Does anyone know why Linus made this, at first glance, boneheaded decision?
Probably because he didn't want to give a carte blanch *in case* a later version would be weak or bad in some way. (At least that was the reason I had once to not include the "or later version" option.)
The problem is presumably that many choose not to add the "or any later version" option to their license.
The reason why people don't add the option is probably because if a later version of the license is weak or bad, they don't want to give users carte blanch to use the new bad license version.
However, since the new license version is, by the look of it, better and stronger, most should have no problem licensing their programs under the new GPL version.
On the other hand, people might quit licensing under the old versions since the new one is better.
Considering how many that are confused about what Turing-completeness means, the grandparent wasn't being a jerk at all. Grandparent was just clarifying things.
Considering that you say that this is a pet peeve of yours, your own grasp of the issue seems to be left wanting. Your mock argument ending with
Computer languages can therefore never be Turing-complete is nonsense since the question if a computer language is Turing-complete is perfectly valid no matter what. Computer languages are just math definitions and higher level languages like Haskell are Turing-complete (I'm sure even though I don't know the details of the Haskell def). In contrast, lower higher level languages like C are not Turing-complete (because the language def imposes limits on most things, e.g. types).
Actually that's the main idea behind RPL, that you are not allowed to use (e.g. link) RPLed programs within a company without making your work availabe. In other respects RPL is roughly like GPL.
So if you don't want e.g. Microsoft or Intel to build on your FOSS and distribute it within the company without giving something back you should consider using RPL.
Here is a link with some info on how the parties and some candidates stand on the issue.
Roughly, the two largest parties, the social democrats (socialdemokraterna) and the right wing party (moderaterna), are in favor of software patents while the rest, the liberals (folkpartiet), the leftmost party (vaensterpartiet), the greens (miljoepartiet),... are against.
(I'm voting for Olle Schmidt of the liberal party but I'm not affiliated with anything above.)
IANAL and I only had a quick look at the cited article but it doesn't seem very authoritative. Common sense (and FSF) say that you can't revoke or change contracts or licenses after they have been agreed, no matter if you pay for them or not. So a change of license will only affect future versions of the software. And the author simply misunderstands clause 9 of the GPL saying that it can take away your rights to use GPL software when it only can be used to strengthen them.
A change of license would affect future versions of some software though. That's why FSF wants you to hand over the copyright of your GPL code to them...
Slashdot Mirror
There is a NYT editorial today titled "Grand Theft Adult": http://nytimes.com/2005/07/22/opinion/22fri4.html
Probably because he didn't want to give a carte blanch *in case* a later version would be weak or bad in some way. (At least that was the reason I had once to not include the "or later version" option.)
The problem is presumably that many choose not to add the "or any later version" option to their license.
The reason why people don't add the option is probably because if a later version of the license is weak or bad, they don't want to give users carte blanch to use the new bad license version.
However, since the new license version is, by the look of it, better and stronger, most should have no problem licensing their programs under the new GPL version.
On the other hand, people might quit licensing under the old versions since the new one is better.
Considering that you say that this is a pet peeve of yours, your own grasp of the issue seems to be left wanting. Your mock argument ending with Computer languages can therefore never be Turing-complete is nonsense since the question if a computer language is Turing-complete is perfectly valid no matter what. Computer languages are just math definitions and higher level languages like Haskell are Turing-complete (I'm sure even though I don't know the details of the Haskell def). In contrast, lower higher level languages like C are not Turing-complete (because the language def imposes limits on most things, e.g. types).
RPL, reciprocal public license, requires you to release back even if you only distibute within the company. See http://www.opensource.org/licenses/rpl.php/
Actually that's the main idea behind RPL, that you are not allowed to use (e.g. link) RPLed programs within a company without making your work availabe. In other respects RPL is roughly like GPL.
So if you don't want e.g. Microsoft or Intel to build on your FOSS and distribute it within the company without giving something back you should consider using RPL.
Roughly, the two largest parties, the social democrats (socialdemokraterna) and the right wing party (moderaterna), are in favor of software patents while the rest, the liberals (folkpartiet), the leftmost party (vaensterpartiet), the greens (miljoepartiet),... are against.
(I'm voting for Olle Schmidt of the liberal party but I'm not affiliated with anything above.)
Here is a positive review from an alpha tester.
Except that Turing's famous paper on computability came 1936...
Yeah, I agree that it would be good to hear what Moglen or someone at FSF has to say about this. Why don't you email gnu@gnu.org the question.
IANAL and I only had a quick look at the cited article but it doesn't seem very authoritative. Common sense (and FSF) say that you can't revoke or change contracts or licenses after they have been agreed, no matter if you pay for them or not. So a change of license will only affect future versions of the software. And the author simply misunderstands clause 9 of the GPL saying that it can take away your rights to use GPL software when it only can be used to strengthen them.
A change of license would affect future versions of some software though. That's why FSF wants you to hand over the copyright of your GPL code to them...
The jist was meeting Nicholas Walker, an 18-year-old high school dropout and maker of GNU/Linux.
How hard can it be?p