Slashdot Mirror
Moglen's Plans to Upgrade the GPL
Nick Irelan writes "Although it most certainly won't be easy, Eben Moglen is attempting to upgrade the GPL. He sees an opportunity to create a version of the GPL that will be able to adequately suit the needs of modern programmers. If they are implemented, his ideas will be the first major change the GPL has experienced since Richard Stallman wrote the original version. Eweek has an amazing article about Moglen's work. Linus Torvalds discussed what he believes should happen to the GPL with Eweek as well."
No need to release it.
Someone set us up the bomb, so shine we are!
no it doesnt....
GPL = General Public License (GNU)
and
GNU = Gnu's Not Unix
liqbase
From the eWeek article:
"Linus Torvalds, the creator of the Linux operating system"
Not a good start for this process...
Mod eWeek -50 Flamebait
Based on the wording of the GPL, you cannot release a GPL program under an old lisence. It states that the program is licensed under either the included version, or any subsequent version thereof, at the discretion of whoever is going to be changing it. The GPL is also "unmodifiable" (Something I personally don't like) I assume, technically, this also forces derivitaves of someone who chose to use GPL3 on a GPL2'd project would forever be locked in GPL3
I hate grammar Nazi's.
Hey, maybe you can use LGPL libs instead, because it was the purpose of the creation of the LGPL - allow the usage of GNOME/GTK+ interface in the prioritary apps. For example, almost all GNOME libs are LGPL now.
user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
*feeds the troll*
If you use GPL, FSF can arbitrarily change the GPL to anything they want at anytime.
From the GPL:
9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time... 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.
(IANAL, just an interested geek.)
Depends on how the developers licensed their work. Most GPLed software is licensed under 'GPLv2 or later' which means that you may chose between GPLv2 or a later version.
The Linux kernel is an exception. Linux Torvalds licensed it only under GPLv2 and he removed the 'or later' part. So the Linux kernel won't be (automagically) GPLv3, as intended by Torvalds. From my understanding he doesn't trust Stallman / FSF on this one. If the Linux people want GPLv3 then they'll have to do major efforts contacting all these people who contributed to the kernel, i suppose (you also stated this but it only applies on the GPLv2-only software which is a minority!).
To sum it up: almost all GPLed software will be compatible with GPLv3 and users may use the GPLv3 license instead. Also, developers may chose to license their newer code under 'GPLv2 or later' 'GPLv3 or later' or 'GPLv3 or earlier' or even 'GPLv3-only'.
WE DON'T NEED NO BLOG CONTROL.
Linking against a GPL library (e.g. cygwin) requires the result to be GPL'd.
An LGPL'd library (e.g. libc) can be used by non GPL'd software so long as you provide the ability to upgrade the LGPL'd library (dynamically linking satisfies this condition, as does providing object files and a link script).
Both of the above assume that copyright actually applies (if, e.g., your work isn't legally a derived work of the GPL'd or LGPL'd code then things are rather different).
rage, rage against the dying of the light
Following my own link and reading a bit more it seems that Apple has already addressed issues like Web Services and even Patent litigation against the licensor (Apple).
Regarding Services (Major Changes in Apple Public Source License 2.0):
1. Licensees will only be required to release source code of Modifications they "Externally Deploy" (new Section 1.4, and Sections 2.1, 2.2). "External Deployment" is defined to cover the external distribution of APSL'ed code or use of APSL'ed code to provide a service (including content delivery) to a third party through electronic communication with that party.
Regarding Patent Litigation:
5. The Termination clause relating to patent suits (Section 12.1(c)) has been narrowed such that the license will terminate only if a licensee _initiates_ an action for patent infringement against Apple. It will not terminate in cases where Apple first sues the licensee and they file a countersuit.
So #5 seems to cover litigious bastards such as the SCO Group, except only for patent litigation. It'd be interesting if at some point this was updated to include copyright infringement litigation as well. And #1 seems to cover Web Services. Maybe Mr. Moglen will reference the APSL in revising the GPL.
--
Join the Pyramid - Free Mini Mac | Free Flat Screens
infested with jello like fishes no melotron wishes
Athletic Scholarships to universities make as much sense as academic scholarships to sports teams.
Has any company ever been forced by a lawsuit to open-source software it distributed because it used GPL'ed code?
No. And there likely never will be. It's the same as your latter statement, if they don't follow the GPL, they have violated copyright law, and could be held liable for damages in a civil suit. Any other remedy would be an extreme and new precedent.
I've had enough abrasive sigs. Kittens are cute and fuzzy.
As far as I know nobody has ever been forced to do this, nor is it likely they would. What's more likely is that they'd be barred from distributing and would need to re-write the product without the GPL components. Most companies who base something heavily on the GPL (like embedded linux on devices) cave without a lawsuit.
The GPL doesn't work that way.
See, you don't have to agree to the GPL unless you do something which would be violate copyright if "All Rights Reserved" were the license.
Downloading and running a program doesn't violate copyright even if all rights were reserved, assuming the person you are getting it from is duly authorized to distribute it to you (lets assume they are).
Using the program internally and modifying it also fall outside the realm of copyright law.
Only distribution triggers copyright law restrictions, and only then does the GPL apply.
So even if they wanted to, the GPL couldn't make the requirement to submit internally used changes back to the maintainer, without putting it into the weaker legal realm of a "click through" agreement, like an MS EULA, which is probably not enforcable except in UCITA states.
I've had enough abrasive sigs. Kittens are cute and fuzzy.
A good discussion of this appears here. One point made is that the headers to the library may be covered by the GPL, and those headers *are* part of the application code. There was some discussion about releasing the *headers* as public domain, which may release any linked code from the GPL, but at that point you'd be better off releasing the library under LGPL.
"How could monetary damages be computed?"
Well, luckily (sarcasm) they don't have to be computed. The big boys have gotten laws passed that include statutory damages. So now their pet laws are causing them trouble? Oops.
A Nony Mouse
Bad example:(source)
GNU Readline is the canonical example of a GPL'd library that makes no exceptions for other free software licenses.
Several companies have released code under the GPL after being threatened with a lawsuit. But no Judge has ever stood up and said "Abide by the license! Release the source, or go to jail!" - to the best of my knowledge, no lawsuit has ever gotten that far before the parties choose to settle.
You are not alone. This is not normal. None of this is normal.
In the United Sates, copyright violation has _laws_ describing monetary damages per infringement. Those are on the order of thousands to hundreds of thousands in statutory damage (which are essentially punishment and are not connected to "profit" or "lost business")
check out section 504(c) for clarification here. Notice it also has provisions based on "lost business", but they are seperate from the statutory ones.
If, for example, Microsoft used a nifty feature copied directly from the HURD (hahaha) in longhorn and was found out two years and 10 million copies later they could easily owe billions of dollars in restitution.
But that's not what's happening. If I use a GPLed library in an application, unmodified, then in what way am I denying people access to the code of that library by not GPLing the rest of the application?
You aren't. But you are denying them access to the code of your own application, which is something you promised to supply at the time you used GPL code in your project. (If you didn't intend to make that promise, then you were simply breaking copyright law)
The current situation is this: You can start with some software that is available by the terms of GPL, and create a derived work. But instead of redistributing it to others, you can just make your derived work available as a service (for example, a web app). You don't distribute the software -- it always runs on your computer, but simultaneously is used by other people.
The question this raises is this: in what way are you then bound to the terms of the GPL?
Since you're not distributing the software in any form (a binary or source or whatever) a lot of the GPL terms don't get triggered. (Some might even argue that you're not bound by the license at all.) You don't have to supply the source to the program to anyone, even though a lot of people may be using it.
The result is that users still end up becoming "hostage" to a single entity, with no way to alter or maintain the software that they use -- even though the software is technically GPLed and the main goal of GPL is to prevent that very thing from happening! Oops.
Thus it's likely GPL3 will have some provision that addresses this situation. The terms will have to be triggered by derived work creation rather than distribution.
The catch here, is that this is sort of idealogical poison. A lot of Free Software people don't like the idea that mere creation of a derived work that you never release, is a violation of copyright which can only be made legal by licensing.
For example, if a programmer creates a patch to some proprietary software, and only distributes the patch (not the original software itself) does the programmer need some kind of icense from the copyright holders of the original software? Ick. There's some tricky and murky and unpleasant stuff going on here.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
no, because it says v2 or any subsequent version. If they did make a hypothetical, evil GPLv99 you could quite happily keep on using your code under v2. You wouldn't have to take it. As a result of this, you can't make the GPL any less free than it already is.
"Success is based on knowing how far to go in going too far"
It's not a contradiction, but a misinterpretation.
The GPL version "2.2" is the DEVELOPMENT VERSION, which Prof. Molgen submitted to RMS for debate. To my knowledge, no one is using it (I haven't even SEEN it yet). The latest PRODUCTION VERSION is version 2, dated in 1991, the version that almost all GPL software is currently licensed under.
... And so it comes to this.
The part that you're citing is in the preamble. The preamble is not controlling - that is to say, it's not actually part of the license. The actual license starts after it says, "The precise terms and conditions for copying, distribution and modification follow."
The only part of the GPL itself that mentions patents is section 7 (note that if you can't cite a section number, what you're quoting is not part of the license itself). Section 7 merely tells you that if a patent license or court judgement makes it impossible to distribute under the terms of the GPL, then you cannot distribute at all. This covers the case of the IBM patent grant just fine - code using those patents can be distributed under the terms of the GPL, and thus you can continue to distribute the entire work under the terms of the GPL, so you're fine.
We ought to have discussion about the GNU General Public License (GPL) v3. The GPLv2 is an important license, the most widely used free software license. We should have critical discussions to help make the GPLv3 better, and of course defining "better" requires understanding the goals of the license.
But there's a profound unfairness in the two articles linked to here. They are filed in the "Linux & Open Source" section on the eWeek website, and not by accident. The GPL was initially written well before either the Linux kernel or the open source movement began and it was written to serve the purpose of furthering software freedom (an issue the open source movement does not want to talk about because it gets in the way of making their pitch to business, this movement's main audience, on "solid pragmatic grounds rather than ideological tub-thumping", as their FAQ says. This name-calling is starkly less insightful than the analysis the Free Software Foundation offers about the open source movement). So, there is simple miscrediting going on here, but it's also ironic that is no "GNU/Linux & Free Software" section at this website. Such a section would be far more accurate for describing stories about the most widely used and most important free software license.
When version 3 of the GNU GPL is released, it will be the first version to come out that had a chance of being edited by someone involved in the open source movement. As far as I can tell, nobody from the open source movement has had a hand in revising any version of the GPL. The GPL was written by people from the FSF (and the listed author is the FSF). Yet the GPL is routinely cited as an open source license by proponents of that movement, essentially taking credit for work that nobody in that movement did.
The Linux kernel is but one program in a complete GNU/Linux system. It's ironic that this license is so pivotal to the development of the GNU/Linux OS but GNU can't get just a share of the credit.
Of the two men featured in articles which are linked to in this Slashdot thread, one is an authority on the GPL and a co-author of the GPL, the other is someone who exhibits no significant insight into how the free software community came to be or what the GPL is here to accomplish. I'm grateful that Linus Torvalds began the Linux kernel and continues to work on the most widely used fork of that kernel, but this is not about the technical inner workings of the Linux kernel, where Linus Torvalds is unquestionably an authority on the matter. Torvalds is no authority on the GPL or software freedom in general. If you point your friends to these two articles, please don't give Moglen and Torvalds equal billing here. Equal billing would either diminish the attention we should pay to Moglen's comments on this matter or give Torvald's comments more attention than he deserves on this topic.
Digital Citizen