Slashdot Log In
GPL 3.0 to Penalize Google, Amazon?
Posted by
Zonk
on Fri Apr 08, 2005 05:57 PM
from the fine-print dept.
from the fine-print dept.
Michael Ferris writes "Is this the start of a shakedown by the GNU folks? Michael Singer writes that Eben Moglen and the folks rewriting the GPL are looking at a proposal where companies would be required to pay money if they use GPLed software, even if they don't redistribute the software." From the article: "The current version of the GPL, which was last updated in 1991, fails to trigger the open source license if a company alters the code, but does not distribute its software through a CD or floppy disk...the [current] rule does not apply to companies that distribute software as a service, such as Google and eBay, or even dual-license companies like Sleepycat."
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Future versions of the GPL (Score:5, Insightful)
Call me paranoid, but I wouldn't want even the faintest chance that some nasty corporation managed to litigate itself in the position of being able to release a future GPL version, as in bold below:
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.
That's handing control of the licencing of your code over to whoever is allowed to write GPLv1851, if I'm reading it correctly.Re:Future versions of the GPL (Score:5, Funny)
I'm worried about GPLv1984 myself.
Parent
Re:Future versions of the GPL (Score:4, Informative)
The proper solution is to use the named character entity references [w3.org] -- in this case ampersand + "eacute" + ; which will work regardless of the code page in use.
As refererence, I'd send you to the remarkable joelonsoftware article that explains why your approach fails to work in many cases:
The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!) [joelonsoftware.com]
Parent
Re:Future versions of the GPL (Score:5, Interesting)
He needs to be watched closely (to prevent blunders like the GFDL), but, he is well-known to have good intentions.
The problem is, if something bad happens to him, it's possible that whatever members of FSF will have the deciding say will push the GPL in a completely different direction. Whoever controls the FSF, controls the vast majority of GPLed software.
I'm not paranoid enough to label FSFians as possible traitors who would follow whoever shakes the purse, hell no -- I have quite a bit of faith in them. However, they may do any modifications to the licenses of software they don't own the copyright to -- it's a huge power. It's dangerous to leave such power in the hands of people not protected by insanity.
Parent
Re:Future versions of the GPL (Score:5, Insightful)
It's dangerous to leave such power in the hands of people not protected by insanity.
most clever coclusion I've heard in years. Funny also because of the inside pardox. But very true.
Parent
Re:Future versions of the GPL (Score:4, Insightful)
Seriously, if something like this happens, it is simply going to make more people say that people like Jonathan Schwartz and Bill Gates were right all along not just about the GPL (few people make distinctions about which version they are speaking of when it comes to the GPL) but also in regards to RMS.
I haven't written -much- under the GPL, but what I have I never intended to force the user to be required to distribute -any- change so long as they never polluted the world with bastardized
Instead of modifying the "GPL" for this purpose, create the sGPL (strict GPL). I'm sure it will be useful for people, but it inherently changes the GPL too much.
Parent
Re:Future versions of the GPL (Score:5, Interesting)
The mere notion of comparing the use of a piece of software that happens to present a more public than usual UI to the distribution of software is beyond absurd. In my opinion the fact that the vague definition of prior GPL versions' distribution clause, which could be interpreted by some to require public release of changes distributed only within a company, is also absurd. In fact, in my opinion, a lot of what RMS spews is absurd.
For example, don't get me started on GNU/Linux. What about MIT's X11? What about KDE and Gnome? Each of those probably represents a larger body of code than the FSF contributed. What about BSD? Why not just go all out and call it RedHat/Debian/MIT/Gnome/KDE/GNU/IBM/insert-the- name-of-a-hundred-other-groups-here Linux?
I'm sure some piece of code I have written to support some obscure, ancient piece of Mac hardware has, in some form or another made it into some fairly public versions of the Linux kernel somewhere (through my generous agreement that anything I wrote under a BSD license for MkLinux could be reappropriated freely for use in LinuxPPC). Can I get my name in there too? Where do you draw the line? I draw it at 'Linux'.
The point is, he isn't 'our' madman, as GP (or maybe GGP) poster put it. He's the madman for the most extremist fraction of free software developers. For those of us right on the border between choosing whether to distribute software under the GPL or another license, RMS's ranting is the biggest reason to consider anything other than the GPL, and this latest GPL proposal makes it very hard for those of us not on the lunatic fringe to take anything the FSF has to say seriously.
Just my $0.02.
Parent
Re:Future versions of the GPL (Score:5, Informative)
In fact, RMS has explicitly said [gnu.org] that a licence that doesn't allow private versions of software is non-free. The original APSL required you to publish any changes you made to the code - much stricter than the GPL which says only that what you do release, you must release under the GPL. RMS quite rightly said that this makes the original APSL non-free. You might like to read what RMS actually says before deciding that he disagrees with you.
Finally, isn't it most sensible to allow GPL version 2 or any later version *at your option*, and let the users decide whether they wish to move to the new version of the GPL when it's announced? If the new version is unreasonable, people will be free to stick with v2.
Parent
Re:Future versions of the GPL (Score:5, Informative)
But if you go on the Internet, and find the code with no version number, then you can use ''that instance of the code'' under any GPL that you like. You could use it under v1, v2, v3, vWhatever.
It sounds silly, because it's the exact same code, just from a different person, right?
But, this is actually how it works.
Consider this: If Disney were to sell you a video under the terms that you could only use it in your own domicile, then that video they give you is indeed under those terms.
But then, perhaps for a higher price, they could sell you ''the exact same video,'' but perhaps with the license option that you are ''also'' allowed to watch it at your friends' domiciles.
You could, if you wanted to, buy both! They'd be two identical copies, in terms of what you are physically holding in your hand. But the legal permissions around them would be totally different.
(I believe. I'm not a lawyer. This is just my understanding of the situation.)
Same with the GPL. If you can find a v1 version of some code somewhere on the Internet, then you are welcome to use it under the terms you found it under.
The basic idea is that: Software developers are likely (in theory) to release their new software under the most recent version of the code. (Provided they don't think it's evil, and all.) They can take all their old GPL v2 code, and automatically upgrade it to v3, without conflict. Now they are release GPL v3 code.
Now, if on some FTP server somewhere, someone finds the v2 code, and wants to use it- that's their right. They can do that.
They can use that v2 code under the v2 terms. If v3 says that you have to either publish your changes or pay someone (or whatever,) they don't have to do that: Because the code they received was licensed v2.
If they want to play by v3 rules, they can. But, they don't have to.
(Again, I'm not a lawyer. But, I also think that this is true. This is my understanding.)
Parent
Re:Stop your FUD (Score:5, Insightful)
It's all very well to trust the current board, but when you say "any future license", you are trusting all future boards, and 20 years from now, every single member will be someone you don't now know.
Parent
Re:Future versions of the GPL (Score:5, Interesting)
Here's the full section 9, a portion of which you quoted:
9. The Free Software Foundation may publish revised and/or new versions of the General
Public License from time to time. Such new versions will be similar in spirit to the
present version, but may differ in detail to address new problems or concerns.
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.
So my question becomes this: What determines whether or not you specified a specific version of the GPL? Most of my comments and the readme file say "Licensed under the GNU General Public License" (no version mentioned), but then included with the distribution is a copy of version 2 of the license. Does that imply strongly enough that version 2 is the specific license under which the software is distributed?
Or do I need to go make some changes and do a commit....
Parent
Re:Future versions of the GPL (Score:4, Funny)
Developer: It doesn't have a version number, so it refers to the earliest version
Business: It doesn't have a version number, so it must be the most recent
Reality: It has no version number, so it would apply to the current version upon creation of the software
Legal: Pay me lots o' cash, and I'll make something up and back it in court. *Crosses-fingers*
Parent
Re:Future versions of the GPL (Score:4, Insightful)
IE... the first version...
Did the first version have a version number in anticipation of future versions?
Parent
Re:Future versions of the GPL (Score:5, Insightful)
Parent
How is this legally possible? (Score:4, Insightful)
For this to work as described, the user would have had to sign a contract. Otherwise, it will be just as unenforceable as a EULA (you already had a right to click the Next button on your own computer, etc).
Parent
They deliver HTML. (Score:5, Insightful)
Google and eBay distribute HTML. That HTML is created by software that uses GPL code. So if I modify a GPL Office Suite, would I have to distribute the code if I email someone a document I made with it? Seems like a bad idea, in general.
I guess people could fork the GPL2.0'd code if the software developers switched to GPL 3.0
Re:They deliver HTML. (Score:5, Insightful)
Hmm. I could swear that the Google appliance in my rack, and the Google toolbar on my desktop weren't just hunks of HTML.
Parent
Re:They deliver HTML. (Score:5, Interesting)
Just because they use Linux in the Googleplex doesnt mean they use it everywhere.
Parent
The GPL must change or die. (Score:5, Insightful)
That's not the same as you distributing a document you created. In that case, INPUT occurs on your computer, PROCESSING occurs on your computer, and OUTPUT occurs on your computer.
The question is: What's important in determining if a program has been "distributed"? Clearly if I give you a CD with the program and you run it on your computer, I've distributed the program. Clearly if I take GPL software and modify it and I run it on my computer only, I have not distributed the program.
But if I take GPL'd software, modify it, and then let YOU run it on my computer...
Did I distribute it, or not?
Your immediate answer is probably "Who cares?" But now what if I charge you to run this modified program on my computer, for example, by charging a fee if you use my auction program? Now I'm using software provided by the open source community for my financial gain, but not returning the modifications I made to that software to the community.
The extreme of this problem is that eventually, the internet becomes so fast and clients become so dumb that software is never "distributed" at all. I take an open source office suite and then modify it. According to the GPL, if I then sell that software on CD, or by download, so that people can actually run it on their computers, I must provide the source to it as well. But what if I'd rather just make money off of the GPL'd software I've taken without giving anything back to the community?
Well, then I just put the software on my own server, and instead of selling CDs or downloads, I let people provide input and receive output over a remote connection to the program running on my server.
And wala! People can modify and essentially provide GPL'd software without having to provide source.
Parent
This doesn't sound like a good idea (Score:5, Insightful)
hails back to the days of the old BSD style licenses. The GPL works
so well now, precisely because it is unobtrusive. My company runs
GPL'd software because we are able to use it and make modification
without either redistributing the source code and we aren't required
to pay for that ability.
Requiring people to open all their changes or pay for them will put a
lot of businesses off when it comes to dealing with GPL'd software. I
don't think that is a good trade off to make, and I don't think it
will be healthy for the open source community in general.
A move like this will make the newer BSD style licenses and / or
licenses like the Python license much more attractive imo. Now that
open source is finally turning the corner, and solid technologies are
finally moving into the enterprise, why would we even entertain making
changes that will certainly hamper open source adoption?
This isn't a consistent position in my opinion. If you are
developing free software, it should remain free. The idea that you
can get someone hooked on software, and then pull the rug out from
under them and start charging them is ludicrous. If this were to
happen, I can honestly see a major fork in the GPL happening.
Re:This doesn't sound like a good idea (Score:5, Insightful)
If you're not distributing your work, there's no reason why you should be forced to open your code or pay some silly fee.
I hope this whole story is a troll...I really do.
Parent
Re:This doesn't sound like a good idea (Score:5, Interesting)
These implications of the proposed GPL3 are certainly troubling. How is this different from "evil commercial vendor lock-in"?
I'm not trolling...just hoping that this interpretation of GPL 3 is wrong.
Parent
FSF? (Score:3, Funny)
Re:FSF? (Score:4, Funny)
Parent
Bad, bad, bad! (Score:4, Insightful)
Re:Bad, bad, bad! (Score:4, Insightful)
How much source code have they released ?
Parent
Re:Bad, bad, bad! (Score:4, Informative)
Parent
Who what when where? (Score:3, Insightful)
I think this proposal is crazy. If you use software as a part of running your business, that software is benefiting you and indirectly providing services to your customers, even if they never see it. So where do you draw the line?
Re:Who what when where? (Score:5, Interesting)
BTW, when does Stallman's Macarthur Foundation Grant expire?
For the humor impaired: It's a joke son.
Parent
Re:Who what when where? (Score:5, Funny)
The wrapper code could then count the number of calls, the function name called, and the company calling them, and this in turn could append a text file that is stored somewhere and automatically sent once a month to the FSF which will then use the file to send the company a bill. The FSF would then distrubute the money to each of the authors of the given API calls.
I will volunteer to make the necessary modifications to gcc. This of course will require all GPL'd code to call my APIs, which will of course be the first to implement this new system.
Parent
Let's get more detailed (Score:5, Insightful)
Im speechless..... (Score:5, Insightful)
"If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."
Basically, in any other language: 'Now youve had time to build a good infrastructure on the current rules, prepare to be shafted'. If this comes to pass, then in my mind they are no better than Microsoft changing EULA terms with a service pack. Now that there are major companies with an infrastructure built on GPL software, the FSF are looking to essentially move the goalposts and if this is applied retroactively to current code (which from the articles wording I think it will be) then I personally think that its going to do more harm to the GPL community than benefit it in code donations as companies scramble to move away.
Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.
Re:Im speechless..... (Score:5, Informative)
Please someone tell me that they cant do this retroactively, that its impossible under the current GPLv2 terms.
Your wish is my command. If you look at source code to any GPL-licensed program, you will see something like:
Hence any code that was licensed under the GPL before version 3 is released may still be used under the terms of the GPL version 2, at the option of the recipient, not the author. Actually, a number of current projects, including Linux, are licensed GPL-2 only and may be impossible ever to convert to a higher version (permission would be required from too many people to reasonably track down). Hope that helps.
Parent
Horrible implications. (Score:5, Insightful)
Regardless of what GPL 3.0 turns out to be, developers are not forced to use it. They can continue to use GPL 2 if they wish, just as they can choose to use a BSD license, Apache license, creative commons, or any other license of their choosing. Furthermore, software that has already been released under GPL 2 cannot be retracted, it remains available under GPL 2 forever.
Submitter mischaracterises the change. (Score:5, Informative)
Re:Submitter mischaracterises the change. (Score:5, Insightful)
This is deeply irresponsible. Any project that ships software under the GPL is going to be spinning their wheels for months over this, and the Microsofts of the world just got a huge weapon to use against OSS usage. After all, now they can say that using GPL software not only costs you in terms of the usual TCO metrics, but there's a potentially hidden and as-yet-unknown cost that can be applied to retroactively!
Grrr!
Parent
Re:Submitter mischaracterises the change. (Score:5, Informative)
Ditto if they attempt to murder strangers in public. Fortunately, neither will happen. If you are using a product under the GPL 2.0 license, then it is yours to use under the GPL 2.0 license.
If someone releases updates to a GPL 2.0 licensed product, but those updates are only available under the GPL 3.0, and you do not want to abide by the restrictions in the GPL 3.0, then you might be SOL for those updates. That's about the worst thing that can happen.
So chill out.
Parent
Where did this mindset come from? (Score:5, Insightful)
Re:Where did this mindset come from? (Score:5, Insightful)
Like voting? And do corporations get the same negative sides as private citizens, like going to jail? If you won't send a company to jail, and give it a vote, you can't equate them to private citizens
Parent
Re:Where did this mindset come from? (Score:5, Funny)
> There code , there license
There cave, there bear, Og go cave, Og kill bear, Og eat bear, Og rest, Og write GPL 3.0 !
Parent
Call it FUD (Score:5, Insightful)
The problem is that it infringes on one of the four freedoms, specifically the freedom to use. If such a provision were to find its way into any license that made it so that companies and individuals were not subject to the same terms, the license would be both discriminatory and non-free.
Simply put, this is somebody making FUD about the GPL. Don't buy into it for a second.
It's not in print yet... it doesn't exist (Score:5, Insightful)
At least, that's what I heard anyway.
When whatever changes come up, they will be reviewed and we can rant and rave about it at that time.
Nothing is founded, no concrete written agendas were tossed out by the article and all we have is a hypothetical situation that would be very different from the current model.
Like many have said and will say, it's usually a good idea to specify the version of the GPL in which you release software. Unless you really don't care what kind of changes are made at a later date.
I'm not saying all of this isn't going to come true, but at this juncture we could very well be required to throw pies to comply with the next redistribution agreement.
Don't get Dramatic (Score:4, Insightful)
And frankly, it's not really a loophole. Web services are not distributing software, they're running a service using software. That's obviously open to interpretation, but I haven't ever heard anyone distribute under the GPL and complain about someone using their software as a web service. There has been at least one derivative license which has addressed this issue.
In the end, GPL 3.0 will likely provide an optional provision which will 'trigger' GPL source distribution requirements for a web service, at the option of the copyright holder; that is really the best choice. Rather than getting into an enormous philosophical debate over whether the idea is "good" or "bad" or "punitive" or whatever, let's simply have two clear licenses and give the option to the copyright holders to decide under what terms they will license their property.
Software for the Rest of Us (Score:5, Insightful)
This will stop many programmers, many of the best programmers, from using source code under that license. We could no longer keep any of the value of the software we created to ourselves. What is a "web service"? Is my email-processing CGI a "web service"? Any software in the same workflow as any other software under this license would have to be released. So many developers won't make small customizations, because that would force us into the source code distribution business, with all its overhead. Or we might just ignore that provision, or the whole license, en masse.
The GPL is successful because it is a fair contract, even though it's revolutionary. Its enforcement teeth are rarely tested, because it's so close to an equitable agreement among peers. Which has resulted in lots of value contributed by profit-driven organizations, despite the claims of many that the license is anticapitalist. Upping the ante, to require private customizations to be published, could stop the rising tide of acceptance that is pushing GPL to be the default, and any proprietary license to be radical. And then the caution it would inspire: investing in GPL'd software might force acceptance of ever-more demanding licenses, like a GPL4.0 that required redistribution of even software that wasn't changed at all, just to get users "to pull their weight".
The GPL2.0 isn't broken. Let's not "fix it" in a way that could destroy its success, and our chances to benefit from one another's work without onerous burdens.
GPL 3.0 and "servevices" (Score:4, Insightful)
Currently the GPL forbids redistribution of GPL-covered code linked with other code to form a larger work, when that other code is not also distributed under the GPL: the whole larger work must be so distributed. It is a "derived work", and copyright law is clear on this: one can not redistribute derivatives of copyright works without permission. It's that permission that the GPL grants, when you abide by its terms.
Of course, you are not prevented from redistributing GPL code *aggregated* with non-GPL code.
The problem arises from distinguishing between derived and aggregated works: what if I distribute a GPL app that I write, with a bunch of GPL shared libraries that I didn't write (complete with source), and a nonh-GPL proprietary library, essential for the application, that I did write, without which the application is useless.
Is that an aggregate, or a derived work of all the GPL code I did not write?
On the one hand, there's no (legal) requirement that my app even run, and, since I distribute source, one can implement a version of the non-GPL library. The fact that it works when my non-GPL library is installed in the right directory (LD_LIBRARY_PATH, anyone?) is a happy coincidence. Redistributing my GPL app, with modifications is fine, but you can't redistribute my "essential" non-GPL library, rendering the app rather useless without it. You add value to the app, I license more copies of the essential library.
On the other hand, the proprietary library, the foreign GPL libraries, and my app together, constitute a derived work of the GPL libraries, and must be redistributed en masse under the GPL.
If the libraries are statically linked into a monolithic executable, there is little argument that the whole source must be redistributed. The situation is more controversial if a mere aggregate of files is distributed: some argue that not distributing the proprietary library under the GPL is a violation, other's don't.
I tend to believe that if the GPL-covered peices can be redistributed indivudually, without the proprietary library, regardless of whether the result is useless, there is no violation: the fact that the code is actually only useful when the proprietary library is present is but a happy coincidence. What if I distributed a GPL version of the library, and offered a proprietary replacement, with far better performance?
This controvery gets even muddier when one considers alternate ways of effecting program linkage. In effect, the "functional derived work" exists only at run-time, and, indeed, the maner in which the parts of the aggregate are combined, can, itself, be the subject of restrive licencing, and patent. What about linkage via a remote procedure call mechanism? Y'all remember RPC/XDR over TCP/IP, right? Suddenly, self-assembling functional derived works become a reality. Protocols like SOAP, used to support "Web Services" exacerbate the problem.
This leaves a big, gaping hole in the GPL: socket wrappers were a common "hack" to "get around" the GPL: just layer a RPC mechansism around the proprietary library, and a GPL wrapper to call it from the app, and you were all set. And, that didn't even address the issue of inter-machine communication: 127.0.0.1 and Unix sockets are ubiquitous.
Of course, the minute one's app "links" with proprietary code on a different server in this manner, the GPL loses all force, for one did non distribute the proprietary part.
It is this area that the GPL v. 3.0 tries to address, IMHO, perhaps by more differentiating between linkage and aggregation. This can be done, of course, but then the license starts to lose some of its roots in copyright law.
The problem with such an approach, though, is, what constitutes linkage: does an exchange of HTTP requests and responses? What if there is ultimately GPL code satisfying that request? Surely, if I use a program provided by a vendor
GPL is a distribution license not a usage license. (Score:5, Interesting)
Changing from a distribution license to a usage license is a VERY significant change in the spirit of the license.
Quote Last para, Term 0
The act of running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the Program
Term 9
Such new versions will be similar in spirit to the present version,
From whence doth your vision stream, Olson? (Score:5, Interesting)
Perhaps he's just managed to read the Affero General Public License v1 [affero.org] and has decided that that's the way that the GPL v3 is going to look? But apparently he hasn't already read the coverage of this rather crappy license that debian-legal gave in 2003 [debian.org] and then informed the FSF (and RMS), explaining that it couldn't possibly be DFSG Free, [debian.org] let alone satisfy the 4 freedoms?
Oh, right. Must not have actually checked all that out. Gee, does Mike Olson even use the GPL at all? Why would he be reviewing it anyway? Well, lets see: hrm... this sure looks like the 3 clause BSD license to me [sleepycat.com]. Yerp. No GPL in sight at all. Ok, so someone who doesn't even use the GPL, (to my knowledge) isn't a lawyer, and isn't a prominent member of the copyleft side of the Free Software movement is reviewing a license that no one else has seen?
I mean, I can understand slashdot editors missing this bit of trivia in their rush to approve/reject a story... but surely Michael Singer at internetnews would have bothered to actually check if Mike Olson was the "insider" he was claiming himself to be?
This story is complete BS (Score:5, Interesting)
First of all the story purports to tell what Eben Moglen is doing but mostly includes quotes from that Olson fellow. The obnly Moglen quote basicly says "let me finish the draft first and then we can discuss it". Then the article continues on pure speculation of what Moglen might be thinking.
Second, the slashdot blurb completely misrepresents the actual referenced article. The shashdot blurb makes it look like the main purpose of the FSF is to hit up google, yahoo, etc. for money. In reality the whole damn controversy is whether the code should be released or not.
Of course if there ever emerges a legal requirement to release the code and the code is not released the FSF might be entitled to damages -- thats just the way the US judicial system works. But that does not mean that the FSF are looking to hit up these companies for money. In every dispute so far the FSF has made it quite clear that they will forgo money damages when the code is actually released.
Now whether GPL derived code used for providing internet services should be released is an very interesting (and increasingly important) issue. However, this slashdot article completely misses the whole issue by making it look like the FSF is just trying to get rich on Google's behalf.
Re:Sensationalism at its finest (Score:3, Insightful)
"If you look at the market, Yahoo, eBay, IBM, Amazon, Google have all sunk millions into the GPL infrastructure," Olson said. "Not only are we changing the rules, we are changing them retroactively. With the new way, it lets the customer pay with either their source code or with their wallet."
Re:Pot meet kettle (Score:4, Interesting)
My god, I hate it when HP,MS or whomever does this. Now the GPL goes this way, UGH. It's not the fact that they are changing the GPL, that's bad enough but the fact they are retroactivly changing is what makes it so bad. This is the kind of shit the the bad boys do.
"They" can't change the license retroactively. Any software licensed under GPL is either "version 2 only" or "version 2 or later, at your option". So, any GPL software that exists at the moment the GPL-3 is unveiled, you may continue to use as long as you want under GPL-2 terms. Of course, this will not be true of new or updated software released afterward under "version 3 only" or "version 3 or later" terms.
I hope this idea goes no where or you can see any the profesional devs go elsewhere really quickly.
If the GPL-3 turns out to be as implied in the article, I completely agree with you. And I think that most free software authors will as well. If the terms of GPL-3 are that bad, no one will use it and it will quickly become irrelevant - nothing to worry about.
But let's wait until we actually see a draft of a proposed GPL-3, not get upset over third-hand rumors.
Parent