No, this time it isn't. The problem is that the wording of the GPL does not justify any other interpretation (in the eyes of the FSF lawyers, that is), whether they like it or not. Not only does the FSF still consider the Apache license a Free Software license, they explicitly included a remark on their page stating that they don't really think that the terms making it GPL-incompatible are actually a bad idea.
In the end, a license is a piece of text. All that matters is what it says, not what the intention of its authors were. Just like with source code - I'm sure the Linux kernel hackers didn't actually intend do_mremap to be exploitable.
Petition the FSF to include a statement to that effect in GPL3, and to release GPL3 by the end of the year.
I think it is pretty likely that the GPL3 will include something on patents anyway, this actually seems to be one of the two major reasons why people feel an update is neccessary (the other one being the "ASP hole", i.e. the GPL not being prepared for a web application scenario)
It being released this year seems less likely, AFAIK it is still in pie-in-the-sky mode. Sure, you can petition them about it, but remember that the consequences of a rushed, ill-concieved GPL V.3 would be way worse than that of a delayed one. "Release early, release often" probably doesn't work that well for legal documents.
Well, if you ask me, you get around the patent issue by ignoring the GPL and just use the Apache license. The ugly thing about this issue is that the FSF had to declare the Apache License 2.0 GPL-incompatible because the GPL is more friendly towards software patent abusers, because this issue is mostly not dealt with in GPL v2.0 at all (except some very broad explanations). I think nobody, including the FSF, is really happy about that (but then, I wouldn't know).
The GPL is, to some extent, viral. Once any new source code interacts with GPL'd source code, the new source code becomes GPL'd.
This is a very polemic description, to the extent that it is just plain wrong. The GPL doesn't just infect other projects like viruses do, it takes a consious decision to create a derived work of a piece of GPLed code to make it kick in. And there are plenty of ways to interact that do not count as creating a derived work.
Not to mention that APIs cannot be GPLed, only implementations.
I chose the headline, the editors are not to blame.
I still do not consider it ambigous or a flame-bait, in fact, I think it is more appropriate than you proposal (because it is just the FSF's opinion that the Apache license is not GPL-compatible, and others, like Roy Fielding, disagree - and if it really is not would have to be determined in court.)
This is not about an individual, or even the FSF as an organization, enforcing their feelings; it doesn't matter if they like it or not. The GPL's wording is clear in that derived works may not distributed under a license that imposes any restriction not found in the GPL itself, and the Apache license supposedly imposes such restrictions (even if these restrictions only affect people who abuse software patents, and the FSF definitely does not like people abusing software patents).
Still, it is certainly the case the it is the GPL that prevents combination of ASL- and GPL-licensed code in a derived work, the ASL is fine with that.
No, it is about the GPL. Whether a license is GPL-compatible or not can be mechanically deduced by looking at its terms, and neither the FSF nor anybody else has the power to declare some license GPL-compatible because they think it is morally or politically good. The FSF doesn't seem to have much of a political problem with the new Apache license, but it is still not GPL-compatible.
This is not the first case when the FSF had to declare a license they actually liked GPL-incompatible, the Affero GPL is another.
Because GPL-compatibility was one of the reasons to develop ASL V2 in the first place (as the article states, by the way). That it didn't work out is kinda disappointing.
In the end, neither the opinion of the FSF nor the one of the Apache Foundation counts. Not only are both licenses used by others that might interpret them differently, in the end it is up to a judge to decide what they really mean.
But it's true, I wouldn't expect any lawsuits between ASL- and GPL-using free software authors any time soon.
C and derived languages have a datatype called "integer" which is actually just a small subset of integral numbers, as well as a datatype called "char" which doesn't hold characters at all, but is a subset of "integers". This doesn't seem to have stopped their adoption, even if it contributes to the aquired stupidity common amongst their users.
Personally, I think having NFS, NIS, PAM, Gnome, OpenOffice, DocBook, freedesktop.org etc. developed or supported by Sun is worth more than them being sued by SCO, but that is just my personal opinion. I just wonder, do I have to deinstall KDE and Qt? Trolltech wasn't sued either.
There is an important difference between RMS and ESR: RMS is nuts because he cares for his idea of freedom (as silly as it may be), and nothing else. He is extremely predictable and straightforward in that. ESR is just an inconsitently flaming, self-loving troll.
Well, actually there is another difference: RMS wrote GCC and Emacs, ESR wrote Fetchmail and a text adventure-like interface for Linux kernel configuration that nobody liked. I guess a good psychoanalyst could draw a line between not having contributed anything significant and having to pose as the alphageek all the time.
I have never seen ESR praised by anyone but himself, actually.
Technically, he certainly is a "father of OSS", as he was one of those (together with Tim O'Reilly and others) who came up with, and implemented, the idea of avoiding the "freedom" rethoric to appeal to suits, and unfortunatly he even is the head of OSI, but I doubt that there are many people who both understand what he is saying and take him serious.
Do you honestly think RMS would write something better than ESR on this matter?
I can easily imagine that. He would likely claim that non-free Java implementations are useless, and that people should support projects like Kaffee and Classpath to create a free one, instead of denying the existence and possibility of these projects as ESR did.
As you probably know (given your learning of Scheme), it gets really cool if you are not bound by the decisions of the language designer in what paradigms you can use.
I am a happy user of Common Lisp, which supports imperative, functional, OO etc. out of the box. But what is more important is that the language itself is extensible. It is a rather common approach to write applications by first extending the language, and such extensions can go quite far. For example, people have integrated prolog style logic programming into Lisp, there are libraries that support nondeterministic evaluation and backtracking, prototype-based alternatives to the native OO system, dataflow, you can use it as a dynamic markup language etc. The nice thing is that all this gets part of the Lisp system just like the builtins, so that a programmer can mix and match all of these, and the features of CL itself.
Paul Graham has written a really cool book on that topic, which is freely available.
Or perhaps another FreeBSD - AMD64 is considered a tier 1 platform for FreeBSD. IA64 is not yet, but from what I've heard, the remaining "stability issues" are more about ABI stability that about not crashing.
I assume that you think of Netscape <= 4 when you say "Netscape". To wich I answer:
Oh my god no, please do not! You can say what you want about Lindows, for example that it is an utter piece of crap and nobody would possibly want to use it if it weren't for the hype generated by the stupid name, but it stil is not as bad as the broken tag-soup interpreter we all know (but nobody loves) as Netscape.
Nobody, not even Lindows, should try to make something "as good as Netscape". Instead, try making something that doesn't deliberately violate open standards, works more than half of the time without crashing regularly. If you care about Unix users, you might even consider not setting the default font size to half a pixel.
I tend to agree that a lot of things going on in the trademark/copyright/patent arena are horribly wrong. However, this case it a little different: Lindows chose that name because it is so similar to "Windows", and because it is a provocation. They aren't innocent victims - getting sued by microsoft, and potentially having to use another, similar, name was part of their business model from day one.
Just because they use a Linux kernel does not mean that Lindows is somehow a "better" company. Lindows is run by weasels just like any other company, be it Microsoft, SCO, Sun, Borland or Novell. If you are a shareholder, this is good for you, if you are a thinking human being however, there is just no reason whatsoever to sympathize with them.
The program here refers to the distribution as such.
A distribution like Mandrake is not a program. Just shipping XFree86 4.4 and GPLed programms is not a problem - it is "mere aggregation" and not creation of a "derived work", and the GPL does not (and, AFAIK, can not) restrict that, and says so explicitly, in the last paragraph of clause (2).
Why should one come up with a new one when there already are plenty XML formats that can be used for knowledge organization, like XML Topic Maps and RDF?
I'm a FreeBSD user as well, I use it both on servers and on my personal desktop, but it is clear that it is way less popular than Linux. And Linux is mostly irrelevant as a dektop OS.
It doesn't matter how many people besides you use the same OS, or how fast its adoption grows. Just use it as long as you personally feel that it is the best choice for you, and when you begin to think that another OS is better, switch to it and be happy because the usefullness of your system improved. The output of "uname" does not matter. Software is a tool, not a religion.
I agree that some programs have a quality that is somewhat close to literature, but maybe not poetry. In particular, I agree with Richard Garbiel that there should be a Master of Fine Arts in Software.
I still claim that software is a discipline of its own, and natural languages and its literature are only very loosely related to it.
There are over 1 billion native speakers of either Cantonese or Mandarin.
Not many of them happen to live in europe, however. The trick is that (some form of) english is understood virtually anywhere. Almost all of the chinese speakers live in china, this doesn't help in international communication.
Slashdot Mirror
In the end, a license is a piece of text. All that matters is what it says, not what the intention of its authors were. Just like with source code - I'm sure the Linux kernel hackers didn't actually intend do_mremap to be exploitable.
It being released this year seems less likely, AFAIK it is still in pie-in-the-sky mode. Sure, you can petition them about it, but remember that the consequences of a rushed, ill-concieved GPL V.3 would be way worse than that of a delayed one. "Release early, release often" probably doesn't work that well for legal documents.
Not to mention that APIs cannot be GPLed, only implementations.
I still do not consider it ambigous or a flame-bait, in fact, I think it is more appropriate than you proposal (because it is just the FSF's opinion that the Apache license is not GPL-compatible, and others, like Roy Fielding, disagree - and if it really is not would have to be determined in court.)
Still, it is certainly the case the it is the GPL that prevents combination of ASL- and GPL-licensed code in a derived work, the ASL is fine with that.
This is not the first case when the FSF had to declare a license they actually liked GPL-incompatible, the Affero GPL is another.
Because GPL-compatibility was one of the reasons to develop ASL V2 in the first place (as the article states, by the way). That it didn't work out is kinda disappointing.
But it's true, I wouldn't expect any lawsuits between ASL- and GPL-using free software authors any time soon.
C and derived languages have a datatype called "integer" which is actually just a small subset of integral numbers, as well as a datatype called "char" which doesn't hold characters at all, but is a subset of "integers". This doesn't seem to have stopped their adoption, even if it contributes to the aquired stupidity common amongst their users.
Personally, I think having NFS, NIS, PAM, Gnome, OpenOffice, DocBook, freedesktop.org etc. developed or supported by Sun is worth more than them being sued by SCO, but that is just my personal opinion. I just wonder, do I have to deinstall KDE and Qt? Trolltech wasn't sued either.
Well, actually there is another difference: RMS wrote GCC and Emacs, ESR wrote Fetchmail and a text adventure-like interface for Linux kernel configuration that nobody liked. I guess a good psychoanalyst could draw a line between not having contributed anything significant and having to pose as the alphageek all the time.
Technically, he certainly is a "father of OSS", as he was one of those (together with Tim O'Reilly and others) who came up with, and implemented, the idea of avoiding the "freedom" rethoric to appeal to suits, and unfortunatly he even is the head of OSI, but I doubt that there are many people who both understand what he is saying and take him serious.
This hasn't stopped anyone from shipping apache, which is also under a GPL-incompatible license.
I am a happy user of Common Lisp, which supports imperative, functional, OO etc. out of the box. But what is more important is that the language itself is extensible. It is a rather common approach to write applications by first extending the language, and such extensions can go quite far. For example, people have integrated prolog style logic programming into Lisp, there are libraries that support nondeterministic evaluation and backtracking, prototype-based alternatives to the native OO system, dataflow, you can use it as a dynamic markup language etc. The nice thing is that all this gets part of the Lisp system just like the builtins, so that a programmer can mix and match all of these, and the features of CL itself.
Paul Graham has written a really cool book on that topic, which is freely available.
Or perhaps another FreeBSD - AMD64 is considered a tier 1 platform for FreeBSD. IA64 is not yet, but from what I've heard, the remaining "stability issues" are more about ABI stability that about not crashing.
Oh my god no, please do not! You can say what you want about Lindows, for example that it is an utter piece of crap and nobody would possibly want to use it if it weren't for the hype generated by the stupid name, but it stil is not as bad as the broken tag-soup interpreter we all know (but nobody loves) as Netscape.
Nobody, not even Lindows, should try to make something "as good as Netscape". Instead, try making something that doesn't deliberately violate open standards, works more than half of the time without crashing regularly. If you care about Unix users, you might even consider not setting the default font size to half a pixel.
Just because they use a Linux kernel does not mean that Lindows is somehow a "better" company. Lindows is run by weasels just like any other company, be it Microsoft, SCO, Sun, Borland or Novell. If you are a shareholder, this is good for you, if you are a thinking human being however, there is just no reason whatsoever to sympathize with them.
Why should one come up with a new one when there already are plenty XML formats that can be used for knowledge organization, like XML Topic Maps and RDF?
I'm a FreeBSD user as well, I use it both on servers and on my personal desktop, but it is clear that it is way less popular than Linux. And Linux is mostly irrelevant as a dektop OS.
It doesn't matter how many people besides you use the same OS, or how fast its adoption grows. Just use it as long as you personally feel that it is the best choice for you, and when you begin to think that another OS is better, switch to it and be happy because the usefullness of your system improved. The output of "uname" does not matter. Software is a tool, not a religion.
I agree that some programs have a quality that is somewhat close to literature, but maybe not poetry. In particular, I agree with Richard Garbiel that there should be a Master of Fine Arts in Software.
I still claim that software is a discipline of its own, and natural languages and its literature are only very loosely related to it.
I'm partial for ancient greek myself.