr_mischief]

The freedom of secondary or tertiary developers to close the source is reduced. The freedom of the initial developer to specify how his hard work is to be treated is increased. The freedom of the developer who would have only received the closed-source version of the software otherwise is also increased. It's a trade-off, and which freedoms matter the most to an individual determine which license that person will choose. Some people use one for some projects and the other for some as well.

Re:Remember!

[phoenix.bam!]

GPL protects the freedom of the user, not the freedom of the distributor. Sometimes the developer is a user and sometimes the developer is the distributor.

Re:Remember!

[Anonymous Coward]

Fuck off dickhead, write your own fucking code if you don't like the GPL.

Re:Remember!

[Anonymous Coward]

What you describe would be considered by many to be an onerous restriction of GPL3 if it were true. The way I understand it the only anti-DRM provision in GPL3 is that you can implement as much DRM as you want with GPL3 software, but you disavow the legal protection that the DMCA offers your DRM scheme, thus allowing your users to try to circumvent (or try to circumvent) your DRM scheme without legal penalty.

Re:Remember!

[Anonymous Coward]

You aren't free to hit me because you've had that freedom taken away from you.

Freedom only makes sense when defined in terms of voluntary association vs. coercion, the two (opposite) modes of human interaction. Freedom means freedom from coercion, or put another way, the right to voluntary association. The concepts have meaning because they are opposite.

Realizing this, the idea of a "freedom" to employ coercion (against others) doesn't make much sense, because a "freedom" to violate freedom would completely ignore the fundamental concept of freedom.

Your friend isn't "free" to hit you because such an act would be a violation of your right to voluntary association -- your right to be free from coercion -- not because some third party (government) came in and declared that particular "freedom" void. The concept of freedom preceded government, not the other way around.

Re:Remember!

[synthespian]

Freedom is a category that relates to people.
Freedom can never relate to an inanimate object, such as code.

Re:Remember!

[tepples]

How would you prevent the Tivo in a copyright-free world? By introducing a competing DVR that has built-in file sharing.

Re:Remember!

[synthespian]

You guys keep repeating phrases the FSF taught you. You refuse to think with your own heads.

What about the "freedom" to be suckered into a dual-licensing scam, such as GPL/Proprietary projects?

Does the GPL level the playing field in that case, or is the field tilted in favor of the owner of a dual-license?

The way you put it, it is as if the GPL were tit-for-tat. The GPL is not tit-for-tat. People are tit-for-tat, as in: a person chooses to cooperate or not with a project. As in: choosing not to cooperate with a dual-licensed project were only a small group has the right to sell the code, where the other group must be of free software. This is the scenario of dual-licenses. The GPL promotes the very same practices it seeks to condemn. You can't fight fire with fire...When one group has secured right the other group doesn't have, you have a losing situation. This happens with proprietary licenses and the GPL/proprietary combo. Futhermore, there's nothing in the GPL, due to copyright laws, that garantees that a further more advanced version cannot be closed under this group-exclusion mechanism. Of course, the same can be said of the BSDL except that, by its very design the same rules apply to everyone - whichever way they go. So, you can exclude me, but I can exclude you. Code stays where it is or it evolves. People choose to cooperate or not. But people are not constrained into cooperating while, at the same time, leaving a loophole for a group exclusion. The GPL can't fix that. Some laws are just above the Church of the FSF. That is why I said that the GPL is not tit-for-tat. GPL doesn't do a thing. People do things. The GPL doesn't garantee ethical behaviour, as I think some Linux developers illustrated recently. Boxing the human mind is a no-go proposition ever since eons ago.

The FSF is really keen on metaphysically-dislocated concepts ("code" is free) and moralistic philosophy ("you must abide by my views of 'freedom' "). GPL v3 is but a further extension of this moralism. This time, they would like to take the power you have to develop exclusive solutions involving hardware.

The FSF is becoming an energy-sucking vortex. Stallman has even warned to expect further developments of the license. People want to go down this path, I'm sorry for them. But, in yet another example of the unfairness of the GPL, only big corporations have the wherewithal for the increasing legalese-junglespeak.

Re:Remember!

[einhverfr]

The GPLv2 I think has been a good license in general. I object to the GPL v3 for a number of reasons:

1) Sacrifice of essential freedoms in the license due to a conflict mentality (the antitivoization provisions, for example, which are after all entirely ineffective since you can use a hypervisor to essentially reduce the functionality of the device with non-certified code without actually interfering with the execution of the binary as required by that license).

Note that in this case, you could tie the upgrade of the certified program to other components. If you modify the code, you can install it, and it can run. It just won't have anything to talk to since those software components will be missing and so won't do anything interesting. Since those software components are only communicating with the GPL3 app through essentially sockets and pipes....

It is *far* better to rely on the market in this regard. Sadly, RMS wants to coerce eveyone to agree with him (essentially firing the Hurd architect over his public objection to the use of the GFDL to force advocacy of the GNU project). RMS supports "Freedom" in a "War on Terror" sort of way. Maybe there is a place for him in the Bush Administration?

2) Incredible complexity which renders the license incompatible with BSD-licensed *dependencies.*

Re:Remember!

[Eivind Eklund]

The GPL protects the poor, downtrodden code from being exploited! Let the code rest, I tell you! Rest!

Eivind.

Re:Remember!

[Anonymous Coward]

>I can't imagine why anyone wouldn't understand that perfectly.

They probably work the same way as you.

Re:Remember!

[Anonymous Coward]

There are countless laws that place no restrictions on freedoms. They do place restrictions on actions (e.g. laws against murder), but there is a question as to whether you were free to take such actions in the first place. Regardless of the moral or ethical position of such laws, many laws have their basis in this idea, and so by definition do not restrict freedoms per se. Yes, using your definition of freedom, there are no laws that I know of which do not restrict 'freedom'. However, such a definition of freedom is so comically broad it could be used it to support the conclusion that the laws of physics restrict your freedom.

Re:Remember!

[PietjeJantje]

I'm not sure if you are being sarcastic or not, but this is the first time a GPL zealot managed to call BSD altruism a flying bullet aimed at its users, in order to advertise their own license. Why, thanks! In a similar line of reason, Mother Theresa was an anarchistic warlord. Not only will their FUD never end, it's getting worse and uglier all the time. But at least BSD people don't have to advertise their cause like that. The GPL people are doing that nicely now, they only don't realize it.

Re:Remember!

[TuringTest]

voluntary association vs. coercion, the two (opposite) modes of human interaction But, but... coercion is voluntary action, that you take in order to avoid some voluntary actions from the other part! So they are not really opposite, and your definition of freedom doesn't make sense.

Re:Remember!

[Anonymous Coward]

coercion is voluntary action

It's important to note that the concept of freedom only makes sense in terms of interaction between human beings. Of course, on the part of the aggressor, coercion is always voluntary. Everything a person chooses for himself is, of course, voluntary. But again, we are not talking about the "interaction" between an individual and himself. We are talking about the interaction between one individual and another individual, and that interaction must be one of either coercion or voluntary association.

The "process" of receiving coercion (on the part of the victim) is, of course, involuntary. Logically, a person cannot volunteer to be subject to coercion, any more than a person can coerce another to volunteer.

So they are not really opposite, and your definition of freedom doesn't make sense.

They most certainly are opposite and mutually exclusive concepts -- in terms of interaction between two or more human beings. Again, we are not talking about the "interaction" between a human being and himself, like the little fairy sitting on your shoulder telling you what to do. A person cannot grant freedom for himself, and neither can he oppress himself -- those concepts only make sense in terms of interaction with others.

Really, this isn't rocket science. This is common sense.

Re:Remember!

[Anonymous Coward]

"Real freedom" would allow you to kidnap and murder people at your own discretion at any time without repercussions, and everyone else to do the same. That sounds like something worth persuing, right?

Re:Remember!

[Anonymous Coward]

Whohoho, boy are we sounding a bit like George W here? Taking rights away to preserve freedom?

Re:Remember!

[mr_mischief]

I'm not repeating anything. The phrases I used were my own. Try to find them anywhere on any FSF site. I do not appreciate your accusation that I do not think.

The owner of a dual-licensed project is either the original author, without whom you'd never have a GPL or BSD version anyway, or someone who has had copyrights assigned to them for some reason that is sufficient for the original author(s). The original author never loses his or her rights as long as they keep the copyrights.

You can always fork anything released under a dual license if you don't want to assign your copyrights away to someone who won't include work owned by others.

Dual-licensed code is still GPLed code. With BSD you get the BSD, and if the original distributor shuts down, you might have BSD + 8 different other licenses or you might have just the 8 other mostly non-free licenses.

Some dual-licensed projects, such as Perl, offer the dual licenses specifically so you, the recipient, can choose to redistribute as open or closed source. The licenses in question are the GPL and the Artistic License. Lots of other projects are going this route, although not too many other major ones yet. It's pretty common especially among projects developed in Perl.

The FSF has nothing to do with my decisions. I don't live by RMS's philosophy. I think there's nothing wrong with GPL, BSD, Creative Commons, MPL, nor a bunch of other licenses. I don't have a problem with proprietary licenses on a philosophical level. I just think they suck compared to freer licenses, and it takes some other attributes of the code to overcome the license shortcomings.

As for having to abide by anyone's particular ideas of freedom, well, that's what you do any time you use a license. You're reinforcing the BSD camp's ideas when you distribute under a BSD license. You strengthen the FSF's positions when you distribute under the GPL. The Mozilla people love to see their license. In order to distribute under the GPL, you must offer the four freedoms. It's not because RMS wills it, but because the license says those four freedoms are available to the recipients of the code.

Yes, the BSD focuses on a different type of freedom. I said that. What part of "It's a trade-off, and which freedoms matter the most to an individual determine which license that person will choose. Some people use one for some projects and the other for some as well." did you fail to understand?

Unbox your own fucking mind. Mine is quite free to think for my benefit, thanks. You seem to be guilty of following Theo like you're accusing others of following RMS. For the most part, that's not how the world works. I'm not a disciple of anyone's licensing camp. I like New BSD, GPLv2, MPL, ASL, CC-SA, CC-Attribution, Artistic License, the MIT license, NetHack GPL, the PHP License, and more. GPLv3 is a bit harsher than I'd like, but not enough so that I'll entirely stop working with software licensed under it.

RMS thinks all software should be licensed exactly as GPLv3 does it. I'm saying both the GPL and the BSD camps have their points of view, and that which individual parts of a license mean the most to you determine which yuo should release your own code under. If you don't have code of your own to license, then you really needn't have an opinion, as it's the author's choice and not yours. Still, you're free to have an opinion even when it doesn't make a damn bit of difference. How does that make me the FSF drone you're saying I am?

Re:Remember!

[cromar]

Why get so worked up about anthropomorphic metaphors used glibly in casual correspondence?

Re:Remember!

[Trinn]

Reverse engineering yes, decompiling then using that code no. There is a difference, but its subtle. Basically if you can deduce something from the communication that happens with it then you can reimplement it "safely", but if you have to disassemble/decompile it, you can't use that code. The one way that that does get used is through a system where one person decompiles/disassembles and writes some documentation, then that documentation is passed on through a number of other hands until someone uses it to independantly write a new implementation. I'm simplifying, and I'm sure I got some details wrong, but suffice it to say that simple decompiling can still be covered by copyright (specifically you can't distribute the decompiled stuff any more than you could distribute its compiled form. you are technically allowed to decompile it (licenses aside, of course, because they could concievably try to disallow that, and usually do))

Re:Remember!

[xappax]

Truly allowing freedom would allow freedom to be taken away, and we can't allow that, so we've taken away some freedom to allow true freedom to flourish.

Ok look: Absolute freedom is a social impossibility, because it can only apply to one person. It's impossible to give everybody absolute freedom (no restrictions whatsoever on their behavior) because this would give us the freedom to enslave each other. You can characterize that as some Orwellian doublespeak, but it's a fundamental reality. The purpose of a "free" society is not allowing everyone to do absolutely whatever they want, it's allowing them to do whatever they want as much as possible without seriously interfering with other people's freedoms.

Freedom isn't absolute - people can't have freedom X unless they're also "unfree" to restrict freedom X. This realization can be frustrating and seemingly hypocritical (and is definitely abused at times), but it's important to understand before shitting all over the work of people who are genuinely working to increase the practical freedom we enjoy today, simply because it doesn't mesh with some philosophically pure concept of freedom.

Re:Remember!

[TuringTest]

We are talking about the interaction between one individual and another individual, and that interaction must be one of either coercion or voluntary association. You are begging the question. In order to prove that freedom depends on coercion versus voluntary actions, you are assuming that all interactions are either coercion or voluntary association.

The "process" of receiving coercion (on the part of the victim) is, of course, involuntary. The process of deciding the terms which the other party will offer in a voluntary association is also involuntary. So how you tell one apart from the other?

Really, this isn't rocket science. This is common sense. Only if you agree with the philosophic principles of libertarianism. For any other ideology, this false dichotomy is not obvious at all; you see it as common sense because you believe in it. Can you make the mental exercise of viewing the logic of your argument from outside your moral frame of reference?

Re:Remember!

[TuringTest]

coercion is voluntary action Also, i'm seeing that my wording was not clear and did not mean what I intended.

My original post should have been: "giving in to coercion" is a voluntary action, that you take in order to avoid some voluntary (but unpleasant for you) actions from the other part.

Re:Remember!

[TheRaven64]

You are aware that GPL (v2) is the single most popular software license right now, aren't you? Actually, I wasn't. Where did you acquire this statistic? I use a lot of Free Software, but it comes under a variety of licenses. The only GPL'd projects I use regularly are gcc and bash. The rest is BSDL (e.g. FreeBSD, OpenBSD), MPL (FireFox, Thunderbird), or LGPL (a huge number of things, including GNUstep, OpenOffice.org, etc). By number of projects, and by lines of code, there is a lot more LGPL than GPL'd code on any of the machines I own.

Re:Remember!

[Anonymous Coward]

Am I free to call you a fucking loser at least?

Whoops.

Re:Remember!

[Anonymous Coward]

If you don't know what anarchism is, please at least google for it before making a fool of yourself.

You may want to look at this: http://dwardmac.pitzer.edu/Anarchist_Archives/kropotkin/britanniaanarchy.html

And maybe even this: http://www.geocities.com/CapitolHill/1931/

Re:Remember!

[grcumb]

You are aware that GPL (v2) is the single most popular software license right now, aren't you? Actually, I wasn't. Where did you acquire this statistic?

Sorry, I honestly thought this was common knowledge, but some informal checking with colleagues demonstrated that it isn't quite so widespread as I thought. Here's a fairly well-researched source:

http://www.dwheeler.com/essays/gpl-compatible.html

Rhetoric notwithstanding, this guy does check his facts.

Maybe I want my code to be used Commercially!

[Anonymous Coward]

I don't care if my code is used by Megacorp, or Minicorp cause I may not want to write code to do something myself if you already did. If you were not smart enough or cared enough to make a profit your problem. This crap about oooh it can't be used cause you make money B.S. is crap. Copy, Paste, Done with Quota.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

The GPL isn't about preventing commercial use. If you bothered to read the license(s) you would know that.

It's about preserving users' freedoms. If a commercial entity uses GPL code and distributes that to end users (even paying ones), they're obligated to give them access to the source code. It's that simple. GPLv3 just adds some extra clauses to prevent companies from weaseling around the spirit of these simple terms by using any software patents or the like.

If you don't care about commercial entities taking your code, making changes, distributing it to users, and then refusing to give those users (which may include you) access to their modified code, then release your code under the BSD license, or into the public domain. It's your choice. Stop complaining about other peoples' choices.

Re:Maybe I want my code to be used Commercially!

[Em Adespoton]

GPLv1 and v2 look at free software being sidetracked or misused by private interests who want to subvert it to their own goals. What GPLv3 has done is tackled the opposite problem... private interests attempting to subvert GPL'd code by contributing code to a GPL project which has other non-copyright restrictions attached to it (eg, patents). The idea here is that someone can contribute a key piece of code and have a submarine patent on the algorithm. The entire project becomes popular, things are modified over time, and then bam! Patent holder comes out asking everyone who uses that entire codebase for patent fees based on the algorithm they added. Often that bit of code can be replaced, but that's no so easy to do when you're dealing with embedded software already packaged or out in the real world.

Re:Maybe I want my code to be used Commercially!

[huckamania]

It's more likely that a patent infringement would be added unwittingly by a third (or fourth) party. The GPLv3 does what it can, but it can't magically give immunity from patent infringement. Well, except for the case you outlined which is probably the least likely to occur.

Still, SAMBA!

Re:Maybe I want my code to be used Commercially!

[pclminion]

It's about preserving users' freedoms.

It is not. The open-source nature of the codebase is only helpful if the user has (or is) a developer to do the required work for him. A non-programmer who has source code has no advantage at all over a closed-source product.

It is only developers who can extract the so-called "value" from open source.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

The GPLv3 does what it can, but it can't magically give immunity from patent infringement. Well, except for the case you outlined which is probably the least likely to occur.

I wouldn't be so sure of that. A patent-holder adding patent-protected code to a GPL project just to destroy it sounds exactly like something Microsoft would do. They've done worse things in their history.

Wouldn't it be nice if every business could be run ethically? Unfortunately, not only is that impossible, but our very largest and apparently most respected corporations are completely devoid of ethics. That's why measures like this are necessary.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

What an idiotic argument.

If you're small-to-medium business GrishnakhCo, and you buy software product XYZ from JunkSoft, and then proceed to use it for many business-critical tasks, like storing all your designs in it, then the company goes out of business, you're screwed. However, if you access to that source code, you can hire programmers to fix problems in the software or modify it to fit your needs.

You don't have to be a programmer, though it obviously helps. All you have to do is hire a programmer. For any business, this isn't a huge cost for a small project, done under contract. Lots of programmers make their entire living working on contract projects like this.

Open source isn't just about home computer users.

Re:Maybe I want my code to be used Commercially!

[Anonymous Coward]

I don't care if my code is used by Megacorp, or Minicorp cause I may not want to write code to do something myself if you already did. If you were not smart enough or cared enough to make a profit your problem. This crap about oooh it can't be used cause you make money B.S. is crap. I'd agree 100% with your proposition that "oooh it can't be used cause you make money B.S. is crap.".

Fortunately, the GPL does not say that. What the GPL says is basically this: "if you take this code (which you did not write) and then use it in a product (re-distribute it), then you must make this source code and any changes you made to it available to anyone who asks for it".

The GPL does not say anything at all about "oooh it can't be used cause you make money".

Re:Maybe I want my code to be used Commercially!

[pclminion]

If you're small-to-medium business GrishnakhCo, and you buy software product XYZ from JunkSoft, and then proceed to use it for many business-critical tasks, like storing all your designs in it, then the company goes out of business, you're screwed.

Why would you be screwed? Does the software disintegrate when the maker goes away?

However, if you access to that source code, you can hire programmers to fix problems in the software or modify it to fit your needs.

If you had serious problems with the software why were you using it? You just said it yourself -- a company and the support they provide could disappear at any moment. You could choose a product that DOESN'T suck.

You don't have to be a programmer, though it obviously helps. All you have to do is hire a programmer.

Which you could have done initially, and gotten a product far better suited to the original need.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

Don't be such an idiot.

Why would you be screwed? Does the software disintegrate when the maker goes away?

Every software maker advertises how important support is, after the sale. If you have problems with a high-dollar product afterwards, you have to have someone to talk to about it and fix the issue. We're not necessarily talking about cheap-ass $50 boxed software; there's lots of software out there that costs 4-5 figures per seat.

If you had serious problems with the software why were you using it? You just said it yourself -- a company and the support they provide could disappear at any moment. You could choose a product that DOESN'T suck.

Another utterly stupid argument. When you're dealing with specialty software, your choices are generally few, and limited to smaller companies. Obviously, you have no idea of what it's like working in a real company.

Which you could have done initially, and gotten a product far better suited to the original need.

The idiocy never ends! Creating a large special-purpose software product can take years. If your company doesn't specialize in that, it makes more sense to spend $30k on pre-made software to do it than to hire a team of programmers for 3 years.

Obviously, you have little concept what life is like outside your parents' basement.

Re:Maybe I want my code to be used Commercially!

[l4m3z0r]

you don't care about commercial entities taking your code, making changes, distributing it to users, and then refusing to give those users (which may include you) access to their modified code, then release your code under the BSD license, or into the public domain. It's your choice. Stop complaining about other peoples' choices.

Translation: If you don't agree with my point of view. STFU.

Re:Maybe I want my code to be used Commercially!

[huckamania]

Not trying to pick a fight, but the open/free community is certainly not without its ethically challenged members.

I think it is entirely possible that we will see GPLv3 code maliciously added to non-GPL code with the express purpose of tainting it. I seriously doubt that MS employees have the same zealotry that the open/free community has. After all, most MS employees are probably only there for the pay check. You definately can't say the same thing about the open/free crowd.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

No, the translation is: why are you bitching so much about the choices other people make? It doesn't affect you, and it's none of your business.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

I entirely disagree, and point to history for support.

I can't recall any case where GPL code was ever added to a project to intentionally taint it. Can you? In fact, I can't recall any such ethically-challenged behavior among the open/free community. Yes, they've had their squabbles. There was the incident where one BSD developer used some GPL code, but that looks to have been a misunderstanding, and anyway, the community worked it out. There's been some incidents where developers were alleged to take BSD code and declare it GPL; I'm not sure what happened with those. But these are isolated incidents involving a few people, and they seem to get worked out when some light is shone on them. None of them seem to be terribly malicious, just people trying to reuse open/free code and not exactly following the rules until their mistake is discovered.

MS employees are mostly there for the paycheck, but the company itself is not: its upper management will do anything for more power and money, legal or illegal, and it has shown this behavior over and over again. MS employees are just cogs in the machine, and do whatever upper management tells them to. They don't even have to worry about lawsuits, as their company protects them and acts as one. So I don't see why MS wouldn't intentionally taint open/free code with patent-encumbered code, under direction from management. It's only to their benefit.

The open/free crowd definitely is zealous, but their zealotry is confined to doing the right thing, and arguing what's better ("GPL protects our freedom!" "No, BSD is more free!"), not doing plainly unethical things to advance their cause (which is a cause that deals directly with ethics). The proprietary crowd doesn't care one whit about ethics; corporations only exist for money, and ethics (and laws too) are just a stumbling block in their path.

Re:Maybe I want my code to be used Commercially!

[l4m3z0r]

why are you bitching so much about why people are bitching about the choices other people make?

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

Because they're bitching so much about the choices other people make and I'm tired of reading all their bitching about it!

Re:Maybe I want my code to be used Commercially!

[pclminion]

Every software maker advertises how important support is, after the sale. If you have problems with a high-dollar product afterwards, you have to have someone to talk to about it and fix the issue. We're not necessarily talking about cheap-ass $50 boxed software; there's lots of software out there that costs 4-5 figures per seat.

For 4-5 figures per seat you'd think you should be able to operate the software without constant assistance. If something is regularly not working for you, here's an idea: HIRE A PROGRAMMER.

Another utterly stupid argument. When you're dealing with specialty software, your choices are generally few, and limited to smaller companies. Obviously, you have no idea of what it's like working in a real company.

I do work at a real company. And yes, there are often few choices, and sometimes all the choices are bad. I do not see how Open Source is the answer to this problem as you claim it is.

The idiocy never ends! Creating a large special-purpose software product can take years. If your company doesn't specialize in that, it makes more sense to spend $30k on pre-made software to do it than to hire a team of programmers for 3 years.

I'm confused. Are we talking about Open Source here or not? First you say Open Source is a benefit because you can hire programmers to work on the code, and THEN you say it's better to spend money up-front so that you... don't have to hire programmers? What the fuck exactly IS your opinion?

I happen to disagree with your stances. You can hate me for that, it's okay. But it doesn't mean I'm a kid.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

I'm confused. Are we talking about Open Source here or not? First you say Open Source is a benefit because you can hire programmers to work on the code, and THEN you say it's better to spend money up-front so that you... don't have to hire programmers? What the fuck exactly IS your opinion?

You do realize that open-source doesn't necessarily mean free-of-cost, don't you?

Back in the old days, when you bought some expensive piece of equipment, they gave you the schematics with it. This is basically the same thing.

It's basically a safety net in case the proprietor goes out of business or is unsatisfactory in their service efforts.

Re:Maybe I want my code to be used Commercially!

[huckamania]

"So I don't see why MS wouldn't intentionally taint open/free code with patent-encumbered code, under direction from management. It's only to their benefit."

Because it's open? Because everyone can look at it? Because it will probably have to be built using GNU tools and libraries?

I could go on. Sprint and Verizon didn't have to taint Vonage for them to have violated a patent.

"but their zealotry is confined to doing the right thing"

I'm sure there are many here on slashdot that would consider tainting Office or Exchange a right thing to do. MS is an Illegal Monopoly and opening up either of those two things would certainly be a boon for open/free land. That's why I think it is likely to happen.

Doing the right thing isn't hard. It's knowing what is right. Once you know something is right, doing it is easy.*

---
*Stolen from Ghandi -er- Ben Kingsley.

Re:Maybe I want my code to be used Commercially!

[pclminion]

Back in the old days, when you bought some expensive piece of equipment, they gave you the schematics with it. This is basically the same thing. It's basically a safety net in case the proprietor goes out of business or is unsatisfactory in their service efforts.

I've never claimed that open source is free of cost, nor have I even said that it's a bad model of software development. I just hate the GPL, specifically.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

Because it's open? Because everyone can look at it? Because it will probably have to be built using GNU tools and libraries?

It's been proven many times that being open doesn't mean a huge number of people are going to look at and inspect some code. Usually, it's only a handful of the project's developers that ever look at that code, even though the source is open for all.

Worse, patented code isn't very easy to spot. How many software patents do you know of off the top of your head? How are you sure an algorithm you're implementing isn't patented? Have you done a patent search? They're not called "submarine patents" for nothing, you know. If some random developer submitted some great new code, but didn't bother to tell anyone there was a patent covering it, how do you think people are going to know?

And what do GNU tools have to do with anything? Compiling something with gcc doesn't change its legal status. gcc is just a tool.

I'm just saying this is a possible threat from certain proprietary interests that very much want to destroy and/or discredit open source software.

I'm sure there are many here on slashdot that would consider tainting Office or Exchange a right thing to do. MS is an Illegal Monopoly and opening up either of those two things would certainly be a boon for open/free land. That's why I think it is likely to happen.

I don't agree. MS Office and Exchange are made by MS behind closed doors, in a very proprietary fashion. The threat of GPL code "virally" "infecting" proprietary code has always been there, as long as GPL code has existed. So MS and other companies have strict policies to protect themselves which forbid developers from using GPL code, in case some programmer wants to take a shortcut.

But even if some GPL code got into Office because of a lazy MS programmer, and this somehow got past whatever checks they may have to prevent GPL infection, the chance for discovery is still pretty low because their code is not open. People would have to somehow prove it's there just using the binaries. There's been some very recent progress on this front, but that's recent (and I don't know the details of that). So I don't think GPL infection is a huge problem for MS or other proprietary companies, unless someone squeals.

However, I don't see how a GPL zealot could ever infect MS code with GPL code; again, the codebase is closed, and only MS employees can add to it. It's not like they take contributions from random people on the internet, like most open-source projects do. I'm sure many GPL zealots would love to do this (heck, I'd like to see it done!), but I don't see how it can be accomplished in practice. This issue seems like a red herring to me.

Re:Maybe I want my code to be used Commercially!

[Grishnakh]

That's fine; don't use it in your projects. But why do you have a beef with people who do? It's a lot better than closed-source code, and I don't see people going around bitching about XYZ software company selling closed-source software (I don't mean MS, I mean smaller non-monopoly companies), while I constantly see people ragging on the GPL.

It reminds me of people who complain about homosexuality. Don't like it? Fine, don't do it. But why do you have to make so much noise about it?

Re:Maybe I want my code to be used Commercially!

[pclminion]

That's fine; don't use it in your projects. But why do you have a beef with people who do? It's a lot better than closed-source code, and I don't see people going around bitching about XYZ software company selling closed-source software (I don't mean MS, I mean smaller non-monopoly companies), while I constantly see people ragging on the GPL.

I'm not talking about a choice between closed-source vs. GPL, I'm talking about, for instance, BSD vs. GPL. If it's a choice between "open" and "not open" I think the choice is fairly obvious.

And I don't see the mystery why I would be commenting about the GPL in an article about the GPL and how people love/hate it.

Re:Maybe I want my code to be used Commercially!

[Anonymous Coward]

RMS doesn't want people to pay for software.

RMS doesn't want people to pay for content.

Therefore RMS doesn't want developers of either to eat!

So 6% adoption in a few months.

[Anonymous Coward]

It'll be a GPL3 world by 2009? (just pointing out the meaninglessness of these numbers)

Re:So 6% adoption in a few months.

[Anonymous Coward]

5.98125% of that 6% is the GNU project. Only 7 actual projects that RMS doesn't have control over the licensing of use the GPL3.

Re:So 6% adoption in a few months.

[TheWanderingHermit]

By 2009? I think a lot of people have strong feelings about not using it.

It's now the Vista of Open Source Licensing.

Oh dear!

[mmcuh]

Oh dear! Another rift in the community, etc. Really, how many articles of this type have been posted to Slashdot in the last few weeks?

And the statement "Just 6 percent of developers working with open-source software have adopted the new GNU General Public License version 3" is obviously false, since the vast majority of GPL-licensed software have copyright notices that say that the software is available "under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version" - which includes GPL version 3.

What is this "Evans Data Corporation"? It would be interesting to see any other press releases they have written.

Re:Oh dear!

[BibelBiber]

A very good reason not to put this kind of "optional" in your license again (whatever you think of v3).

Re:Oh dear!

[cromar]

I am wondering who they are, as well. If it helps, they also did this study about PHP.

Ooh, and here's one where they say Linux is gaining ground over Windows in certain areas...

Re:Oh dear!

[RightSaidFred99]

It's not obviously false. The GPLv3 is a more restrictive license, the copyright holder, if he wishes to enforce this higher level of requirements for distribution, must distribute his code under it. It's rather silly to say that since the person wishing to make use of the license can choose a more restrictive (Gplv3) license - what's the point?

The whole point of Gplv3 is that the FSF and several assorted righteous nutjobs thought they would "get one over" on Microsoft. They didn't, but Groklaw and other assorted retards really think they were quite smart about it. Regardless, for a company to be bound by Gplv3 the software they distribute must be explicitly licensed under Gplv3. The "or later" clause only makes sense if the "or later" version is _less_ restrictive, in which case a distributor can point to that clause and the newer, less restrictive version of the GPL as a defense if accused of copyright infringement.

Complicated, I know, and I'm sure you just don't get it.

Re:Oh dear!

[mmcuh]

The "or later" clause only makes sense if the "or later" version is _less_ restrictive, in which case a distributor can point to that clause and the newer, less restrictive version of the GPL as a defense if accused of copyright infringement. No, the "or later" is important regardless of whether later versions are more or less restrictive. If a piece of code is distributed under "GPL version 2 or, at your option, any later version", you can use parts of it in your own program that you distribute under GPL version 3 or later. You could not do this if the code you wanted to use was licensed under GPL version 2 only.

Re:Oh dear!

[A beautiful mind]

Oh dear! Another rift in the community, etc.

Someone should do a study about the correlation between articles predicting a rift in the community and slashdot posters sneaking in goatse'd links. It might make more sense and possibly be more entertaining than the study in TFA.

Re:Oh dear!

[Anonymous Coward]

What is this "Evans Data Corporation"? It would be interesting to see any other press releases they have written.

"Our Data in the News
The following is a sample of recent articles featuring Evans Data."

Click the links at top of page for more info on your first question.

Re:Oh dear!

[moderatorrater]

So what, you're saying that basing your decisions on the ongoing wisdom and good intentions of a group of people you're not part of isn't smart? You might as well tell me to repeal the Patriot Act!

Re:Oh dear!

[mr_matticus]

But this is not the same line of argument as the "GPL2 or later *is* GPL3" approach--it's not. The copyright holder has not moved to GPL3, so it's still GPL2. Unless the copyright holder changes the license, the more "liberal" GPL2 license is still the controlling one.

It's not the order of the licenses that determines importance. Given multiple discrete licensing schemes available at the user's option, only the least restrictive license controls. Think of it this way: if you were presented with a choice of being able to keep 90% of profit or 50% at your option, the licensor cannot complain until you try to take 91%.

In short, anything that differs from GPL v2 can't be enforced, and any terms which are contrary or more restrictive than others concurrently available are simply invalid. That software stays GPL2. You can elect to make your derivative work GPL v3, but that doesn't change the fact that the owners surveyed here have elected not to transition to GPL v3.

Re:Oh dear!

[einhverfr]

Almost. IANAL, etc.

Licenses are disjoint. Hence the GPL v2 or later means that people can use or distribute the code under either license. However, since the GPL v3 is far more strict about dependency licensing, I am willing to bet that a lot of projects cannot effectively move even if they are GPL v2 or later.

In this case, however, the GPLv3 only grants a subset of the rights granted in the GPLv2. No new rights are granted. So it doesn't seem to be a big change.

Re:Oh dear!

[yt.rabb at gmail]

What is this "Evans Data Corporation" Seems like they do a lot of surveys on developers, java, php ... The interesting thing I found was that the CEO John F. Andrews was just appointed to some DRM advisory council. http://biz.yahoo.com/prnews/070917/aqm002.html?.v=22 Flame on.

Re:Oh dear!

[Azuma Hazuki]

Even if the 6% figure were true, it's still three times more than a previous article's citation for the corporate adoption rate of Windows Vista =P So GPLv3 can't be that much of a turd.

Re:Oh dear!

[quantaman]

And the statement "Just 6 percent of developers working with open-source software have adopted the new GNU General Public License version 3" is obviously false, since the vast majority of GPL-licensed software have copyright notices that say that the software is available "under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version" - which includes GPL version 3. Well speaking of "obviously false" please explain how you equate the option to adopt a license with actually adopting it. Hey while we're at it why don't we add the entire list of GPLv3 compatible licenses to your list as well, it makes sense that tomcat and apache be counted as GPLv3 projects!

For a bit on anecdotal evidence I should note that in my little 3 person project we considered an upgrade to GPLv3 from v2 (we have the "or later" clause), however decided against it since one of the devs was strongly against the idea. The fact is that GPLv3 is a different kind of license from GPLv2, v2 basically just says you need to allow distribution and modification and stuff, but the GPLv3 also covers usage with its anti-DRM clause. Not to mention the fact that v3 is way longer than v2. Until the FSF comes out with a GPLv4 that somehow manages to appease both camps I really suspect that you're going to see both licenses in wide usage, however I'm not sure that unity is possible. I think the fundamental debate between between GPLv2 and GPLv3 is similar to the old debate between GPLv2 and BSD licenses, it depends on what restrictions you're willing to enforce to ensure what you consider freedom.

Re:Oh dear!

[mr_matticus]

Hence the GPL v2 or later means that people can use or distribute the code under either license. However, since the GPL v3 is far more strict about dependency licensing, I am willing to bet that a lot of projects cannot effectively move even if they are GPL v2 or later. Precisely. And since the copryight holder has no way of knowing which license was accepted by the licensee, there is no legal way for them to enforce GPL3 terms at all, since they are contrary to and more limited than the broadest license available.

You can't enforce anything beyond the broadest license simultaneously available. All the other side has to do is say that they licensed it under GPLv2 and be done with it. What they choose to license their derivative work under has no bearing on with which license they received the original code.

I'm confused...

[packetmon]

'Developers are confused and divided about [the restrictions GPLv3 imposes], with fairly equal numbers agreeing with the restrictions, disagreeing with them, or thinking they will be unenforceable The conundrum... If a packet falls in an OSPF forest of Spanning Tress does it send an ICMP-Unreachable

Re:I'm confused...

[Bigjeff5]

The conundrum... If a packet falls in an OSPF forest of Spanning Tress does it send an ICMP-Unreachable I think it actually finds a nice Spanning Tree Loop and lives out the rest of it's days in un-inhibitted tree-swinging bliss.

No need to phone home when you can keep on going forever!

FUD, and not well founded....

[nweaver]

The GPLv3 is full of FUD, a lot of it well founded. There is a reason I won't use it, but then even GPLv2 isn't as free as I like (BSD forever, yada yada yada)

This article is also FUD, and unlike other GPLv3 fun, it is NOT well founded.

That 6% are actually using GPLv3 already is not an indication of failure, but a pretty big success, and suprisingly high. That 40% will never touch GPLv3 is not suprising, because how many of them are BSD zealots rather than GPL zealots anyway.

Re:FUD, and not well founded....

[Anonymous Coward]

Aren't these the same as the projected adoption rates for Vista?

Re:FUD, and not well founded....

[jrumney]

That 6% are actually using GPLv3 already is not an indication of failure, but a pretty big success, and suprisingly high. That 40% will never touch GPLv3 is not suprising, because how many of them are BSD zealots rather than GPL zealots anyway.

They mostly seem to be Apache zealots, judging by who won the OSS popularity contest that was part of the same survey.

Maybe for Now...

[cromar]

Anyone who pays attention to the discussions on ./ would know there is division of attitudes toward GPLv3. I'm fairly confident it will gain more support, what with the FSF behind it, and as new projects come out.

By the way, anyone have more information about just who exactly the Evans Data Corporation is, and whether they are a respectable source of research? (I noticed this is a press release and not an independent article...)

Re:Maybe for Now...

[rucs_hack]

anyone who pays attention to the discussions on slashdot will be entertained, but the likelihood of finding out completely accurate and useful information in a comment thread is rather low. I do sometimes find very interesting things, but I've never considered slashdot as a port of call in a software decision process, news, yes, flamewars yes, even intelligent discussions, but that's all.

I changed my project to gpl3 on the day it came out, my software was gpl2.0 or later anyhow, and I wanted to make the change formal, including the new license as text. I'm sure the few people who actually like my software won't feel inhibited by the license.
Its more important for very successful projects with a real risk of being indiscriminately ass raped by proprietary companies. For almost all gpl projects the gpl3 won't make any difference whatsoever.

That won't stop the arguments, but speaking as someone who actually studied the license in detail, I think its a lot better than gpl 2.

I rather suspect that a lot of people who disagree with gpl3 are doing so because they spend their evenings photoshopping Linus Torvalds head to Tam River screenshots, taping it to their monitors and fapping away while the serenity soundtrack plays in the background.

There you go, a reasoned response, didn't I do well :-)

Re:Maybe for Now...

[just_asgard]

i think that developers shouldn't use a license that hasn't proven it's reliability on practice. and it's obviously why most of developers were made such sollution. they're just waiting for some real proves.

No Margin of Error

[DragonWriter]

The survey was based on in-depth interviews with 380 open source developers and no estimated margin of error was given.

No doubt because it wasn't a random sample in the first place, so a "margin of error", which reflects the sampling error, would be meaningless.

Re:No Margin of Error

[mathimus1863]

If you replace all instances of "GPLv3" in the article summary with "Microsoft Vista" this would be a repeat of an article a a couple months ago.

Re:No Margin of Error

[Anonymous Coward]

It was a random sample of Open Source developers. If they had sampled Free Software developers, I am sure GPLv3 would have fared better.

Previous post..

[JCWDenton]

A previous post by a fellow /.-er also on the GPLv3 he outlined how right and wrong and not clearly definable. Im not sure if it was also a humerus attempt but I would really liked to find it again.
It was more or less a list of what to expect in upcoming versions of the GPL.

1. Webservices communicating with the gpl licensed code also need to be opened up.
2...
|
6...

it was truly genius..

Re:Previous post..

[Anonymous Coward]

I know what post you're thinking of, but it wasn't genius. It was a blatant troll. By the way, item number one has already been considered and rejected for inclusion in GPL.

Re:Previous post..

[david_thornley]

1. Webservices communicating with the gpl licensed code also need to be opened up.

Considered and rejected for the base GPLv3, put in the Affero GPLv3. The two licenses are explicitly compatible, so code under either can be mixed and matched. This was one of the more controversial provisions in the original draft.

Frankly, it's not clear to me that it's enforceable. The GPL is a license that is based on extending normal rights under copyright (although not as far as some would like). It seems to me that, if I have legally acquired software, without any further restrictions, I can probably run it on my web site for general use. (Or would this be public performance?) Therefore, I wouldn't have to accept the GPL in this case. Moreover, if it restricts rights normally available under copyright, it has to be a contract rather than a license, which brings up a lot of legal issues that aren't present normally.

(Note: as should be obvious, IANAL. This isn't legal advice. It isn't even illegal advice.)

Vista-ization

[Anonymous Coward]

GPLv3 is the Vista of Open Source Licenses!
Maybe it's time to coin a new term like Vistafied, or Vistication.

Not a Problem

[maz2331]

This isn't a problem, just a different choice of license. I personally do some LGPL v2 stuff, and won't use v3 since it's revision of LGPL is too restrictive for my tastes, while BSD is too free.

I can't say I'll never use or join a v3 project. It's all a matter of what goals the project has.

Not everyone wants to follow the full-blown Stallman ideas of socialist software but still wants to publicly collaborate and protect copyright. Others want to follow the BSD model, and some are into the full-blown over-the-top protectionist idea of proprietary code.

GPL-2 vs GPL-3 just like

[Anonymous Coward]

Fidel Castro vs. Fidel Castro's son.

Fud-mongering

[Vainglorious Coward]

Only 6% of developers...have adopted GPLv3...Two-thirds say they will not be adopting GPLv3 anytime in the next year, and 43% say they will never implement the new license

Interesting to compare this "shunning" with Vista :

less than 2% of UK-based firms have already upgraded all their desktops to Windows Vista. Just shy of 5% said that they have begun a Windows Vista desktop upgrade program. 6.5% said they will upgrade in the next 6 months; 12.6% in the next 12 months; 13% in the next 18 months; and 18% in the next two years

Summary : GPLv3 is more popular than Vista

Re:Fud-mongering

[magictongue]

Setting humor aside, there is lots of FUD being thrown around between the licenses. If one reads both license they are far more similar than different. GPLv3 just adds some clarifications, patent safeguards, and DRM safeguards. Over the years companies and individuals became quite creative in attempting to making open source code, close sourced. This is due mainly through creative reading and rationalizations that most like will not hold in court. Making the license clearer is a benefit since it reduces the attempt to read into it eliminating conflicts and misunderstandings. Just the other day I heard of a company that differentiated between static linking and dynamic linking. They argued that by dynamic linking the work is no longer derived. Hence, one can take open source code, compile it into a dynamic library, and then disregard the GPL. I am curious as to the real reason for the shunning? Is it FUD, do the developers have a real grievance, or are developers simply procrastinating.

Re:Fud-mongering

[Anonymous Coward]

For me, it mostly comes down to the hoops of the anti-tivoization stuff. A software license shouldn't attempt to define the hardware it can run on (spin it any way you want, a hardware restriction or a software restriction is still a restriction on use... users can't use it if you can't distribute it to them). Throw in the overly complex commercial versus home use verbiage which then says it's ok to not stick to the ideals of the entire "anti-tivoization loophole closure" if you're making devices for commercial use (however, if you take a commercial device home and use it there, the developer's protection under the commercial device is nullified even if the device was never meant for personal usage). Tack on the fact that those restrictions make the GPLv3 incompatible with the GPLv2 as well as the need for some distributors to fork GPLv2 code (or possibly switch to entirely closed code) because of the legal incompatibility. Mix in a little rush to change the license of major projects like gcc by a firm deadline before people can figure out how the files with exemptions will work under the new license.

Basically they made the license more obfuscated rather than more clear in several ways and drove a wedge of legal incompatibility between the GNU code base (and those who will follow the FSF like Samba) and the rest of the existing GPLv2 code base in an attempt to leverage that code to get other people to change their license to something they think is inferior but allows them to continue to do their work without having to fork side projects. All because RMS got pissed off that he was losing control and relevance in the community he helped start. Did I mention that, despite being a very strong GPLv2 proponent for about 15 years, the whole move made me quite anti-RMS and anti-FSF (especially Alexandre Oliva's attempt to browbeat the Linux devs on LKML and when he was losing the argument, threatening that someone could just relicense the entire kernel v3 (specifically the v2 only parts) regardless of what Linus, et al wanted). That said, the GPLv2 is still my license of choice and I'll lick the goatse guy's ass before I license anything I wrote under v3.

Evans Data = Marketing Research Firm

[Anonymous Coward]

Brag list of their clients and page includes their address phone numbers. Links to more info on the rest of their site included on the web page. They seem to specialize in doing Marketing Research within the tech community, have they ever talked to anyone here? Opinions?

Vista?

[athloi]

You mean the GPL3 is the Microsoft Vista of the open source licensing world?

Re:Vista?

[A beautiful mind]

Vista was just the marketing codename for that operating system.

Didn't you hear that Microsoft announced yesterday that the official designation for the OS is "XP, Millenium+7 Edition"?

Re:Vista?

[Anonymous Coward]

That's what the FUD-sters who wrote this article would have you believe.

Re:Vista?

[f8l_0e]

Be careful, Stallman might track you down and kill you for a comment like that.

GNU and Samba

[Random BedHead Ed]

All things GNU are going GPLv3, as is Samba. That's a pretty big and influential body of code, and it hasn't been established under the new license for very long. I'd give it time and I think you'll find version 2 will become more of a minority license. It could take years, and it will never be 100%, but I'd hardly call it a rift.

Why I don't like the GPL3

[larry bagina]

Much of my programming knowledge (especially API usage) came from looking at source code (this was before "Open Source" and the GPL was so popular). I was interested in learning and using, not redistributing, so the license didn't matter. Heck, once I learned assembly language, I didn't even need the source code.

I personally prefer a BSD-style license. For some purposes, a GPL (never closable) license is preferable. I think we've all heard the arguments enough times. But the GPL 3 isn't about sharing code, it's about enforcing a philosophy that I don't agree with.

Gee duh.

[heli_flyer]

The basic problem with this article is that it confuses Open Source with Free Software. They probably polled BSD, MSPL, Mozilla, etc developers and asked if they were planning to switch to the GPLv3, and as would be expected, most said "no". To me, it's just the obvious restated as something insightful.

shunnnnnnn

[Raleel]

shuuunnnnnnnnn the non-believer! shuuuuunnnnnnnna

First Charlie the Unicorn and now GPL3

[DigitalReverend]

What's next?

Pale-faced android... and TPS reports.

[fo0bar]

I first read "[Evans Data's CEO said]" as "[Even Data's CEO said]".

"Yeah.... I'm going to have to ask you to stay on duty an extra shift... Oh, and if you could Make It So, that'd be great. Yeah...."

The takeup is actually pretty strong

[ribuck]

A six percent takeup in a few months is pretty strong, and one-third of those polled haven't ruled out adopting it in the next year. Give it a few more years, and I think GPLv3 will be the dominant GPL version.

After all, how many projects still use GPL version 1?

Re:The takeup is actually pretty strong

[phantomlord]

After all, how many projects still use GPL version 1?

GNU GENERAL PUBLIC LICENSE
Version 1, February 1989

GNU GENERAL PUBLIC LICENSE
Version 2, June 1991

GNU GENERAL PUBLIC LICENSE
Version 3, 29 June 2007 Mind to venture a guess how many non-FSF/GNU projects were created in the 28 months that the GPLv1 was the current license versus how many projects were created in the 16 years that the GPLv2 was the current license? Predicting the dominance of the GPLv3 based on the current usage of GPLv1 is a little disingenuous.

As for the GPLv3 being the dominate license in a few years, I've read that RMS already wants to get a GPLv4 out soon. If it's within the next 3 years, the GPLv3 will barely have time to catch on and the rapid license changes will make the GPL look unstable to non-FSF zealots.

Re:The takeup is actually pretty strong

[Anonymous Coward]

I hope not. It's draconian idealism from an egotistical leader personified and adopted by zealots. Given the choice between GPLv2 and GPLv3, if GPLv2 is abandoned, I hope a widespread adoption of GPLv3 kills off FOSS.

Re:The takeup is actually pretty strong

[Whiney Mac Fanboy]

I've read that RMS already wants to get a GPLv4 out soon.

Where did you read that? I'm guessing it was in a slashdot comment made by someone just as gullible as you,

Re:Microsoft FUD, I imagine.

[Anonymous Coward]

No one cares what you imagine, Twitter.

The world operates on facts, not your paranoid fantasies.

Not exactly

[paladinwannabe2]

Something licensed that way can be used by both GPLv2 and GPLv3 projects, but can't use GPL3 code itself without converting to GPL3. It's still under GPLv2 until then.

Re:Not exactly

[JesseMcDonald]

Something licensed that way can be used by both GPLv2 and GPLv3 projects, but can't use GPL3 code itself without converting to GPL3.

I know that. I meant the code is released under the GPLv3 in addition to whatever other licenses apply (like the GPLv2). Of course it's still available under the GPLv2 terms, at least until someone accepts a non-GPLv2 patch (GPLv3-only / GPLv3-or-later). For that matter the historical versions will remain under (at least) the GPLv2 no matter what patches are later accepted.

It's still under GPLv2 until then.

Not quite: it's under GPLv2 and GPLv3.

Re:Not exactly

[sumdumass]

What? How can you claim something falls under a license when I can ignore a good portion of the limits the license imposes?

No, Section 9 says you have the option of following either version. You would have to change the license in order to have it count as GPLv3. Otherwise there would be nothing stopping me from ignoring your GPLv3 restrictions and just using the GPLv2. The GPLv2 says that derivative works much use that license (GPLv2). It also says no further restrictions so I'm not sure if you can even use GPLv3 code with it. Although some are saying the original work, unless changed by the copyright holders can have the GPLv3 code with it. But that would fly against the FSF's own compatibility matrix that says the two licenses are incompatible and cannot be used together in the same project.

So, RMS may have found the limit

[GroundBounce]

That's his job.

Re:Confused... M$? (or not)

[joe 155]

couldn't say who actually coughed up the money for this one, but they do list M$ as clients. We all know M$ aren't above what we might (generously) call "interesting" techniques when it comes to dealing with the GPL (not least, IIRC, calling it a "cancer"). Evans also list some (what I would call) nicer companies though - especially from the open source POV - including but not limited to RedHat and Sun. You can check out the full list here;

http://www.evansdata.com/company/clients.php

But does this actually mean anything?

[Timothy Brownawell]

43% will never use GPLv3. How does this compare to the number who will never use GPLv2 either (ie, the BSD folks)?

23% might use GPLv3, but not this year. Because there's really no reason to rush and/or the "improvement" might just not be worth the hassle of relicensing.

Using GPLv3 scares away twice as many people as it attracts. Using it instead of what? If this wasn't specified then people are probably comparing it to their "normal" license, in which case BSD people will be scared away and GPL people mostly won't care.

Only 6% have adopted GPLv3. Because really, what's the rush?

Really, I don't see that this says much of anything. They should ask better questions next time.

Uh...

[Keyper7]

GPLv3 is controversial because it imposes restrictions on what you can do with programs implemented under this license.

Isn't that what licenses usually do?

I know the rest of the paragraph clarifies what he meant, but isolating this sentence was not a very good idea.

"Or later" nullifies restrictions

[Chemisor]

Any project released "under the terms of the GPL v2 or later" gains nothing from GPLv3 since anybody can accept GPLv2 instead, and likely will. This means that all those new restrictions imposed by GPLv3 only apply to those people who want to abide by them, which obviously excludes any company trying to Tivoize that project or sue it for patent infringement. So to state that any "or later" project is now under GPLv3 is definitely incorrect. A chain is only as strong as the weakest link, and a multilicense project is only as restricted as its least restrictive license.

Math

[Anonymous Coward]

2/3 (~66.67%) + 43% + 6% > 100%

Who TF is Evans Data?

[LorenzoV]

I have no idea who Evans Data is, but the report referenced by the article reminds me a lot of the studies that Microsoft has paid for. As in: "He who pays the piper calls the tune".

Even if the data are correct, I'd suggest that there simply has not been sufficient time since GPL3 has come into being for new releases, hence new licensing, to have adopted it.

Of course, YMMV, ICBW, and all that.

Re:Who TF is Evans Data?

[Pop69]

Funnily enough, if you look, http://www.evansdata.com/company/clients.php, guess who one of their clients is ?

Re:Who TF is Evans Data?

[nerdacus]

Funnily enough, if you look, http://www.evansdata.com/company/clients.php, guess who one of their clients is ?

I'm waiting with baited breath...who? Red Hat? IBM? Apple? Oracle? Sun? Oh, I'm guessing you mean Microsoft... I suppose you're implying that because they have had Microsoft as a client at some point in the past that they are Microsoft's butt monkey or something. Are they also butt monkeys for the other names I listed? How can they be biased for all of their clients at once, when those clients are pushing diametrically opposed positions? Perhaps because they are impartial?

I happen to know someone who works at Evans Data. Yes, their clients are always looking for the results they'd like to see, but a good research company also needs to look out for its own reputation or risk losing credibility. They are scrupulous in their work, doing the research their clients ask for, but without putting a particular spin on the results. In the particular case of Microsoft, Evans Data has done at least a couple of studies for them. The first study they did came out with unfavorable results for MS, and MS subsequently dropped them as a client. MS came back at a later date after licking its wounds and commissioned further research because EDC is a reputable and legitimate company, and skilled at the kind of research they do.

Re:Who TF is Evans Data?

[Anonymous Coward]

Sorry, "baited" should have been "bated", I was too quick to hit teh submit button.

oops, wrong link.

[Anonymous Coward]

Try this one. Apologies for not checking my link before posting.

Now I have to wait to post the correction.:=(

There's a difference

[GroundBounce]

The change from GPL1 to GPL2 was more of a no-brainer. Even Linus adopted GPL2, and he's pretty much on the "business-friendly" side of the spectrum.

The thing that's new about GPL3, is that it tries to not only keep the code itself open and free, which I believe is a valid goal of a software license, but it tries to control *other* behaviors of an organization that are more marginally related to the code itself, such as patent cross-licensing agreements, etc. If a piece of software does not violate any known patents, then the license for that software should not restrict or control how a user conducts their patent cross-licensing or other aspects of their business.

Even though I may agree with some the philosophical aims of GPL3, I have a problem with a software code license that tries to reach out and control general business behavior of individuals or organizations. The GPL3 is basically an attempt to try to force organizations not to just keep the specific code open and free, but to align philosophically with more general business practices.

Re:There's a difference

[Ginger Unicorn]

That may be your interpretation of the motivation of the FSF, but RMS and Eben Moglen have stated quite clearly a number of times that reason they came up with GPLv3 was because GPLv2 has exploits involving patents and Tivoisation that can be exploited by unscrupulous people/businesses who want to circumvent the freedoms that are the raison detre of the GPL.

So the GPLv3 is a bug-fix. That's all. If jackasses like tivo and MS/Novell weren't trying to hack the GPLv2, I doubt this revision would have even crossed their minds. They arent trying to tell people what to do with their software - they're just trying to ensure everyone gets access to the source code, no matter what underhanded schemes other individuals might try to pull.

Re:There's a difference

[nmos]

I felt the way you do about the early drafts of the GPL3 but I mostly like the final version.

The thing that's new about GPL3, is that it tries to not only keep the code itself open and free, which I believe is a valid goal of a software license, but it tries to control *other* behaviors of an organization that are more marginally related to the code itself, such as patent cross-licensing agreements, etc.

I don't think that's the aim of the GPL3 at all. The purpose (as I understand it at least) is to keep the USERS of the code free to actually use and modify the code, not just look at it. The purpose of the patent wording is to keep someone from technically fulfilling their obligations under the GPL on one hand by releasing the relivent code while on the other hand threatening a patent lawsuit against anyone who actually tries to take advantage of those rights.

So, do you disagree with the goal above or do you just think they could have done it differently?

Re:There's a difference

[zotz]

"Even though I may agree with some the philosophical aims of GPL3, I have a problem with a software code license that tries to reach out and control general business behavior of individuals or organizations."

This is fairly interesting. Is the power in the present copyright laws to allow this? If so, is it there as a result of FSF pushing for changes to the law to make it so?

Mind you, if you accept free market ideas, and accept that there can be free markets in goods protected by copyrights and patents, (the latter of which I don't) I should be free to try the Paint Your House Red License. Those who don't want to paint their house red can choose not to use my software. (I don't really agree with what I just said, but...)

To bad we can't easily write meta licenses:

Meta GPL.

list four freedoms.

Instructions to lawyers and judges...

Insert clauses needed for your jurisdiction to ensure the four freedoms are protected with reguard to this code or any part of it in every possible case. If you find a loophole, plug it.

End of meta license.

all the best,

drew

Re:There's a difference

[GroundBounce]

Firstly, I need to admit that I haven't read every word of leagalese in the GPL 3. I'm going primarily from analyses and summaries of others. I think the main problem is using a software license to determine more than just how the code is used and redistributed. To me, the issue isn't threatening patent lawsuits - if a software author has violated a patent (no matter how much I feel software patents shouldn't exist), then (s)he is open to a potential patent lawsuit no matter what the license. The issue is that the GPL3 tries to tell me what kind of arrangements I can make with my customers *outside* of just how the code is treated. It's a bit like saying "you can't redistribute this code if you redistribute it along with porn, because *I* think porn is wrong".

In summary, as I mentioned previously, I agree with some of the goals and intentions, but I don't believe it is right for a software license to attempt to dictate behaviour other than how the code itself is treated.

Re:There's a difference

[einhverfr]

And the antitivoization provisions are entirely ineffective. All they say is that one cannot interfere with the running of a GPL v3 binary in any way. It does *not* mean the system as a whole has to do anything meaningful once that is installed. This can be accomplished by merely aggregating GPL v3 components with those which talk to the more interesting parts of the system and then forcing the upgrade as a unit. The GPL v3 components can be added but then cannot do anything interesting since there are other components of the system missing on the other ends of pipes.

This sort of scenario could also be used to implement robust DRM in such a way that replacing open source components would not compromise the system any more than sniffing traffic in an SSL session would give you credit cards. (RMS is factually wrong about the inability to lock users out of content in systems containing GPL v3 components.)

Those missing components can be required to be digitally signed with an encrpytion key to ensure they haven't been tampered with.

Re:There's a difference

[nmos]

The issue is that the GPL3 tries to tell me what kind of arrangements I can make with my customers *outside* of just how the code is treated.

??? What makes you think the GPL3 does anything like that?

In The Real World...

[BSDetector]

In the real world where real people do real things with real computers - they dont give a damn about any of this. Hell, they probably have no idea what you all babble about.

SourceForge says yes

[Chemisor]

> 43% will never use GPLv3. How does this compare to the number who will never use GPLv2 either (ie, the BSD folks)?

We might get some numbers from SourceForge:

There are currently 105950 projects registered.
Of those, 68143 are using the GPL and 11979 are using the LGPL, for a total of 80122 projects.
The remaining 25828 use different licenses, which I am too bored to break down.

43% of 105950 projects is 45558.

If we assume that only people who currently use the GPL (and not LGPL) will consider GPLv3, then only 7751 of them will be opposed to it, or 11%, which is certainly less. If the above assumption is true, it just means that the GPL camp will lose 11% of developers, or 25% if you include the LGPL people.

> Only 6% have adopted GPLv3. Because really, what's the rush?

When you release a new version of your product and only 6% of your customers decide to upgrade, it might be a good time to fire the person responsible for it.

Re:SourceForge says yes

[zotz]

"When you release a new version of your product and only 6% of your customers decide to upgrade, it might be a good time to fire the person responsible for it."

Not sure why. I think I like the ideas I think are in the GPLv3. But... I am busy. I have not even taken the time to read the new license carefully enough to know if I should switch my few unimportant projects over. Plus, most of the objections I see are from people who seem to want loopholes in the protection of the four freedoms.

all the best,

drew

Soup and Gravy

[cryfreedomlove]

What's the difference between soup and gravy? That's the question that comes to mind whenever I hear someone trying to describe the difference between open source and free. There's a lot of fud, flame, and passion out there and its very confusing for most people. Most people just want shit that works.

Re:Soup and Gravy

[Anonymous Coward]

>There's a lot of fud, flame, and passion out there and its very confusing for most people

That could be applied, mutatis mutandis, to just about any philosophical or political issue under the sun.

>Most people just want shit that works.

That has nothing to do with whether there is a valid distinction to be made between open source and free software. I'm sure most people don't care whether their furniture is made of fiberboard or plywood, but as someone with experience in carpentry, I can say it certainly makes a difference. Your entire post is essentially a fallacy of relevance.

Re:Soup and Gravy

[Angostura]

One is a method of distributing software, the other is an ideology.

Countersurvey

[heli_flyer]

In a survey of 1 Slashdot user, 100% of the users were found to think Evans Data are idiots for asking Open Source developers if they plan to switch to a Free Software license.

Re:Countersurvey

[Anonymous Coward]

83.72% of all statistics are made up on the spot.

They are from Santa Cruz!

[HaeMaker]

They are a Santa Cruz Operation!

Medieval Code: "Help! Help! I'm being repressed!"

[ClosedSource]

"GPL protects the freedom of the code"

I think this "freedom of the code" idea is nonsense. Both the GPL and the BSD allow the original work to be freely available. The primary difference between the licenses is how they treat changes or additions made to the original code. BSD doesn't require new work to be licensed under it and the GPL does. No anthropomorphism is required to understand the difference, although the creation of a non-existent "moral" issue might have some propaganda value.

Not quite

[einhverfr]

LedgerSMB is GPL v2 or later. At the moment there is no plan to change this to V3 or later.

Now, this means that if you fork LedgerSMB, you can decide to use the GPL v3 for your license if you can meet the terms of that license (which by my reading would require removing any dependency on BSD-licensed code since the additional permissions cannot be meaningfully removed in accordance with section 7 of the GPL v3). This probably means a major port from PostgreSQL, or at least the stored procedures, etc.

Re:Not quite

[JesseMcDonald]

. . . if you can meet the terms of that license (which by my reading would require removing any dependency on BSD-licensed code since the additional permissions cannot be meaningfully removed in accordance with section 7 of the GPL v3)

An important point, but if you re-read my original comment you'll notice that it does not apply to such cases, since not all the commits were GPLv2-or-later.

Re:Not quite

[einhverfr]

All of *our* code is GPL v2 or later.

However, the GPL v3 requires that certain dependencies are under the a license which allows for "relicensing" under the terms of the GPL v3 (see section 7). It is in the dependency area that you would run into trouble.

Re:Not quite

[fsmunoz]

ow, this means that if you fork LedgerSMB, you can decide to use the GPL v3 for your license if you can meet the terms of that license (which by my reading would require removing any dependency on BSD-licensed code since the additional permissions cannot be meaningfully removed in accordance with section 7 of the GPL v3). I'm beginning to think that the BSD licence needs to be tried in court, since recently I've heard so many different things I can't do with it.

More seriously I disagree with your view (I have had a good discussion about this and other GPL/BSD issues here, but i'll repeat it). Although not immediately apparent section 7 is actually more compatible with the BSD licence than GPLv2 - at least to those who support the view of the BSD/GPLv2 code sharing incompatibility:

(...)Point 7 of the GPLv3 has *two* different concepts: "aditional permissions" and "aditional requirements". "Aditional permissions" can in fact be removed, this is the situation e.g. in dual-licensing, someone received code under the GPL with aditional permissions that are however removable since they don't afect the fact that the code is under the GPL, only that whomever receives the code can also make use of the aditional permissions. Now, "aditional requirements" are different and that's where the BSDL requirements fall, and they can't be removed as aditional permissions can. The requirements listes cover the general requirements in BSDL/ISC/MIT et. al. and include the possibility of having them as a separate license file or mixed in in a single file.

Now, its probably not illegal to combine the licenses, however, there is a problem with that combination, and it might open you up to other attacks from both downstream users (who might accuse you of misrepresenting the legal rights you had to the software, particularly if they removed the BSD terms and then were sued by the original copyright holder of the BSD-licensed portion), or possibly the copyright holder of the BSD-licensed portion more directly under theories other than copyright violation (perhaps tortious interference with contract.) I see your concern but as per above I don't think that is the case. As in this article [fsf.org]:

A GPL licensee may place an additional requirement on code for which the licensee has or can give appropriate copyright permission, but only if that requirement falls within the list given in subsection 7b. Placement of any other kind of additional requirement continues to be a violation of the license. Additional requirements that are in the 7b list may not be removed, but if a user receives GPL'd code that purports to include an additional requirement not in the 7b list, the user may remove that requirement. Here we were particularly concerned to address the problem of program authors who purport to license their works in a misleading and possibly self-contradictory fashion, using the GPL together with unacceptable added restrictions that would make those works non-free software.
(...)
The list was made to acommodate licenses which were not compatible, so it wasn't made for the BSDL/MIT/ISC licensed but for others which had other minor requirements that were not restrictive but merely incompatible due to the wording of the GPLv2 (this following the FSF reasoning on the compatibility of the BSDL license with the GPL, which you disagree with). Cheers. So I don't think Section 7 means what some people make of it - thoug part of the fault is in defining two different concepts and naming the section after only one of them.

Re:Not quite

[fsmunoz]

Mangled the quoting a bit and left the other poster comments in the middle, but more importantly left out the link to Opinion on Additional Terms in the FSF site.

Re:Not quite

[JesseMcDonald]

Very well; I see I was being too narrow by using the term "commits". What I meant was all the code used in the project, whether part of the main program or any of its library dependencies; basically anything the final program could be considered a derivative work of.

Re:Not quite

[einhverfr]

The problem is far more subtle than most realize.

The basic issue has to do with whether or not the BSD code can be "relicensed" (in RMS and Eben Moglen's words) as GPL code without making any changes. While it is clear that copyright-worthy changes can be under any license, the question is whether the original code licensed under the BSD license can be. IMHO (IANAL) this talk of "relicensing" seems like legal mumbo-jumbo devoid of any accepted meaning.

The problem is that the GPL 3 is only compatible with licenses which allow this "relicensing" *independant of* other copyrights being enforced. This is clear in the Rationale documents, and in the opinions of both RMS and Mr Moglen. In other words, it requires that I can extend the requirements of the GPL3 to any and all parts of the code and any dependencies not specifically excluded from the Corresponding Source requirements without enforcing any of my own copyrights in the process.

I think that the BSDL always follows copyrighted elements released under that license and cannot be removed because this would require removing the permission grant from the code. Hence the additional permissions cannot meaningfully be removed without adding substantive code to the file. Because the GPL 3 requires that this is possible by merely conveying the software, it seems to me that this is a big problem. The authors of many of our dependencies agree and out of respect for them, I won't support moving the license.

Re:Not quite

[einhverfr]

One of the problems you could also run into (IANAL) is that 7(b) "reasonable" legal notices are not defined. It is arguable that the requirement to contain a permission grant in a license which also requires that the permissions can be arbitrarily revoked outside of the copyright license, might not be "reasonable" for the purpose of the license. Ie. the purpose of the legal notice is to essentially add restrictions as to the excersize of the permission to remove additional permissions according to section 7.

Re:Not quite

[fsmunoz]

I continue to disagree, additional permissions and allowed adittional restrictions are clearly enough defined (i.e. defining them more would defeat the purpose).

In any event here is the fresh opinion of SFLC:

(...)he GPL clarifies its copyleft requirement by explicitly prohibiting imposition of "further restrictions" on downstream recipients' exercise of GPL-derived rights. A condition in a non-GPL license covering some incorporated code, however liberal or simple such a license is, is certain to be different from the terms of the GPL in at least a literal sense. However, the meaning of "further restrictions" under GPL version 2 (GPLv2) has not been read in a literalist fashion, but rather has been elaborated over time by the communities developing, distributing, modifying and using code under that license, as a matter of custom. The treatment of notice preservation requirements in non-GPL licenses is a case in point.

The kinds of notice preservation requirements commonly found in permissive licenses are different from counterpart requirements in the GPL, but they are, as a rule, similar in nature and purpose and no more burdensome than the GPL requirements. For example, section 1 of GPLv2 requires that anyone making or distributing a copy of the program "publish on each copy an appropriate copyright notice" and "keep intact all the notices that refer to ...the absence of any warranty". The GPL also requires that distributors accompany all copies with the license text. The existence of such requirements in the GPL justifies regarding the comparable requirements in permissive licenses as not being "further" restrictions in relation to the GPL.

Section 7 of GPL version 3 (GPLv3) codifies this GPLv2 interpretive tradition, explicitly allowing contributors to attach "non-permissive additional terms" to the material they contribute if those terms fall within a list of acceptable categories. Such terms supplement the requirements of GPLv3 itself and are not considered "further restrictions". Among those categories are terms "[d]isclaiming warranty or limiting liability differently" from the disclaimers in the GPL and terms "[r]equiring preservation of specified reasonable legal notices or author attributions". (...) They are lawyers, although I'm not of the opinion that one should surrender ones opinion just because of that. It should be taken into account though.

Re:Not quite

[einhverfr]

The question is not answered by the SFLC in the above document.

Basically, does a notice of permission grant, i.e. "Permission is hereby granted, free of charge, to anyone obtaining a copy of ths software and the accompanying documentation...." constitute a reasonable legal notice for the purpose of section 7b especially since it abridges the ability to excersize all rights under section 7?

If so, can I add arbitrary license exceptions and force notice of these even if the GPL v3 does not allow me to ensure that the license exception are guaranteed downstream under section 7.

Re:Not quite

[fsmunoz]

The "Permission is hereby granted, free of charge, to anyone obtaining a copy of ths software and the accompanying documentation...." constitutes a reasonable legal notice. And the GPLv2 does ensure that the licences are guaranteed downstream, you can't remove restrictions, and that permissions grant is a restriction: more exactly the fact that you must display it.

If I'm not understanding you correctly is likely my fault though... perhaps a further example would be helpful (this is not in jest).

Re:Not quite

[einhverfr]

My concern is that section 7 allows anyone who merely conveys the software to remove additional permissions beyond those of the stock GPL v3 from "any portion of" the covered work (see section 7, paragraph 2). This is not the same as encumbering a BSDL work with new code under a different license because modification of the code is not required to excersize this right.

My argument is that the notice of permission grant effectively abridges the granted rights in section 7. I.e. one cannot then just declare the BSD permission grant void just by virtue of the fact that you distribute the software. In short it functions as an additional restriction beyond the mere requirement to include the notice (i.e. it restricts the ability to remove permissions granted to the code as granted in section 7, paragraph 2). For this reason, I don't think it is safe to rely on 7(b) legal notice exceptions for a compatibility argument in part because the notice is in conflict with the rest of the license (and therefore might not be "reasonable").

anti-GPL != anti-oss

[Anonymous Coward]

NT

Re:anti-GPL != anti-oss

[Ajehals]

I didn't suggest otherwise.

I plan to move my project to GPLv3, but....

[anwyn]

I plan to wait a year to let the bugs in this license shake out. I believe that GPLv3 is an improvement, but I am willing to let somebody's else's project be the guinea pig.

One more word

[texwtf]

WTF?

Re:One more word

[Salsaman]

Don`t worry about it. It`s part of M$ current FUD program, which is to make it appear like there are splits and rifts in the Free Software movement.

Nothing to see here.

Remember? Relinquish, rather

[Anonymous Coward]

That is so flabbergasting that I wonder if you're mocking your own argument with hyperbolic satire.

"It is important to take freedom away to protect it. Truly allowing freedom would allow freedom to be taken away, and we can't allow that, so we've taken away some freedom to allow true freedom to flourish."

What doublespeak. Submission is freedom if War is peace. "True freedom" you say? The idea of freedom painted by that comment is perverse, and not widely held by humanity (but nonetheless, is apathetically acquiesced to). That claim is "No true Scotsmen", and the limitations on freedom you suggest to make it "true" must be heinous (sorry-- coincidence with the name) indeed to place it so far away from what most English speakers mean by "freedom".

"Those restrictions are for your freedom."
No; those restrictions are UPON your freedom, as you've already stated! Doublethink. Those restrictions are for the freedom of the code itself, and the freedom of computer code or any other non-living, let alone inanimate, let alone ABSTRACT thing is far different from the freedom Patrick Henry held more important than his very life. Using the word "freedom" to ascribe rights to computer code perverts the word in the minds of others, and if you're still more careless, its meaning in your own mind as well.

"And I'm sure I'll get modded down..."
I hope not; there's no option for "-1, I disagree" or "-1, misguided" etc. Flamebait or Troll maybe, but the idea you presented is Stallman's reasoning exactly. It's not ineffective, but it is disingenuous for him or the zealots he's attracted and programmed to claim any sort of moral high ground despite limiting the freedom of their users. This is exactly the sort of thing we ought to discuss on Slashdot, since we've already decided it's worth our time to dig through these discussion threads.

Stallman's idea of freedom is wrong-headed, and it does make him, ipso facto a fringe thinker. Fringe thinking may be advanced or perverse, but in this case it's perverse. He created a truly great hack that used the bait of costless software to gain a legal beachhead. But, his rants (calmly prepared, but still rants or litanies in spirit) do damage to the image and cause of non-commercial software even as the GPL expands its veil of coercion virally from developer to developer with the code.

CAPTCHA: gullible

this will end sometime...

[Anonymous Coward]

...the shunning of gpl3. Once the great software patent wars start in earnest. So far, it's just been scouting expeditions and the occasional sniper, wait until "large & clearly threatened" multi billion dollar software industries realize they are on the way down. SCO was a joke of a contest, wait until the real action starts. FUD or no FUD, it is going to happen, and the way the legal system is, you could be 100% in the clear and still be forced into complete bankruptcy. gpl3 was in part written to address this issue before it becomes necessary to have the protection. Ignore that at your own peril.

Linus is a great coder,but he is a bonehead of a tactician and insists on living in the past. Both those things are truthiness, said without malice, just as obvious observations. You need the long view and to keep your head out of the sand. You may want to ignore reality, but in $oviet deep-pocket$ amerikkka, reality is NOT going to ignore YOU....

Re:this will end sometime...

[dbIII]

Linus is a great coder,but he is a bonehead of a tactician and insists on living in the past. Both those things are truthiness

Three things - linux is not a gnu project so the tactics are somebody else's problem, I really do not understand the doubleplusgood newspeak and broad software patents are an astoundingly stupid idea that get in the way of everyone's code no matter what you do.

How many?

[Anonymous Coward]

"After all, how many projects still use GPL version 1?"

That's a terrible argument. Did version 2 represent a predominantly compelling improvement over version 1? Was the size or value of any such improvement bigger or smaller than, the difference between versions 3 and 2? Is that "improvement" actually negative?

It seems to me that version 3 is merely RMS revealing his anti-commercial motivation*, and this new light and the restrictions imposed by the new version are starting to make developers think more seriously about the "GNU General Public License" instead of merely "the license that SOMEONE calls 'General Public License' which seems pretty popular, more out of habit or convenience than practicality or ideology".

Seems to me...

[Anonymous Coward]

There are no hobbyist anymore, the way the Great Linus (& some would say Bill) started
Few, if any, study CS
All open source projects, of note, have been adopted by M$ competitors
GPLv3 Needs heavy duty REAL lawyers (with REAL lawyers prices) to interpret
People shout up Python over assembly, FFS!
Knowing (a/any) Linux (or other S/W of significance) inside out AND knowing how to use it in some other realm, has long been a team effort
People shout down MONO!
No one knows how to write games for the PS3
Everyone is ignorant of threads despite 15+ years of industry use
All s/w contains killer bugs which result in no DRM no internet safety
All s/w is sold/ given away with contracts/conditions of use/licences which are self-denying, unenforceable and unreadable (it is assumed)

You have two choices, use your BIG BRAIN to study physics/maths or study LAW
Studying CS means you have one real employer, M$, or none, OS

Re:Seems to me...

[erikvcl]

What the hell are you talking about? What is your point?

End of the GPL monopoly?

[Kawahee]

Clearly this marks the downward spiral of GPL's monopoly on FOSS!

Re:Microsoft FUD, I imagine.

[Anonymous Coward]

That FUD fest you link to in your sockpuppet's journal seems to suffer from the same shortcomings.

Please undertand the GPL first

[Anonymous Coward]

"The GPL isn't about preventing commercial use. If you bothered to read the license(s) you would know that."

The GPL does, indeed, explicitly allow the sale of the code it licenses. However, it also explicitly attempts to remove any value from the creative or technologically constructive act of creating such code. The license strictly requires each recipient to provide subsequent recipients with the source code and binary under the same conditions! The value of creating the code to zero because the cost of licensing the code goes to zero, leaving only the cost of distribution (plus, optionally, the cost of supporting or modifying the code under contract).

So yeah, you can sure sell GPL code, but after EVEN ONE person does so, the license has freed that buyer to give away the binary and the source to anyone else as well! If you bothered to read the license(s), you would know that. It was designed to introduce infinite elasticity to the supply-and-demand formula to drive the code production component to zero. It was well-engineered, and it works. It lets the cat out of the bag on the very first release under the licensing terms, as Richard Stallman intended.

The GPL is all about preventing commercial use. If you bothered to think about the license(s) after reading, you would realize that.

Re:Please undertand the GPL first

[Grishnakh]

The GPL is all about preventing commercial use. If you bothered to think about the license(s) after reading, you would realize that.

You're misinterpreting the words "preventing commercial use".

What you really mean is "steal someone else's code, make a few modifications, and sell it as your own." That's basically stealing, because you've added no real value.

What I mean by "commercial use" is "a company is allowed to use GPL code and incorporate it into their own product, as long as they release the changes to customers. That way they don't have to reinvent the wheel."

If you're Linksys, for example, it's cheaper and easier to use Linux as the base OS on your router, rather than writing your own OS from scratch. But part of the bargain is that you have to share the modifications you make with your customers. This doesn't mean you have to provide your customers with schematics to the device, or even the source code to proprietary software that runs on top of Linux, but the portion that you borrowed from someone else (Linux), you have to provide the changes.

However, if you're CrapSoft Inc., and want to make a special version of Linux to sell to the government for a juicy contract, but you want to keep your changes secret so the customer is dependent on you, that's not allowed. Do you honestly see a problem with that? If so, I suggest you get to work writing BSD-licensed workalikes to all GPL software, because you have no right to tell other software developers under what terms they're allowed to share and distribute their code.

Just wait

[Chris Snook]

Once adoption of version 3 slows, they'll just release version 3.5 and we'll all have to buy new books anyway.

Re:Just wait

[Dr. Photo]

I attempt to disbelieve!

It seems to indicate success

[Burz]

And adoption of GPLv3 will never reach the lions' share of projects.

Why? Because the 'split' the article talks about indicates that a significant set of coders on either side want different terms. And they are now both served by different versions of the GPL, instead of sticking with v2 and leaving a sizable chunk of hardware-freedom advocates without a common license. IOW, the addition of v3 has successfully expanded our options.

WRT to the owners, free hardware was all we had up until DRM and in fact the whole body of free software discussion starkly embodies that assumption. Now it can no longer be taken for granted and so freely re-programmable hardware must be asserted as a prerequisite for free software culture: The GPLv3 does just that.

The forgotten survey question...

[3seas]

Do you understand teh GPL v3?

Re:The forgotten survey question...

[Anonymous Coward]

People, even F/OSS proponents, can understand the GPLv3 and still disagree with it.

Re:The forgotten survey question...

[VGPowerlord]

Do you understand the words that are comin' out of my keyboard?

That's an interesting question. Let's have at it.

[fyngyrz]

Can you think of a better wording for a license that will protect the four freedoms in all situations?

The better choice is PD, as in, I release this source code as PD - you can do anything you like with it. The FSF's four freedoms are:

1. Run the program, for any purpose
2. Study how the program works, and adapt it to your needs
3. Redistribute copies so you can help your neighbor
4. Improve the program, and release your improvements to the public, so that the whole community benefits

For #1, PD allows you to do this, and it also allows an end user to run it as a client of a closed source, commercial project, which the GPL disallows.

For #2, PD allows you to do this as well, only the "your needs" that you can adapt it to are broader; you can do anything that you could do under the GPL, but you can also implement a fully commercial, closed source modification (or use) of the PD code in distribution.

For #3, you can redistribute or not, just as under the GPL. But you can also redistribute as closed source, modified or not, which you cannot do under the GPL.

For #4, you can do anything the GPL allows here, and you can also improve it (or just use it) and release it in proprietary executable form so you can benefit people in a way that the GPL does not allow.

Clearly, if the four freedoms are the critical issue, then PD is a far superior mechanism for implementing them. You said you were looking for a mechanism to protect the four freedoms in "any" situation; it is obvious that the GPL protects them in considerably fewer situations than does PD. As near as I can figure out, nothing protects them in "all" situations, not even PD.

What the GPL actually accomplishes is the erection of a legal structure that must be navigated or avoided by those people who choose to be downstream clients of GPL'd source. This means that lawyers have an opportunity to earn money, people are restricted in how they can use such projects in ways they would not have been if the project had been PD, and (here we get into the actual reason for the GPL, or so my instincts tell me) any changes or improvements to the code must be made available to everyone if the code is redistributed in any form.

So a value exchange — new changes going to the community in exchange for the use of the unmodified project — is a behavior forced upon the users of GPL'd code. The freedom to do this of course exists with PD, but it isn't freedom to do this that the GPL is actually working to implement; it is the requirement to do it.

As other posters have said before me, by all means, if the GPL is what you want to use, then use it.

Just don't think it is freedom that you're getting. What you're getting is restriction at the expense of freedom. Which may be precisely what you want. Just be sure it is. If the four freedoms are what you want to protect and foster, then PD is a much better choice.

Me, I'll either go commercial, or I'll go PD. The former is a means to earn a living with what I consider to be the strongest potential to earn; the latter is a means to benefit the developer community in the broadest possible sense, which in turn can benefit others downstream from the developers. The GPL, as far as I am concerned, isn't even in the running.

compared to Vista adoption...

[m2943]

Seems to me like compared to Vista, people are falling all over themselves to adopt GPLv3.

Seriously: people don't change licenses over night. In fact, projects developed under a GPLv2-only license often simply have no possibility of changing to the GPLv3. If only 43% say they will never change, that's good news because that means that nearly 60% of developers potentially could change. That's actually higher than I would have thought.

who's responsible?

[falconwolf]

It is the legal system's fault normal people need a legal interpreter in order to conduct official business.

No, it's "normal people's" fault because they allow it to be that way, at least they don't protest and vote to uphold their values.

Falocn

freedom

[falconwolf]

You aren't free to hit me because you've had that freedom taken away from you.

Your (My) freedom to hit someone (you) ends where their (your) nose begins.

Falcon

You might as well tell me to repeal the Patriot Ac

[falconwolf]

Is this Justice Roberts? If so please strike the Patriot Act down.

Falcon

truth, damned truth, and statistics?

[swordgeek]

How many ways can a survey be skewed, either deliberately or accidentally? Lots. More than anyone likely cares to count.

However, the question of the survey is entirely valid, and I think the results are likely true as a philosophy.

Lots of people in the community are really unhappy with the bent of the GPL3. That's the take home message here. Is it 65% or 63% or 97%? I don't really know or care, because that doesn't change the fact that it _is_ controversial among the very group that it's supposed to empower.

Re:truth, damned truth, and statistics?

[einhverfr]

In my (somewhat limited) contact with other FOSS developers, I have heard from GPL supporters and detractors. Nobody in my immediate social circle of developers likes the GPL v3 though.

No, you're wrong.

[Anonymous Coward]

You are misquoting the GPL. It states:
If the Program specifies a version number of this License which applies to it and
"any later version", you have the option of following the terms and
conditions either of that version or of any later version published by
the Free Software Foundation. If the Program does not specify a
version number of this License, you may choose any version ever
published by the Free Software Foundation.

While this states that you may, optionally, follow the GPLv3 licence on any piece of GPLv2 software, it does *not* mean that GPLv2 software is bound by what is in GPLv3 and nor does it allow you to replace the GPLv2 licence with GPLv3, whether it is for your own internal use, redistribution or otherwise. It is up to each and every individual who obtains a GPLv2 package whether or not they want to decide if they will abide by GPLv3 for it or not.

Re:That's an interesting question. Let's have at i

[zotz]

Please, I am not that dense and it is not that late...

"For #3, you can redistribute or not, just as under the GPL. But you can also redistribute as closed source, modified or not, which you cannot do under the GPL."

And there go the freedoms in some situations. If you want to honestly try to answer the question, fine. If you just want to have the freedom to restrict the freedom of others, let's not argue, we just disagree.

all the best,

drew

No, no.

[crhylove]

The GPLv3 might actually WORK in some cases. Vista, on the other hand....

There is no "freedom to enslave".

[Anonymous Coward]

I'm not sure if you are being sarcastic or not, but if you are, then please be aware that taking freedoms away to protect other freedoms is the basis of all law. You aren't free to hit me because you've had that freedom taken away from you.

Unless you are an anarchist, you really have no basis for criticising the GPL in this regard, because you agree with this logic applied to different areas.

Nice of you to preclassify dissent on that manner, but you are wrong. Anarchism is not the antithesis of your position -- freedom itself is.

"Taking away freedoms to protect other freedoms" is the kind of Orwellian contradiction that has resulted from the failure to grasp that freedom is *logically* self-defining. It is not society or law that draws the line between your nose and my fist -- it is the logic of the idea of individual rights itself. It is simply this: if rights are based on the nature of man -- "inalienable" and all that -- they must apply to everyone equally. Freedom of action, properly understood, ends where the next guy's freedom begins. That is the meaning of "your right to swing your fist ends where my nose begins"; it is part and parcel of the principle itself, before the questions of law and society even come up.

Under your notion, the reason why we don't have slavery is because the government takes away the "freedom to enslave". If you can't recognize that for the contradictory dead-end it is -- a dead end to which your premise logically leads -- then it's no wonder you like GPLv3.

Freedom is not the mere physical capacity to do something; it is a moral principle that defines what capacities a man has the *moral right* to exercise. Freedom is not what society deigns to give the individual; it is antecedent to society. Society is not its source; it can secure it, or destroy it, but it cannot alter its nature. Freedom is what protects the individual from society; it is a constraint upon it, one that the latter has no power to abrogate under conditions of uncompromised political freedom. Liberty is not defined or delimited by any person or group, but is a logically self-defining principle, properly understood, and as such is therefore antecedent to any law. It is the source of the rule of law.

It is most definitely NOT anarchy.

Re:That's an interesting question. Let's have at i

[fyngyrz]

And there go the freedoms in some situations.

None of the freedoms are in the least compromised by that as far as I can tell. In no way is the code prevented from being distributed in any of the original ways from the state it was in at the time. If you think so, specify exactly how you see that happening. Claiming it does without specifying how is specious.

If you want to honestly try to answer the question, fine. If you just want to have the freedom to restrict the freedom of others, let's not argue, we just disagree.

I *did* honestly answer the question. Don't be rude.

Proxy

[tepples]

No, nor the GPL v2. The FSF recommended including that clause, but it's silly to release your code under a license you haven't seen. GPLv3 section 14 addresses this by introducing a new role in the licensing of a work of free software, the "proxy":

If the Program specifies that a proxy can decide which future versions of the GNU General Public License can be used, that proxy's public statement of acceptance of a version permanently authorizes you to choose that version for the Program.

Does GPL violate itself?

[Enderandrew]

I'm honestly not trying to troll here, but ask a serious, and simple question.

Doesn't the GPL state that once rights are granted, you may never take away rights? Isn't the GPLv3 more restrictive than GPLv2? Aren't projects that switch from GPLv2 to v3 like Samba now removing rights previously granted under the old license, and thusly violating that license?

Re:Does GPL violate itself?

[Rhapsody Scarlet]

Doesn't the GPL state that once rights are granted, you may never take away rights? Isn't the GPLv3 more restrictive than GPLv2? Aren't projects that switch from GPLv2 to v3 like Samba now removing rights previously granted under the old license, and thusly violating that license?

I believe how it legally works is like this:

Software starts out GPLv2 only. Agreements are gotten from all developers to re-license all code under both the GPLv2 and the GPLv3 (or code from developers that don't agree is rewritten under the GPLv3). Extra code is then added under the GPLv3 only.

All code that was under the GPLv2 is still under the GPLv2 and can be used as such, but is also licensed under the GPLv3 and the whole program can only be distributed under the GPLv3 because some code is GPLv3 only. So no violation since no extra restrictions have been added to any GPLv2 code.

I came to this conclusion after a long and very unproductive debate on whether all versions of MAME are under the GPLv2 because they're derivative works of MAME 0.10 (which was under the GPLv2).

Re:That's an interesting question. Let's have at i

[zotz]

I am not being rude. If you have issues with copyleft in general, say so. If you like copyleft, but have issues with the GPL itself, say so.

Just in case you missed the context:

"The thing is, I don't disagree with you. I'm fine with the GPL in any form, because frankly, it's the author's decision to use it and I respect that. My point is that it's a confusing muddle to wade through,"

and

""I'm not saying don't use it. I'm not even saying it's bad. I'm saying it is murky, and the article makes perfect sense to me because of that murkiness."

To which I responded:

Could be. Can you think of a better wording for a license that will protect the four freedoms in all situations?

Then you come along with:

"The better choice is PD, as in, I release this source code as PD - you can do anything you like with it. The FSF's four freedoms are:"

Did you misunderstand the context or my question that badly?

PD may indeed give the four freedoms to the first generation as does the BSD. But neither protect those freedoms for subsequent generations.

I asked a question in the context of an agreement with copyleft but seeking for better language. You come along with the old BSD vs GPL argument with PD as a stand in for the BSD.

But, just so you know. If the government wants to do away with copyright and put everything in the public domain, I will kick up less over that than I do now over how the system is run.

Now, if you would like to answer the question in light of better language for a copyleft license, fine. If you do not like the idea of copyleft, that is not an argument I am looking to have in this thread. Perhaps we will run into each other in another thread.

all the best,

drew

"The SCO Group" is from Utah, not Santa Cruz

[walterbyrd]

I think that's what you were refering to. The company that filed the bogus lawsuit against IBM, and told all Linux users to pay them $699 per CPU, is *not* Santa Cruz Operations i.e SCO.

Caldera simply adopted the name "The SCO Group" specifically to confuse the public, and the courts. In their latest filings to the bankruptcy court, The SCO Group, carries on about how "SCO" (what they call themselves) used to earn $230 million revenues. But the "SCO" they are refering to is a different company.

No.

[keeboo]

The copyright holder may license each different version as he pleases, he may even make a parallel commercial release.
The one able to violate such license is the one who receives the code already under such license.

Re:No.

[Enderandrew]

That simply is not true.

I can't take GPL code and then relicense it under a closed-source license. The entire purpose of the GPL, is that once you write code under the GPL, every right is forever protected, and can never be taken away.

"You may not impose any further restrictions on the recipients' exercise of the rights granted herein."

With BSD licensed software, you can relicense it largely, since BSD offers very little in the way of protection.

Re:No.

[Enderandrew]

To clarify, and I should have put this in my first response, there is no commercial restriction in the GPL, so yes, you can have a commercial release at any time. The GPL doesn't say you can't make money off software, but even commercial GPL products must release source code for free, and follow the GPL completely.

If you take GPL code and then reuse it, it must remain GPL protected, and rights can't be removed.

So let's take Samba. Earlier versions were GPLv2, and when they made later versions of Samba, they included GPLv2 code, thusly it must retain all rights and freedoms of the previous license.

Everyone is so quick to point out the freedoms it protects, but it does so by placing restrictions. The new restrictions, such as the Tivo/Hypervisor clause effectively remove rights from the previous version.

Given that the GPL says you can't remove any rights already granted, isn't this a contradiction?

Re:No.

[keeboo]

That simply is not true.

I can't take GPL code and then relicense it under a closed-source license. The entire purpose of the GPL, is that once you write code under the GPL, every right is forever protected, and can never be taken away.

We're not talking about the same thing.

GPL code is under the copyright of someone, right?
A work simply being copyrighted means that nobody - except the copyright holder - has distribution rights.

Then the copyright holder decides to make it free open source software.
He releases his copyrighted work under the GPL, thus giving rights to the one who receives that. Now you (as an user) have some rights but, still, do not hold the copyright of that work.

The copyright holder cannot revoke that specific release under the GPL, you cannot lose those rights previously granted (unless you violated the license, but that's another history).

Still, the copyright holder remains being so. As such he may release a different version under a different license, or even release it as closed source. He's not closing what was previously opened, he just took his (let's say) "internal version" and decided not to license it under the GPL as the other version.

"You may not impose any further restrictions on the recipients' exercise of the rights granted herein."

As I said, it applies only to code released under the GPL.
Unless the copyright holder states that his root development tree is under the GPL aswell.

With BSD licensed software, you can relicense it largely, since BSD offers very little in the way of protection.

Are you suggesting you can replace BSD by GPL?
I used to think that way, but the BSD license says nothing about such right. You may distribute under the GPL, but the BSD is still there.
Granted, in practice the result works like a code licensed under the GPL only.

Re:No.

[keeboo]

Let me put things this way:

You hold the copyright of XYZ software, no license applied at his point.
You release a derivative (or claim it is a derivative) of that work, but release under the GPL license.
What is changed to the root version of yours? Nothing.

Now, you take the root version (not the GPL-licensed one) and release as a closed software. -- You're the copyright holder and that version had no license attached to it, the GPL is simply not present here.

Re:No.

[Enderandrew]

Yep, and Samba (as an example) wasn't coded by one individual. The Samba page lists 47 people, though I'm not sure that counts all contributions from all coders in the history of the project.

When it was discussed on the LKML about relicensing the kernel under GPLv2, Linux keeps bringing up a major reason it likely won't ever happen. Linus can relicense his code, but he can't relicense the contributions of other people.

Did every single contributer to Samba over the years each stipulate that they have signed off on relicensing their work?

Regardless, I'm not sure even the original copyright holder can remove rights granted by the GPL, given that software licenses are contracts.

From the GPL:

"This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language."

Later versions or forks, are derivative work under copyright law, even if they are developed by largely the same team. The license stipulates that certain rights are granted, and may not be removed. Where does the license stipulate an exception that the original author can suddenly revoke rights for derivative works?

Re:No.

[keeboo]

Yep, and Samba (as an example) wasn't coded by one individual. The Samba page lists 47 people, though I'm not sure that counts all contributions from all coders in the history of the project.

Right, and who's the copyright holder?
Think of a closed software developed by persons A, B and C hired by company XYZ. The XYZ company holds the copyright and it's the one who have a say on that software.

A contributor to a FOSS project isn't that much different license-wise. Unless he contributes the code he claims copyright.
Or, a similar situation, when you merge a FOSS project (assuming the respective licenses are compatible) with different copyright holders, in that case the resulting software have both as copyright holders.

Later versions or forks, are derivative work under copyright law, even if they are developed by largely the same team. The license stipulates that certain rights are granted, and may not be removed. Where does the license stipulate an exception that the original author can suddenly revoke rights for derivative works?

But who's saying the author is closing an already GPL'ed version? He may claim he has the original free-of-any-licenses version, and that the GPL version is just a derivative of that original version.
He, then, releases the closed version claiming it was derived from that original version. No GPL involved at all.

Re:No.

[Enderandrew]

"Right, and who's the copyright holder?"

With most FOSS projects, each person maintains copyright on their contributions.

Regardless, I've been looking around, and from everything I'm finding, even original copyright holders relinquish portions of their rights by releasing their code under an open license, which they are now bound by. A license is a contract that binds both parties, which is why the GPL stipulates the no warranty clause.

The bulk of your argument is that if you code a project, you're not bound by the license you release your code under since you ultimately own it. I don't claim to know 100%, but from what I can tell, I'm not sure that is the case. I'd like to see a proper clarification on this.

If you are bound by the license, then it would seem the GPL are in fact violating their own license, though in practice they aren't going to go after themselves. However, it does add a layer of hypocrisy to projects shifting from GPlv2 to GPLv3, and it casts a legal doubt of whether or not they should be able to switch.

Re:No.

[einhverfr]

Regardless, I'm not sure even the original copyright holder can remove rights granted by the GPL, given that software licenses are contracts. Is there any reason the original copyright holder can't take the code and use it in a closed source application?

I.e. you can't revoke the license for a given release. You *can* take your own code and do whatever you want with it since you can't violate your own copyright.

Re:No.

[Enderandrew]

Honestly, I think this is the crux of the question, and I would like to see some legal clarification.

Is the original author bound by the terms of the license under which they release software?

I see license as a two-way binding contract, however it is possible I don't understand the legality here.

Re:No.

[jstomel]

Did every single contributer to Samba over the years each stipulate that they have signed off on relicensing their work? Actually, yes. I think. Samba is a GNU project and doesn't the GNU have a policy of not accepting code unless the copyright for the code is donated to the FSF?

Re:No.

[einhverfr]

Sure, the license is a 2-way binding contract (otherwise arguments of requiring specific performance in GPL violation cases would be obviously nonsense). However, the licensor is not obligated under that contract to avoid using his code under other licenses. He/she is merely obligated to allow people to use the GPL'd releases under that license.

Look at SugarCRM for example. They have a free/ open source version. And they have more advanced closed source versions. They can't decide one day that the GPl doesn't apply to the GPL'd version. They can, however, continue to use their own GPL'd code in their own proprietary versions.

IANAL, etc.

gah, more mindless meme amplification

[Anonymous Coward]

The BSD license protects developer freedom? Like when I take your BSD code and roll it into my proprietary product and deny you and everyone else any information about how I was able to use your work? Thank you for giving me my "freedom" at your expense, I suppose.

The only "freedom" the BSD license gives developers that the GPL does not is the freedom to take freedom away from other potential developers. Do the arithmetic for your shortsighted freedom and what do you end up with? Not much.

And what is this "freedom for code" nonsense? Code doesn't care about freedom. The only people who care about the licensing terms of source code are developers. Not code, not users, but people who use code: developers. Every free software license out there is for developers, period.

Why do people keep regurgitating the same senseless ideas like a bunch of blinkered zombies? There should really be a word for this behaviour. Meme amplification. memeamp. mamping. maximeme. Bleh, those suck. Someone have a better name? Like dogs howling, cats caterwauling...

Re:gah, more mindless meme amplification

[Lotunggim Ginsawat]

You can take my BSD code and make it proprietary, but you can never take my ownership of the code. Plus, my name is on your About page/screen anyway (or else I will sue for copyright infringement).

The only "freedom" the BSD license gives developers that the GPL does not is the freedom to take freedom away from other potential developers. This is so untrue I do not know what to say. Those potential developers can take my BSD code and do what they want with them (as long as my name is on About page). No freedom is lost unlike what you may think.

Re:Pale-faced android... and TPS reports.

[Anonymous Coward]

Captain Lumbergh?

Notice the Language and Framing

[Anonymous Coward]

Anybody notice the language and framing? Let me highlight the important words:

Developers are confused and divided about [the restrictions GPLv3 imposes], with fairly equal numbers agreeing with the restrictions, disagreeing with them, or thinking they will be unenforceable.

Also notice: "disagreeing with them" carries the pronoun (them) not the noun.

This language is meant to leave three concepts in your brain:

1. GPLv3 is confusing and divisive.
2. GPLv3 imposes restrictions.
3. GPLv3 is unenforceable.

Had this been another license under discussion, the terms would be very different:

1. "Confusing" would be "Complex"
2. "Restrictions" would be "Terms" and they would "apply" not be "imposed".
3. "Unenforceable" would not even appear.

Somebody is trying to say something about the GPLv3 on a lower-level channel. Are you listening? (Really listening?)

Not true

[einhverfr]

You just can't have your DRM'd hardware interfere with the execution of the GPL3 binary.

Allowing the GPL3 code to run in a sandbox where it doesn;t have access to unencrypted streams of data would not violate the GPL3 in the slightest. RMS is factually wrong about the GPL3 having any mitigating factor on DRM. Hypervisors, or even DRM built into the sound card or video card, effectively can prevent the GPL3 app from getting around it.

The GPL is for developers

[Anonymous Coward]

I'm a GPL fan, but I don't buy this user vs. developer dichotomy. Only developers care about the license terms attached to copyrighted source code; these terms mean nothing to end users.

The difference between the GPL and BSD license terms is that GPL terms protect the original developer's access to work derived from their own, BSD terms do not. GPL terms provides that you increase the size of the free software ecology, BSD terms do not. Neither license prevents the original developer from producing and distributing proprietary versions of their own code of course; but the GPL does prevent subsequent developers from doing so without negotiating terms with the original author. I use the GPL to restrict your freedom to use my software in a proprietary manner. Do I want to promote "less freedom"? Hell no. I do this because I want to encourage other developers to license their code in a similar manner. How am I, as a developer, "more free" when I cannot use the code you derive from my own? You do nothing to promote freedom when you do not freely license your work. When you do freely license your work, your work may benefit developers the world over.

The GPL is more free. The GPL is for developers.

Re:The GPL is for developers

[Lotunggim Ginsawat]

Hell no. I do this because I want to FORCE other developers to license their code in a similar manner. There, fixed it for you.

Re:The GPL is for developers

[Anonymous Coward]

Quite right. As an author of thousands of lines of code released under the GPL, I always make it a point to go to other developers houses and point a loaded gun at their head, forcing them to release their code under the GPL, too.

The reason I don't use the BSDL is because the vast majority of BSDL proponents are insufferable pricks.

Re:The GPL is for developers

[Lotunggim Ginsawat]

Man, you should have known that with GPL, derivatives will always be GPL. Care to show me how this isn't the case? If I use your thousands of lines of GPL'ed codes, it will always be GPL.

Prove me wrong, or can you?

Q on GPLv3 and anti-Tivoisation

[SiliconEntity]

There's something I've never understood about GPLv3 and its anti-Tivoisation clause which maybe some expert here can answer. Section 6 contains the relevant language, which says that keys, etc. have to be provided so that users can modify their software and still have it work. But it only applies to a "user product". Let me quote the definition:

A "User Product" is either (1) a "consumer product", which means any tangible personal property which is normally used for personal, family, or household purposes, or (2) anything designed or sold for incorporation into a dwelling.

Now it occurs to me that maybe, strange as it seems, this part of the license does not apply to software(!). The terms I quoted only apply if one of two tests are met. Either the product has to be "tangible personal property" or it has to be something for "incorporation into a dwelling". Well, "tangible" means something that can be felt with the sense of touch, which software obviously cannot. As for "incorporation into a dwelling", "incorporation" usually means a kind of mixing of substances, so this would suggest something which gets built into the walls or flooring of housing. Again, this would hardly describe a pure software product.

It might be, then, that this language only applies to things like physical appliances that you buy and bring home, or which get built into your house like plumbing. Any software which is part of these devices, as we see more and more commonly, would be affected by this language. It would only apply to devices which are designed to have their software updated periodically, but if that is the case, under GPLv3 the manufacturers would have to supply any keys necessary to make sure the device works as well with user modifications as with manufacturer-approved ones.

However, it would seem that pure software products, including in particular software designed to implement Trusted Computing or make use of the TPM chip in many PCs to condition access to network resources to only certain versions, would not be affected by this. In fact it does not seem that GPLv3 touches Trusted Computing at all, at least if the software to enable the TC features were delivered separately, in intangible form.

Has anyone ever seen discussion of this point? Thanks!

Re:Q on GPLv3 and anti-Tivoisation

[dbIII]

They put that little bit in because the implications of the draft wording would have meant that script kiddies would have as much access to the encryption keys in your router as you and the company that manufactured it - all the fun of remote firmware updates by virus writers! The change in the wording gets around this but still allows restrictions on things like those that make embedded video recorder devices. It leaves those that make routers and have made major direct contributions to things like the linux kernel alone. Linus has to live in a world where things like identity badges, passwords and encryption keys are important and recognises the importance of not making it easy for people to hack routers - RMS does not and is loudly opposed to those things and that is where this part of the licence came from.

The software improvements these people and companies make ARE available under GPLv2 - hating them for making money and changing the licence to stop them is counterproductive and in my view really brings back the stupidity of the emacs fork.

Agreed, actually

[einhverfr]

Unfortunately it is also a part of a pattern.

Ask yourself why the GFDL has an invariant sections clause. Research the answer and you will find out that it has its roots in the desire to force advocacy of the GNU project by forcibly attaching the GNU manifesto to technical documentation. Free Software == Free Speech, Mr Stallman? Or only *your* free speech?

When the then-main architect of HURD spoke out publically against this sort of forced advocacy, Mr Stallman asked him to resign from the project, saying that all GNU participants must publically support GNU policies. Is this the sort of Free Speech we are to associate with Free Software?

Waiting for GPL V4

[asscroft]

It doesn't seem fair to attack Microsoft with GPL V3 without also going after Google. Google gets to use GPL software without ever having to release source because they're not selling software, they're selling services. If Microsoft did that GPL V4 would come out faster than you can say Free Software.

The question is...

[Moraelin]

The question is, however, whether that comparison even holds any water in the first place. The whole concepts behind modern western law are some (supposedly) inalienable human rights, and how to best strike a balance to preserve them. E.g., you have a right to life, but I don't have some fundamental right to take yours, so we get laws against murder.

The GPL just isn't preserving any fundamental human right (much RMS and ESR like to rant and rave about "freedom of speech" vs "free as in beer") in any commonly recognized moral or philosophical system. I'm sorry, it's just a contract, not some fundamental human right. Asking for your code in payment for using mine in a commercial project is just a method of payment, in the end, not preserving some fundamental liberty. It's like saying that to use my digital photos, you have to send me a sixpack of beer, or that you can use my thrash can if you march around the town for at least one mile chanting "Moraelin is hung like a horse" loudly.

It's just "if you want my X, I want your Y". It's contract law, not the freakin' bill of rights.

Let's talk about "freedom of speech" too. Freedom of speech just means noone will send you to Guantanamo for saying something. The _spirit_ of it was also: something (true) about the government that it would like to keep hidden. That's its use against tyranny that those founding fathers had in mind. Noone can come round you up because you said the King did this or that, or that some other political system would be better. But let's not even go that far. Let's stick to: noone will send you to Guantanamo for saying something. For code, I guess that means publishing it.

That's all. That's "freedom of speech". You put your own code on some FTP, and the FBI or CIA didn't come kick your door in for it. No more, no less.

Pray tell, why do you need GPL to do that? And how does the GPL protect you there? Does sticking the GPL on a 'printf("The government has been supplying the same arms to terrorists that they now use against us.")' totally prevent censorship of that code, or what?

Additionally, get this: anything I ever said or wrote, including this message or whatever code I might ever release, is automatically copyrighted by me. I don't need to stick the GPL on it to be mine from the moment I wrote it. The notion that it prevents some evil corporation from taking my code and forbidding me from using it ever again, is blatantly absurd. Any evil corporation trying to apropriate my code is already violating the law as it is, if I didn't explicitly grant them a right to.

Now I may or may not have the means to fight them off. But that's an entirely different topic, and just sticking the GPL on it won't change it one bit. If I don't have the means to fight off, say, SCO for a piece of my code without GPL, I still don't have the means with GPL on it too. OK, I know, I can now also write off my code to the FSF and let them fight it out. Again, it's nothing but a bit of contract, not some fundamental freedom: I give you X if you fight off evil corp Y for me.

So basically it's all about contract, all about "but you'll give me your Y if you want my X". No more, no less. Gimme your changes, if you want my code, for example. It may fit some personal ideological crusade, but some fundamental human right it ain't.

And there it just becomes a case of "well, am I interested in getting that in exchange for my work?" No more, no less. Maybe I don't want your code at all. Maybe I just want my name in bold letters up there, because that's the kind of vain person that I am, and signing off my own code to the FSF just doesn't serve _my_ interests there at all. Maybe I want a sixpack of beer for it. Maybe I want to enforce my _own_ moral code there, not RMS's code. Or whatever.

Rights and liberties aren't that flexible or tied to one particular kind of contract. Your right to, say, criticize the government (i.e., "freedom of speech") is universal, not a consequence of slapping a GPL on a piece of speech.

Re:The question is...

[rubycodez]

no, it's not contract, it is copyright law

maybe just maybe

[sentientbrendan]

it's because the GPLv3 is *a bad license*?

I've *always* had a problem with the GPL when compared to other licenses like Apache and BSD. It seems more about advancing Richard Stallman's personal politics than about writing a good license. Just look at how long it is! It feels more like a EULA than an open source license.

The entire point of the open source is to give the user the right to use source however they wish. Stallman's doublespeak will make you think that's what the GPL is all about. It doesn't make any sense though! If I want my users to be free to use my source however I want, why do I make it impossible to use with software licensed under other licenses? There are tons of weird ass restrictions on what sort of linking you can use (none if it's straight up GPL as opposed to LGPL, but LGPL isn't much better!).

The GPL and especially the GPLv3 is like poison to most companies, even pro open source companies like google. Even the LGPL isn't used in a lot of situations because of ambiguities in the license! It has too many restrictions on how the source can be used and too much weasalease and ambiguous wording in the license. From my point of view, that doesn't increase the freedom of the user, it restricts it.

Why should large companies contribute to or use software licensed under GPLv3? It seems like suicide for the open source community to embrace a license that's going to alienate them to the fringes of hobbiests and zealots. Why would I release code under a license like that when I know it's just going to prevent people from using it?

Re:maybe just maybe

[Anonymous Coward]

If you don't like the GPL, then don't use it and STFU.

Re:That's an interesting question. Let's have at i

[Rick Genter]

Just don't think it is freedom that you're getting.

As I understand it, the GPL was never about the developer/user getting freedom; it is about the software getting freedom.

you must be new here

[sentientbrendan]

it is called open source vs free software, and that rift has been going on for a long time. It's gotten worse and worse as Stallman has tried to make the GPL more restrictive and difficult to use, and as people have analyzed the old GPLv2 more closely and noticed a bunch of ugly restrictive clauses.

Also, as others have mentioned the "any later version" statement doesn't mean that existing software is under GPLv3, but that existing software is *compatible* with GPLv3.

Really though, if Stallman could change the terms of the GPL license at whim, do you really think that would be a good idea? Copyright for an enormous amount of GPL software is already turned over to him as part of FSF rules. I have no idea why some people trust him so much with their property, but I can't say it speaks highly of their intelligence.

Not what I heard

[Anonymous Coward]

I have heard that Linus Torvalds is quite disappointed that as many people have adopted GPLv.3 as their license of choice. The thing is: in a typical Linux 'distribution', there are about 10,000 programs. Some are small applications and others are major suites. When a single piece of the distribution is licensed as GPLv.3, then people averse to distributing *ANY* GPLv.3 software (hello Novell with their Microsoft deal), are hosed for the entire distribution. Last I heard, *MORE THAN HALF* of the software on a typical distribution is GPLv.3. The GPLv.3 really gives folk (read lawyers) over at microsoft massive skull cramps. They read the license, turn red, get angry, hold chair throwing contests with Steve Ballmer, get nearly as sweaty as Steve, then re-read the license, get redder and along with it massive skull cramps. Their quick advice to the marketers over at M$: avoid this like a cross between H5N1 and ebola.

Just as a clarification

[Moraelin]

Just to clarify why I wrote the previous long rant: because redefining liberties like that is essentially one step towards losing them.

The root of all recent erosion of liberties and human rights is... people not knowing what their rights or liberties are, or what a human right or freedom is in the first place. The easiest way to take something from someone is when they don't know they had it in the first place.

Some while ago there was a poll in the USA, regarding what people think the first amendment covers. Turns out that most thought it covers everything _except_ their relation with the government. They think they have a right to call the neighbour names or to troll a privately-owned message board, an woe if anyone tries to moderate them. But the government? Nah, of course the government can censor them. Duh, that's what a government is supposed to do.

Sad.

So redefining such private crusades as "protecting freedom of speech" or such, is just muddying the waters some more. And I think we could all do without that.

You want say that corporations are evil or that all source should be open? Fine. Say it. That's freedom of speech. No more, no less.

But redefining "I can force you to show me _your_ code" as freedom of speech, is bogus. It's already claiming the non-existent right to tell someone else what to do. I can see the legality of it, and even the morality, as essentially a method of payment in a contract. Sure, go ahead, demand whatever payment you wish for your work, and the market will decide if it wants to pay your price. But some sacred human right it ain't, and freedom of speech it ain't.

And I'd rather see more people know what "freedom of speech" really is, and when their government violates it. Instead of thinking that freedom of speech violation means when <insert firewall manufacturer> stealing a bit of GPL-ed code. The latter is copyright and contract violation, and despicable in its own right, but it's not a human rights violation. And thinking it's the other way around, is the first step towards losing the real human right.

And "we have to restrict some of your real rights, to preserve some bogus 'right' we just invented/redefined" is a slippery slope in and by itself. It ranks up there with "we have to restrict some of your rights, to preserve your right to live in communism." Once you accepted that it's ok to give up some rights for RMS's personal ideological crusade, then, pray tell, why not for Marx's?

Ok, so maybe Marx is a bad example there, as it's a discredited ideology already. The same applies to any other ideological crusade, though. "We have to restrict some of your human rights, to preserve your right to live in a christian fundamentalism and stop the Islam" for example is exactly of the same calibre. Or insert any other ideology.

Simply put, equating "ideology" with "human rights" (or "freedom") and seeing nothing wrong with giving up some of the later for the sake of a bit of the former, is a dangerous frame of mind. I'd rather people understood "ok, it's a contract, I'm demanding X as payment for my Y" than get used to nodding that it's really no different to giving up some of their rights to gain a bit of ideological feel-good.

Re:Just as a clarification

[ConceptJunkie]

You know this kind of clear thinking isn't going to get you anywhere around here. This is certainly one of the most insightful posts I've read in a long time.

Oh, and don't let the cat out of the bag on Marx... a lot of people haven't caught that headline yet. ("1848 called. They want their ideas back.")

Flawed Logic & Propaganda

[Anonymous Coward]

> "Just 6 percent of developers working with open-source software have > adopted the new GNU General Public License version 3...

    May it have something to do with it being new brainiac?

> Also, two-thirds say they will not adopt GPLv3 anytime in the next
> year, and 43 percent say they will never implement the new license.

    Is that so? And do any of them know the winning numbers for the
    next lottery drawing? ..please let me know.

    What those developers say doesnt matter too much at the moment.
    What matters more is how many future projects will be GPLv3, and,
    how this amount of projects will have an effect on computing in
    general.

> Almost twice as many would be less likely to join a project that uses > GPLv3 than would be likely to join... [Evans Data's CEO

    I doubt that. There are many "cross-developers" who work on both
    BSD and GPL projects.

> said] 'Developers are confused and divided about [the restrictions
> GPLv3 imposes],

    Granted. But if youre trying to imply that the license isnt good
    because people dont understand it, just take a look at the people
    mouthing off about the BSD license --still without knowing too much
    about it either, or about patenting/licensing/copyrights in general.

> with fairly equal numbers agreeing with the
> restrictions, disagreeing with them, or thinking they will be
> unenforceable.'"

    Loaded language.

    "restrictions"

--Anonymous Coward

It doesn't dictate the hardware

[Anonymous Coward]

Just where the software can be used.

Much like US code export restrictions. It isn't dictating that the Iranian hardware cannot use the code, just that the code cannot be put on Iranian hardware.

Stop fudding yourself.

And to stop you

[Anonymous Coward]

rewording the documentation so that the only original bit is the header.

There are invariant sections defined by BSD: the license text being one and the (c) information being a second. Does this mean that you discard utterly the BSD?

Re:And to stop you

[einhverfr]

It is a question of intent.

The BSD license intends that the permission grant follow the software. I have no problem with that.

I *do* have the problem with the idea that technical documentation should be forcibly coupled with political propaganda. This is a different issue entirely.

And that's copyright for ya

[Anonymous Coward]

Try including some of MS's Shared Source code in your project. Guess what, they'll FORCE you to obey their license. The combined work MUST be released in terms that are compatible with it. Same with the GPL. The combined work MUST be released in terms that are compatible with it.

So what can cause these two very different licenses to act similarly? What common thread is there? Copyright. Copyright requires that a derived work must be released under terms compatible with all licensed parts released in that derived work.

Don't like it? Then get copyright changed.

Re:And that's copyright for ya

[Lotunggim Ginsawat]

And that's why the word 'ENCOURAGED' used by the parent poster I replied to is simply wrong. FORCE is the correct word. GPL will FORCE you to use GPL for derivative works. Are you the same AC that misleadingly used the word 'ENCOURAGED'?

Re:And that's copyright for ya

[Anonymous Coward]

No, they are not the same AC, that would be me, who correctly used the word "encouraged". The misleading polemicist here is you. If you don't like my licensing terms, don't use my code. I would encourage you do to so, but I certainly can't FORCE you to do anything. If you don't like people to use licensing terms that restrict what you can do, then stop licensing your code to people who would make it entirely unavailable. I won't allow that with my code. I demand that people who use my code give the same freedoms to others. I demand in the same sense that I demand you stay out of my house unless invited. Are you "less free" because you can't take my car whenever you like? Are you "less free" because we have laws which prohibit you from trampling over other people's rights like a selfish asshole? BSD licensing terms do not insist on giving freedoms to others; GPL "restrictions" do. The "more free" bullshit you BSD people keep spouting is nothing but sophomoric sophist crap.

Re:And that's copyright for ya

[xappax]

GPL will FORCE you to use GPL for derivative works.

I guess, but you're not forced to use my GPL code in your project. If you'd like to take the library I've worked hard on and use it to build your own program, you have to license your program as GPL. If you'd rather not license your code under GPL, you're free to write your own library, or find a different one that doesn't include that requirement.

The idea that the GPL is "restricting people's freedom" by "force" is a stretch that requires forgetting that using GPL code in the first place is an entirely voluntary decision.

Re:And that's copyright for ya

[Lotunggim Ginsawat]

You are correct, but your post further invalidate the point of the GP poster, that says GPL will encourage people to license code in similar manner. By nature of the GPL, ENCOURAGE is not a option, FORCE is more appropriate.

Re:And that's copyright for ya

[Lotunggim Ginsawat]

If you don't like my licensing terms, don't use my code. Then please explain how GPL will encourage people to license in similar manner. Encouragement should be done in voluntary manner, not by using copyright law to do so.
In the end, you misuse the word ENCOURAGE, you should have use DEMAND, as per your post above.

BSD licensing gives more freedom to developers. If I have thousands of lines of BSD code, some company like Microsoft can come in and take all my code away and make it closed source. But what most people does not realize that the copyright of the code is still mine, and not Microsoft. In fact, Microsoft has to name me somewhere in their product to satisfy the copyright requirement. I lose nothing, I still have my code (many GPL people thinks I will lose my code when it is clearly isn't the case) and other developers (2nd generation, 3rd generation, whatever) still have access to my code and can do whatever they want with it, with attribution of course. Even you can take my code, even after Microsoft has done so.

This is why GPL is less free than BSD.

Re:And that's copyright for ya

[Anonymous Coward]

Encouragement should be done in voluntary manner, not by using copyright law to do so.

Then stop copyrighting your code and licensing under the BSD - just put it in the public domain. Why are you using the BSD to restrict my freedoms?

Maybe we should abolish government too, because it restricts your "freedoms". Doing the right thing should simply be "voluntary", right? Yes? No? You'd rather avoid the topic and just keep repeating yourself?

I think you should try reading a book, instead of merely regurgitating the tripe you read on /. Maybe start with Lord of the Flies. Maybe try studying some political history. Maybe watch the excellent Ken Burns WWII special currently on PBS. Maybe someday you'll figure out that freedom isn't free; that freedom does not derive from some utopian libertarian ethos that seems to permeate the minds of ignorant bootlicking Ayn Rand sycophants.

Re:And that's copyright for ya

[xappax]

Look at it this way: If I set out to write a program, I have to decide what license I'll want it to be under. If I choose a restrictive copyright, then I'll only be able to use code from projects that are compatible with that license. If, however, I choose the GPL, then all GPL-licensed code will be available to me. Being able to use all that code serves as an incentive, or encouragement for me to license my program under the GPL - it does not force me to make that decision though.

The only time force comes into play is if I were to choose to use GPL code in my program and then refuse to license it under the GPL. In that case, I could be forced to comply with the license, yes - but that's true for any license. If I tried to use BSDL code without acknowledging the author, I would be forced to comply with that license too. So maybe in some technical, pedantic sense the GPL could conceivable end up forcing someone to do stuff - but honestly, we're talking about a legal construct. All legal constructs involve force - that's the point.

Re:And that's copyright for ya

[Lotunggim Ginsawat]

I would license my code under public domain, but there are additional efforts to do to truly makes it public domain, like paperworks to truly declare it public domain ACROSS THE PLANET. Plus, if I declare my code under public domain in one country, in another country my code will be automatically copyrighted (to me) and there is nothing I can do about it.

Please do think the world consists of America only (or the Western world).

With BSD, I can just publish my code with it and be done with it. It will be BSD across the world. Uniformity is good. With PD license, that will apply only in selected countries. With BSD, it applies across the world.

BTW, what freedom BSD license takes from you if you use my BSD code?

Re:And that's copyright for ya

[Lotunggim Ginsawat]

GPL is a fairly restrictive license, compared to Apache or BSD or MPL. Considering that the most successful open source software in the planet (Mozilla Firefox) uses dual-licensing, it does not seem that GPL is that successful.

Explain to me why GP AC says that licensing his/her code in GPL will encourage people to do the same. At best, the only reason GPL will be used if his/her code is truly godly and there are no alternative. Thus, the best way to encourage people to use GPL is to license the best quality codes under GPL. This is because, if all parameters are the same, between 2 codes that is licensed between GPL and BSD/Apache/MPL, programmers will most likely pick the latter.

Re:That's an interesting question. Let's have at i

[fyngyrz]

Did you misunderstand the context or my question that badly?

I don't think so. You asked for better wording for a license to protect the four freedoms. I responded that PD protects the four freedoms better (which is really a very direct answer to your question), and then I outlined in precisely what ways that was so. PD is very much "a license" in this context because it describes, precisely as a license does, how one may proceed with regard to distribution and use of code.

Then I pointed out that protecting the four freedoms best wasn't really what the GPL did; what it does is restrict freedom, specifically the freedom to subsume a GPL'd work inside another work without having to give up your new work, or the manner of integration. I noted that PD does not do this; so PD stands as better protecting the four freedoms, plus adding others. What PD does not do is restrict the freedom to subsume a work inside another work in any way, shape or form.

The question isn't limited to copyleft, though you might prefer it to be. Not if freedoms, even just those four, can be addressed better by other methods. And in fact, they can be. The best move to protect them is to go from the GPL to PD. If the intent, as you posed it, is to protect those four freedoms. Not to mention some other useful freedoms.

PD may indeed give the four freedoms to the first generation as does the BSD. But neither protect those freedoms for subsequent generations.

Ok, two things. First, PD protects the freedoms applied to the "first generation" (the work the original coder did) *better* than the GPL does, as I outlined. It protects them in every way just as the GPL does, then protects additional freedoms.

Second, PD does not allow the first generation coder to apply conditions to the second (and later) generation coders, protecting MORE freedoms. The GPL, at this juncture, applies restrictions, not freedoms. The original work (first generation) was a product of, let's say my work. Now I put it out there for the public to benefit. If I put it out there PD, you can add to it if you like, and release 2nd gen mods for everyone's benefit. Or not. This is a freedom; a choice. Your choice, and, I might add, your choice applied to your work. This freedom continues forward for every path that continues in PD. However, in the same situation, if I GPL it, you must give out your work if you redistribute. This is not a freedom; it is not a choice. It is a restriction. This restriction propagates endlessly with the code, like a poison pill, holding back freedom at every generation, restricting every generation. At no point, even in the first generation, can this GPL'd code be subsumed in any form into another project and then distributed in closed form. The PD version, however, can — even if you decide to keep yours closed, the first gen can still be used and forked into all manner of uses and modes and distribution, as can subsequent, more advanced generations that continue in PD through other people's hands.

Here's the core of the matter. PD says "I hand this code over, have at it, this is what I did, you can use it, improve it, even screw it up. It's a gift. A free one. No strings." GPL says "I provide this code, and you can use it. But if you distribute it, then any changes you make must also be handed out under the same terms as the ones I used. In other words, not a gift, but an exchange; I give you code; you give me compliance with my idea of how you should license your code." This is copyleft.

Now, I do not have a problem with copyleft, as you keep coming back to. None at all. I wouldn't use it, but that isn't because I have a problem with it, it is because it doesn't solve any problem that I have nearly as well as other available choices. PD completely and cleanly solves the problem when I want people to benefit the maximum amount without prejudice or c

Re:That's an interesting question. Let's have at i

[fyngyrz]

the GPL was never about the developer/user getting freedom; it is about the software getting freedom.

Well, that's an interesting idea, applying "freedom" to an inanimate object with no emotions or feelings to hurt or resources to damage that can be a building block in all manner of structures.

Funny thing, though, the GPL restricts which structures the software may participate in, or be a building block in, based on classing the behavior of those who utilize the software. So if the GPL is about giving the software freedom, I'd say it was a complete and utter failure. What it does is provides restrictions on how the software may be used. This isn't freedom for said inanimate object. This is limitation.

On the other hand, If I release software object X as PD, it can be incorporated in any structure, in any manner. If it is incorporated one way here, it may be incorporated another elsewhere. And so on, ad infinitum. If it is subsumed inside a closed source project, it still exists in unmodified form (the original creation) and can still be incorporated endlessly as released.

Nowhere in a PD release stream does the developer earlier in line get to restrict the freedoms of a developer later in line. This (if you're concerned about freedom) is a good thing. Also, nowhere does a developer later in line get to restrict the freedoms of a developer earlier in line. Again, a good thing. X is still 100% available to everyone.

Now, a closed source fellow may have produced XY, and he's not sharing. But so what? You still have X. You didn't write Y, so in what way can it be construed that you (or I) should have control over it if freedom is the issue? You might produce XZ and share it; no problem there, either. Now we have X and XZ in PD, or X in PD and XZ in GPL or BSD, and XY closed (but benefiting those users who end up with it anyway.) The guy who did X did what he wanted; the guy who did Y did what he wanted; the guy who did Z (you) did what he wanted. The world ends up with X, XY, and XZ. X is in no way compromised. Freedom all around. No one telling anyone else what to do with what they write. We simply see people specifying what to do with what they write. And no lawyers. How nice and fuzzy. :-)

Re:That's an interesting question. Let's have at i

[zotz]

You can play all the fancy word games you want.

PD allows someone in the chain to take away freedoms for derivative works should anyone in the chain desire. As such, it does not "protect the four freedoms in all situations" which was how I worded my original question.

I get your points, you seem to want to answer a question I am not asking or to convince me that protecting the four freedoms in all situations is not worth doing.

I can't think of another way to protect the four freedoms in all situations with respect to copyright than via copyleft (anyone have any other ideas that do so and not ones that don't) and as such, my specific question to the person I asked it of was if they could word a license (I assume this must be a copyleft license, but I may be wrong and would be happy to be as it would give me more options) that will protect the four freedoms in all situations.

If you want to take a shot at that, please do. If you want to convince me that copyleft is ill advised, please take it private or better yet, leave it for another day.

all the best,

drew

Flamebait!

[Anonymous Coward]

Mod parent down!!

And that is why...

[protomala]

I'm looking for a new license.
Anything simpler than GPL and that keeps my source open always.
Suggestions anyone?

Sugar CRM 5.0 is GPL3

[butlerdi]

Do not really know why, but Sugar CRN v5 is now under GPL3 ferom their old Sugar license. Reasonably large project with a large user and developer base.

Where's the FUD tag?

[Vexorian]

Really, it is taking too much time to get added.

"No error estimate was provided", I am not sure if this poll actually has happened, also could it be the 300 were mostly Novell's or other sold outs?

GPLv3 is getting quite a big quantity of new projects licensed into it, specially for a license that is less than a year old...

Too soon to judge?

[Ngarrang]

It seems a tad too soon to judge the adoption as being shunned. New licenses take time to be used by new projects and for old projects to adopt (if they were going to adopt). I mean, how old is the GPL2? That is a lot of traction to overcome in a short period of time.

You can change the licence, it says so.

[Bazar]

The legal text at the core of this mini-debate

Unfocused:

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 2 of the License, or
(at your option) any later version. Focused so that it reads out only what applies when we want to redistribute it under version 3

you can modify and redistribute it ... under the terms ... of version [3] Thats pretty clear to me.
That allows me to modify the license to version 3, removing all traces of version 2. Which would make it illegal to treat any part of that distributed code as version 2. Regardless of what it was originally distributed as.

GPL3 is Shunned, says Freedom-fearing Business rag

[PaulGaskin]

Don't be afraid of freedom! It's not so bad.

Not so fast

[einhverfr]

First, you have no authority to force downstream users to use the work under the GPLv3 only until you add copyrighted elements of your own to it. Secondly, I also don't think that it would be acceptible to represent the software as being licensed by the *author* under a different license than it has.

If you intend to change it (i.e. are starting a fork), I see no reason why it would be worth fighting over exactly where the line is to be drawn. But here you are only talking about distributing it (i.e. verbatim), and this would seem to be clearly over the line.

This is a *big* problem I have with the GPLv3 in that it requires that all meaningful dependencies outside the operating system are under licenses which allow this sort of "relicensing" by parties who have no copyright claims to the code.

Although LedgerSMB is under the GPL v2+, we depend on a lot of BSDL libraries, and since I don't believe that one can effectively remove the BSDL permissions grant from BSDL code without adding copyrights of one's own... All I can say is, have fun porting to another RDBMS.

Re:Not so fast

[TheRaven64]

The question, as I see it, is whether the later version bit means:

You may distribute this code under the terms of GPLv2, or you may distribute it under the terms of GPLv2+n or:

You may distribute this code under the terms of (GPLv2 or GPLv2+n) If it's the former, then there is no problem. You simply exercise your right to distribute it under the terms of GPLv3. This then introduces a problem, because by exercising this right you have lost the right to distribute it under the terms of GPLv3.1. If it means the latter, then you would have to distribute it under the terms of both GPLv2 and GPLv3, although it is unclear what terms a derived work could be released under. It seems that the most recent thinking (coming from the GPL and BSDL dual-licensing question) is that derived works would have to be distributed under the same terms, so moving to pure GPLv3 is not an issue.

I guess this is why the FSF asks for copyright assignment...

Re:Not so fast

[einhverfr]

The question, as I see it, is whether the later version bit means:

You may distribute this code under the terms of GPLv2, or you may distribute it under the terms of GPLv2+n or:

You may distribute this code under the terms of (GPLv2 or GPLv2+n) Or does it mean (as I maintain):

You may distribute this code as allowed by either the GPL v2 or the the GPL v2+n (at your option)?

If it's the former, then there is no problem. You simply exercise your right to distribute it under the terms of GPLv3. Does this mean "Distribute as permitted by the GPL v3?" or "Changing the license to the GPL v3, forcing all downstream recipients to ignore the author's permission grant?"

HAHA, Vista-esque adoption rate...

[slashkossucks]

What a turd the GPL3 has turned out to be... I think we can safely call it a "FAILURE", because, well, we call VISTA a failure when it gets similar results...

Re:That's an interesting question. Let's have at i

[fyngyrz]

PD allows someone in the chain to take away freedoms for derivative works should anyone in the chain desire.

I ask you again, exactly how does PD allow freedoms to be taken away? Please take a little time and explain. I really don't see how this could possibly be the case. So I'm asking you to educate me. If I disagree, I'll tell you. If I am enlightened, I'll thank you.

I think I need to change my license...

[s_p_oneil]

For all the software I've published under the BSD license, I need to change the license to say "You can't use any of this code in a GPL project". Too bad earlier versions weren't protected by that clause, but I don't suppose there's much I can do about that.

It burns me up to think that someone can take code I want to share with everyone, fix a few bugs in it, perhaps add a few minor features to it, label his additions as GPL, and then try to sue me when I get around to fixing those same bugs or adding the same features in my code. Don't even bother trying to tell me no one can try to sue me for that. You can "try" to sue anyone for anything in this country. Even if I win, it will still cost me a lot of time and money I don't have.

The point is that if I'm nice enough to share my code with everyone, I shouldn't have to worry about being sued by some asshat who's stingier about sharing than I am. (Disclaimer: I used the word "asshat" because of the frivolous lawsuit, not because he likes GPL.)

Re:I think I need to change my license...

[Anonymous Coward]

So you don't want someone else to take your code and relicense it? Why the hell did you choose the BSD license in that case? That's what the ``freedom'' of the BSD license is all about! Blaming the GPL for that is the most ironic thing I've ever heard of. The whole point of the GPL is to prevent such nonsense, after all.

Re:I think I need to change my license...

[Anonymous Coward]

Use the old BSD license, which is incompatible with the GPL.

Re:That's an interesting question. Let's have at i

[zotz]

"I ask you again, exactly how does PD allow freedoms to be taken away?"

OK, I release program A as PD (if that is indeed possible)...

You make a new program B based laregly on A and release it (C) all rights reserved.

Third person gets B from you and now does not have the four freedoms with respect to B. Was it really not clear that I was getting at this?

all the best,

drew

Actually there is

[einhverfr]

There is one case I could think of where a non-copyright holder could effectively convert the license from GPL v2 to GPL v3. This case involves a patent holder granting a patent license in accordance with the GPL v3 but not the GPL v2. The patent holder could then enforce the GPL v3 as a patent license.

FUD over GPLv3

[Snotman]

I do not think developers work on open source because of the license wrapped around the code; they do it for the code irrespective of the license. Although, I imagine there are some developers that are politically divided over types of license, I doubt most are. So, this article just seems like FUD about GPLv3.

Dan

Nope.

[mr_matticus]

Setting aside the fact that the GPL is *not* in any way a patent agreement, the patent holder would have to switch to GPL3 in order to do that, which the developers have explicitly NOT done.

If they were to do so, it would be a moot point, since that would involve them taking their work to GPL v3. Yours is a cyclical argument.

Re:That's an interesting question. Let's have at i

[fyngyrz]

OK, I release program A as PD (if that is indeed possible)...

Of course it is... what an... unexpected remark. Huh. Have you truly never run into a PD release? Here's one for you, a multiple read-client / single write-client flat file database written in python that implements a very useful subset of SQL and comes in at under 20k. In fact, zipped, the database engine, two database examples, a test / example program, and the docs come in at about 13k bytes and don't use any library that isn't part of python itself (re and os are used.) There it is, you can do anything you like with it. And I hope you do. :-) At least download it and read the docs, so you can see the terms (none) and see what it can do. So there's some actual PD software; mystery solved.

You make a new program B based laregly on A and release it (C) all rights reserved.

A still exists, unmolested, as PD at this point. So no harm or change of status has come to A at all. B is a new work, so it could have any type of release. Copyrighted, GPL'd, PD, BSD, etc. There is no harm to A in the fact that B contains A to any proportion. To the extent that B is new work (let's say it is mostly A, as in your example), that new work should be controlled by the author. A, which is not new work, is still out there and still controlled by the fact that it is PD and available to anyone.

For a very specific example, my database engine, above, is "A". You take it and incorporate it in something you write, unspecified, we'll just call it B and stipulate that B contains A, or perhaps it is more enlightening to call your aggregate production "AB", indicating some of my stuff, A, is in there. Now, my database engine, A, is still available. You wrote new stuff - I have no claim on what you wrote. That's the PD outlook. You wrote your stuff, it is yours. I chose to release A as PD, but that was my choice. Why should you be obligated to do the same? It isn't as if your incorporation of A into AB devalued A somehow. Quite the opposite. You added value and your own personal twist, now you are, and should be, responsible for the disposition of AB.

Third person gets B from you and now does not have the four freedoms with respect to B.

No, no, just wait a minute! They never had the four freedoms with respect to B; they had the four freedoms with respect to A, and they still do. I didn't give them A, I gave them B. They can still get A, presumably from the same place I got it. Or perhaps even from me. A != B, after all; doesn't hurt me a bit to hand out copies of A, no matter what my plans and intent are for B, or how much, or how little, I've changed it. So nothing has been lost, and of course, the community has gained B in whatever form the author, me, has chosen to bring it to them based on the new work. Also - in your scenario, B is "based largely on A" and so your access to A gives you - largely - what is in B. Presumably, this would allow you to create your own B without any particular trouble. So there is very little sense in the idea that A should control B in this case, even if you're just trying to take the work that B represents. But there's another side of this coin - when B is by far the larger part of the code. Then where does the justification for A controlling B come from?

It seems to me that what you are arguing for is that by writing A and releasing it as GPL, A gets to control all software that incorporates it downstream, regardless of the magnitude of additional effort or anything else for that matter.

It appears I've established that your outlook, that A should control B, is without merit other than it gets you additional free stuff you didn't have to write yourself. Which certainly has value, but as it is coercive — A controls B by force o

Re:That's an interesting question. Let's have at i

[zotz]

"Of course it is... what an... unexpected remark. Huh. Have you truly never run into a PD release?"

Of course I have, I have also seen people question whether they will hold up. Have you never seen such questions? I seem to run into them fairly often.

"No, no, just wait a minute! They never had the four freedoms with respect to B; they had the four freedoms with respect to A, and they still do. I didn't give them A, I gave them B."

Bingo, and so you gave them B without them having the four freedoms. That is my point exactly.

Now, if you want to actually try and answer the question I asked and am still asking you. Can you think of a cleaner wording for a license that will do what the GPL does / tries to do?

You do get that it is trying to bring about a world where these four freedoms exist for all users with respect to all the code they have? (I just worded that off the top of my head, so I hope I got it about right.)

"It seems to me that what you are arguing for is that by writing A and releasing it as GPL, A gets to control all software that incorporates it downstream, regardless of the magnitude of additional effort or anything else for that matter."

(minor nit - A is the first program, not the first programmer)

The first programmer can only control via the GPL license on A, what the copyright law allows. So, The second programmer, (I removed the me and you to try and ease up) wants to exert even more control over B and anything that may come from B than the first person does with A. Also using that same copyright law that A uses via the GPL.

"It appears I've established that your outlook, that A should control B, is without merit other than it gets you additional free stuff you didn't have to write yourself."

Nope, but I do agree that there is some "judo" going on. Or some hoist and petard action.

It is using copyright law to try and reverse the effects of copyright law in a fairly narrow domain.

Like I said, if you feel like trying to get copyright laws abolished so that all of our code can be Free, I may not think that the best, but I will not fight that too hard unless I learn something new.

But please, don't come back with arguments that do not relate to the real question asked.

all the best,

drew

Disagree with your position

[einhverfr]

The GPL v2 aims to include an implied patent license. Even if this is not the case, downstream distributors cannot distribute software under the GPLv2 if people are making patent claims they might enforce in ways unacceptible to that license. If a patent license is issued under the GPL v3 only on a GPL v2+ application, it becomes effectively GPL v3 only since those are the terms it can be distributed under.

It *would* give a patent holder an ability to go after people who violated the GPL v3 but not the GPL v2.

Re:Disagree with your position

[mr_matticus]

The patent holder couldn't issue a GPL v3-only "patent license" without migrating to GPL 3. Thus, they would no longer be in the "I am not adopting v3" camp.

It's not possible to fulfill both in your scenario.

If use was granted by GPL v2, that's it. It is forever available as GPL v2 in its entirety.

Re:That's an interesting question. Let's have at i

[fyngyrz]

Of course I have, I have also seen people question whether they will hold up. Have you never seen such questions?

No, never, and I've been writing software since the early 1970's. What's to hold up? Take that python database for example: I wrote it, I released it, I gave you the link, I maintain no claim on what you can do with it. I explicitly disclaim all obligations, and there was no exchange of anything from you to me that obligates me to you in any way. I kept the appropriate records when I wrote it, so as to unequivocally establish the release date — there's notarized and sealed source code in multiple bank vaults as per our IP lawyer's recommendation, just as we maintain for our commercial software. So... what's to "hold up"? How can you steal it? How can you take it "away"? How can you break an obligation to me when you don't have any? How can I break an obligation to you when I don't have any? Where's the problem?

Bingo, and so you gave them B without them having the four freedoms. That is my point exactly.

With a PD release mechanism, AB is mine to give; so that's not a valid issue unless I choose to continue AB in the same vein of release as A; that's a freedom of mine. The only difference between AB and A are the differences (B) I put in it; as you are perfectly free to have A without the B differences, you have no legitimate claim on AB (or B, which is really what you get when you get AB, as you already had A) unless I choose to offer that option.

Can you think of a cleaner wording for a license that will do what the GPL does / tries to do?

Easily. A better wording for the GPL would drop the nonsensical smoke and mirrors about "protecting freedoms" and simply say:

This license formalizes an exchange. If you use this code and distribute it in any manner, you agree to exchange the right to distribute what is provided here for (1) a commitment to distribute any new and/or derived and/or duplicated version in the same manner and (2) extend this commitment in such a way as it remains in force for each successive generation.

In the legalese equivalent, of course. Have to give the lawyers their ragged, bleeding, abused pound of flesh.

That's the entire point of it, after all. The "four freedoms" are actually diminished by this type of license. It is disingenuous to claim they are looked after as if there weren't far better ways to do so, such as PD's easily demonstrated superior abilities. The GPL should just say what it is actually doing — passing along a restriction — and leave it at that.

Copyright Holder Has the Power

[The Monster]

Is the original author bound by the terms of the license under which they release software?

The license grants permission to do some things that can't be done without the copyright holder's permission. So long as the person who has accepted the terms of the license abides by those terms, that permission is irrevocable. It's not really a 'contract', it's a specific grant of permission.

But there isn't a word in the GPL or any other license that enjoins the copyright holder from granting some other package of permissions to someone else. If I own the copyright on software, and release it under GPLv2, I can also release it under GPLv3, BSD, a commercial license that grants to my customers no redistribution rights at all, and weird stuff like "before you can use this software, you have to send a post card to this kid I know who likes to collect postcards" if I want.

Re:GPL3 is Shunned, says Freedom-fearing Business

[dosius]

GPL is free, but BSD is freer. And while I do use GPL (v2, I'll prolly never use v3 specifically on anything), I prefer to use BSD, and more often than not, I release under the 3-clause BSD license.

-uso.

Re:That's an interesting question. Let's have at i

[zotz]

"No, never, and I've been writing software since the early 1970's. What's to hold up?"

Try google. Could be something to do with no consideration and automatic copyrights, but I can't remember and I am a bit weary of this thread.

Also, perhaps the cc-licenses list.

It is not that I don't think it should be possible to put something in the public domain.

I don't think we should have this automatic copyright deal myself, but we do.

"With a PD release mechanism, AB is mine to give;"

Bingo, and so the person getting AB does not have the four freedoms with it unless you choose to give them. If someone is not ok with that for their A, I see some form of copyleft, copyright license as being one clean way to combat that.

"The only difference between AB and A are the differences (B) I put in it; as you are perfectly free to have A without the B differences, you have no legitimate claim on AB (or B, which is really what you get when you get AB, as you already had A) unless I choose to offer that option."

Sure, but since the guy writing A is making use of the same laws, write your own A if you don't like his license for his A. And don't forget where he gets this power.

"This license formalizes an exchange. If you use this code and distribute it in any manner, you agree to exchange the right to distribute what is provided here for (1) a commitment to distribute any new and/or derived and/or duplicated version in the same manner and (2) extend this commitment in such a way as it remains in force for each successive generation."

Nope.

You can use it without agreeing.
This doesn't permit private derivatives.
Could be more issues.

"In the legalese equivalent, of course. Have to give the lawyers their ragged, bleeding, abused pound of flesh."

Bingo. And you get something like the GPL out of that. Youu think the lawyers are going to make something simple? I keep asking around. No one ever comes close from what I have seen.

"That's the entire point of it, after all. The "four freedoms" are actually diminished by this type of license."

Yes, they are. For the first generation. But the goal is to reduce them as little as possible while ensuring they exist for future generations. PD gives the most to the first generation while doing nothing to ensure them for subsequent generations.

all the best,

drew

An analogy

[paladinwannabe2]

Saying that 'GPLv2 or later' code is already under the GPLv3 license is like saying BSD code is already under a GPL license. Sure, GPL coders can use BSD code and re-release under the GPL, but it's not being distributed under the terms of the GPL under someone does that. Likewise, 'GPLv2 or later' code isn't distributed under GPLv3 until someone changes the license to GPLv3.

Non issue

[ShorePiper82]

I've fixed this issue in my own coding practices. I strictly use GPLv4.