The GPL has always quite explicitly stated that the license changes as the FSF sees fit. Section 9:
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.
I don't know they could have been more explicit that the GPL is the GNU Public License where GNU is project under the guidance of the Free Software Foundation.
Under your scenario the far cheaper thing to do is buy a commercial license for that one GPLv3 piece.
But even if they don't, I'm hard pressed to see the downside here. The company goes all commercial and starts heavy licensing fees which they have to pass on, which make their products more expensive than companies that are less open. The board is then confronted with a realistic type cost:
say $3k per developer per year, $1m in backend license fee support costs and $12 / device. If they think staying closed is worth that much, OK.
The reason companies paid developers to work on the Linux kernel was because they wanted to get out of writing their own kernels. It was from good will. And they hated the GPL even while they did enormous things to make it clear that large scale cooperation worked best under the GPL.
Which is the same argument were were having 15 years ago for GPL2. Companies didn't like those terms much either. They didn't have a choice as more and more open source software was GPLed. Apple would have loved when they were building the original samba extension in if Samba had been BSD and they could have created "iMacWin Share" as a sellable product.
The second is patents. That's a serious problem, but the best solution is patent reform - not 'I'm taking my ball and going home'. I guess this could work, but depriving Mac users of Samba over unrelated patent issues (I assume Apple has no patents relevant to Samba) is getting beyond the goal of 'keeping the code free'. Software patents are mostly an abomination, but GPL3 is an awfully blunt instrument to attack them with.
GPL2 was very blunt instrument to attack the semi interoperability as a result of the Unix wars. It worked and everyone is better off.
Not always but almost always. He's got a pretty good track record.
Some of his architectural choices haven't proven so wise, his priorities were less than ideal. But damn he's got 3 decades of some decisions that worked out very well.
I would argue that Version 3 is a violation of copyright as it prevents corporate contributors from having access to code that they contributed to.
Authors aren't bound by licenses including their own. They have slightly more freedom than they would have under the BSD type scenario. If you are an author who released code under GPL2 you can turn around tomorrow and release it under BSD, MIT, PD...
As for being anti-commercial, read back the discussion in the late 1990s early 2000s regarding GPL2 vs. BSD. Same arguments. Yes companies like maximum freedom. Too bad for them.
No you can't because a slavery contract cannot be enforced. If you sell yourself to me as a slave day 1, you can leave day 2. If I try and stop you physically that's a battery, kidnapping... The most you can do is sell your labor in advance under guidelines of US labor law which prevent you from selling it too far in advance.
Even things like no compete clauses the courts have interpreted very strictly.
No it doesn't impact my software. It just impacts the version of my software that all the end users actually run and can run. See the history of X for details.
The system apple has for iOS for code signing is already great. A provisioning file requires a crytographic key involving the device and the signature of the application and the signature of the person sending it. The person installing the application needs to first copy the provisioning file into iTunes. Just make that private key public and you have open system.
So even if I infect myself how do infect others. I don't have the Samba Team's private key so I would need to use another private key that wouldn't match the supposed author.
Then they are going to have a real problem. Think about how complex the software is that runs on macs vs. what runs on tablets.
Lock it down and they kill the platform. People will accept their phones being limited because they always have been (RIM being something of an exception). They may accept tablets being limited but already their is a bit of a backlash. No way will it be accepted on computers.
That wouldn't present a problem. If the Samba team, or someone not Apple, released Samba for OSX via. the Apple Store as a free application then Apple doesn't pick up a GPL3 patent issue in the same way a book store doesn't pick up a copyright problem for a book in good faith that they sell.
All of the GPL is political. Always has been. The GPL3 is about plugging holes in GPL2. Patents can and have been used to effectively limit freedoms under GPL2. Ergo...
Actually at this point Microsoft seems more free than Apple. Apple was never a huge fan of open source, they were desperate. Generally companies in their prime can afford to "go it alone" and try to. Its the rest of the time that open source works so well.
I wouldn't worry too much this is a lot like the arguments 15 years ago when the GPL was replacing BSD/MIT style licenses. The fact is as the GPLv3 spreads companies will reluctantly go along, and contribute resentfully at first that they have to give back.
Yes and no. I think each new generation needs to be educated in how free operating systems came to be closed. Particularly the history of the Unix wars of the 1980s. This subthread started with someone from the BSD side throwing out that can of worms, regarding "the GPL isn't really free".
I agree the debate is over mostly now that the GPL is around 70% with BSD around 7%.
Because the code as used doesn't match the code as spec. A great example is what happened to X before XFree86. There was a free X implementation that didn't actually run on any particular box, at least not well enough to be usable. What the free X implementation (MIT license) did was give a wonderful roadmap for operating systems manufacturers to see how various parts of X could be implemented and thus cut their costs and increased standardization. But everyone using X was using commercial products with commercial licenses, for end users the "free" code was worthless.
It wasn't until (a) XFree86, (b) commercial vendors stopped improving X, (c) a lot of time passed that we had an effectively free X again. You think of the free code as being worthwhile because you grew up in the world of the GPL and LGPL where people had to share back their improvement.
Oh please. During the days when licenses like the MIT, BSD and TeX license were dominant what we saw was: academics wrote open source software, companies took that open source software and wrote closed extensions that they didn't share back. The end result was the systems end users actually used were completely closed. The reason GPL is the dominant license is because GPLed software forced cooperation and collaboration.
We have hundreds of test cases of both styles of license and the results are pretty clear.
I agree with your statements about a,b and c. Of course c has an agenda but that agenda is often quite different from b's. And as for a, a reflects the bureaucracy not the elected it is often a very good source of information. It may be swayed a bit but the quantity makes it hard to eliminate counter evidence.
(c) is what provides the oversight you are looking for. And (c) can be quite good because i) They don't need to get interviews the next week. ii) They aren't looking to advertisers, who are often lined up with (b).
I remember. I had a 386-40 (AMD) that I bought with 8m ram and had upgraded to 20m. I replaced it with a Pentium-60 (1st gen) with 16m. I think that's that was the only memory downgrade in my life in terms of systems I personally owned included when I switched to laptops.
And you are confusing the NYTimes company with the NYTimes. In addition to things like NYTimes syndicate You are including:
International Herald Tribune The Boston Globe and the Globe Website. Telegram & Gazette of Worcester, Massachusetts The Gadsden Times of Gadsden, Alabama The Tuscaloosa News of Tuscaloosa, Alabama Petaluma Argus-Courier of Petaluma, California (weekly) The Press Democrat of Santa Rosa, California The Gainesville Sun of Gainesville, Florida The Ledger of Lakeland, Florida Sarasota Herald-Tribune of Sarasota, Florida Star-Banner of Ocala, Florida The Courier of Houma, Louisiana The Daily Comet of Thibodaux, Louisiana The Dispatch of Lexington, North Carolina Times-News of Hendersonville, North Carolina The Star-News of Wilmington, North Carolina Spartanburg Herald-Journal of Spartanburg, South Carolina about.com boston.com,
17% of Boston Red Sox, Fenway Park and Liverpool Football Club New England Sports Network (NESN) (80%) Donohue Malbaie, Inc. (49%) with Abitibi-Consolidated Madison Paper Industries (40%) in Madison, Maine
Slashdot Mirror
The GPL has always quite explicitly stated that the license changes as the FSF sees fit. Section 9:
I don't know they could have been more explicit that the GPL is the GNU Public License where GNU is project under the guidance of the Free Software Foundation.
Under your scenario the far cheaper thing to do is buy a commercial license for that one GPLv3 piece.
But even if they don't, I'm hard pressed to see the downside here. The company goes all commercial and starts heavy licensing fees which they have to pass on, which make their products more expensive than companies that are less open. The board is then confronted with a realistic type cost:
say $3k per developer per year, $1m in backend license fee support costs and $12 / device. If they think staying closed is worth that much, OK.
The reason companies paid developers to work on the Linux kernel was because they wanted to get out of writing their own kernels. It was from good will. And they hated the GPL even while they did enormous things to make it clear that large scale cooperation worked best under the GPL.
Which is the same argument were were having 15 years ago for GPL2. Companies didn't like those terms much either. They didn't have a choice as more and more open source software was GPLed. Apple would have loved when they were building the original samba extension in if Samba had been BSD and they could have created "iMacWin Share" as a sellable product.
GPL2 was very blunt instrument to attack the semi interoperability as a result of the Unix wars. It worked and everyone is better off.
Not always but almost always. He's got a pretty good track record.
Some of his architectural choices haven't proven so wise, his priorities were less than ideal. But damn he's got 3 decades of some decisions that worked out very well.
Authors aren't bound by licenses including their own. They have slightly more freedom than they would have under the BSD type scenario. If you are an author who released code under GPL2 you can turn around tomorrow and release it under BSD, MIT, PD...
As for being anti-commercial, read back the discussion in the late 1990s early 2000s regarding GPL2 vs. BSD. Same arguments. Yes companies like maximum freedom. Too bad for them.
No you can't because a slavery contract cannot be enforced. If you sell yourself to me as a slave day 1, you can leave day 2. If I try and stop you physically that's a battery, kidnapping... The most you can do is sell your labor in advance under guidelines of US labor law which prevent you from selling it too far in advance.
Even things like no compete clauses the courts have interpreted very strictly.
No it doesn't impact my software. It just impacts the version of my software that all the end users actually run and can run. See the history of X for details.
The system apple has for iOS for code signing is already great. A provisioning file requires a crytographic key involving the device and the signature of the application and the signature of the person sending it. The person installing the application needs to first copy the provisioning file into iTunes. Just make that private key public and you have open system.
So even if I infect myself how do infect others. I don't have the Samba Team's private key so I would need to use another private key that wouldn't match the supposed author.
Then they are going to have a real problem. Think about how complex the software is that runs on macs vs. what runs on tablets.
Lock it down and they kill the platform. People will accept their phones being limited because they always have been (RIM being something of an exception). They may accept tablets being limited but already their is a bit of a backlash. No way will it be accepted on computers.
That wouldn't present a problem. If the Samba team, or someone not Apple, released Samba for OSX via. the Apple Store as a free application then Apple doesn't pick up a GPL3 patent issue in the same way a book store doesn't pick up a copyright problem for a book in good faith that they sell.
All of the GPL is political. Always has been. The GPL3 is about plugging holes in GPL2. Patents can and have been used to effectively limit freedoms under GPL2. Ergo...
Actually at this point Microsoft seems more free than Apple. Apple was never a huge fan of open source, they were desperate. Generally companies in their prime can afford to "go it alone" and try to. Its the rest of the time that open source works so well.
I wouldn't worry too much this is a lot like the arguments 15 years ago when the GPL was replacing BSD/MIT style licenses. The fact is as the GPLv3 spreads companies will reluctantly go along, and contribute resentfully at first that they have to give back.
Yes and no. I think each new generation needs to be educated in how free operating systems came to be closed. Particularly the history of the Unix wars of the 1980s. This subthread started with someone from the BSD side throwing out that can of worms, regarding "the GPL isn't really free".
I agree the debate is over mostly now that the GPL is around 70% with BSD around 7%.
Well yeah, the GPL doesn't assume everyone has good will. Rather it gives the downstream authors a choice:
a) Contribute to the open source world
b) Buy another license from the author on more favorable terms, and thus contribute financially to open source.
Because the code as used doesn't match the code as spec. A great example is what happened to X before XFree86. There was a free X implementation that didn't actually run on any particular box, at least not well enough to be usable. What the free X implementation (MIT license) did was give a wonderful roadmap for operating systems manufacturers to see how various parts of X could be implemented and thus cut their costs and increased standardization. But everyone using X was using commercial products with commercial licenses, for end users the "free" code was worthless.
It wasn't until (a) XFree86, (b) commercial vendors stopped improving X, (c) a lot of time passed
that we had an effectively free X again. You think of the free code as being worthwhile because you grew up in the world of the GPL and LGPL where people had to share back their improvement.
BSD gives total freedom to software implementers and takes that freedom away from end users (indirectly).
Oh please. During the days when licenses like the MIT, BSD and TeX license were dominant what we saw was: academics wrote open source software, companies took that open source software and wrote closed extensions that they didn't share back. The end result was the systems end users actually used were completely closed. The reason GPL is the dominant license is because GPLed software forced cooperation and collaboration.
We have hundreds of test cases of both styles of license and the results are pretty clear.
I agree with your statements about a,b and c. Of course c has an agenda but that agenda is often quite different from b's. And as for a, a reflects the bureaucracy not the elected it is often a very good source of information. It may be swayed a bit but the quantity makes it hard to eliminate counter evidence.
(c) is what provides the oversight you are looking for. And (c) can be quite good because
i) They don't need to get interviews the next week.
ii) They aren't looking to advertisers, who are often lined up with (b).
I remember. I had a 386-40 (AMD) that I bought with 8m ram and had upgraded to 20m.
I replaced it with a Pentium-60 (1st gen) with 16m.
I think that's that was the only memory downgrade in my life in terms of systems I personally owned included when I switched to laptops.
gong, gong.
See above.
And you are confusing the NYTimes company with the NYTimes. In addition to things like NYTimes syndicate You are including:
International Herald Tribune
The Boston Globe and the Globe Website.
Telegram & Gazette of Worcester, Massachusetts
The Gadsden Times of Gadsden, Alabama
The Tuscaloosa News of Tuscaloosa, Alabama
Petaluma Argus-Courier of Petaluma, California (weekly)
The Press Democrat of Santa Rosa, California
The Gainesville Sun of Gainesville, Florida
The Ledger of Lakeland, Florida
Sarasota Herald-Tribune of Sarasota, Florida
Star-Banner of Ocala, Florida
The Courier of Houma, Louisiana
The Daily Comet of Thibodaux, Louisiana
The Dispatch of Lexington, North Carolina
Times-News of Hendersonville, North Carolina
The Star-News of Wilmington, North Carolina
Spartanburg Herald-Journal of Spartanburg, South Carolina
about.com
boston.com,
17% of Boston Red Sox, Fenway Park and Liverpool Football Club
New England Sports Network (NESN) (80%)
Donohue Malbaie, Inc. (49%) with Abitibi-Consolidated
Madison Paper Industries (40%) in Madison, Maine