"No, I'm sorry, you can't integrate my small component into your giant proprietary and ITAR-restricted satellite system unless you agree to give away the entire code to your giant ITAR-restricted satellite system to anybody who wants it. But hey, RMS says that freedom is good, so you can do that, right?"
So, you're another one of those "I don't like the GPL, but I haven't actually read the license" type of people, eh?
Well, to save you the trouble of actually reading the license, I will point you to a quick summary on Wikipedia.
They are the ones who decided on this policy, and they claim a moral (yes, moral) high-ground because of it. But really, they only mean this within the context of their own ideological community. It's really a difference of attitude.
Nonsense. The GPL is a very pragmatic license. People who write and distribute their code decide how they want to allow it to be redistributed. There is no moral claim there at all. Don't confuse this with Richard Stallman's claim of the morality of Free Software, which is a completely different subject.
Anyway, the claim a lot of people are making here is that advocates of the GPL should contribute back to BSD-projects under a BSD-license because otherwise they are being hypocritical. I think that is fairly silly, because then you are basically saying, this is a BSD-project, this code can't be used by GPL-projects, and that is contrary to the BSD spirit. Either you care about how your code is redistributed or you don't. You can't say, "go ahead and use our code, we don't care what you do with it" and then a little bit later say, "except we don't want our code used in a GPL-project."
Of course, this whole discussion is a non-issue in the first place. The original commit was a mistake and has been fixed. The code derived from was dual-licensed under BSD and GPLv2, which explicitly allows it to be redistributed under GPLv2. The mistake was that the commit omitted the dual-licensing part, which a number of people on LKML noted, and was fixed.
Well, I'm not sure what the GP's printer is, but many of the Brother laser printers are Postscript printers. The problem isn't the printer, or CUPS, it is the way that printing is setup in Ubuntu. The config tools just don't setup the printer profiles correctly. I've found that if I bypass the printer config tool and just use the CUPS web-based tool, the printer works perfectly. Meanwhile, the last time I was playing around with Fedora, their printer config tool worked perfectly. This is really just an Ubuntu problem, although I'm not sure what exactly the problem is because I'm using the same Postscript driver with the same PPD for my printer no matter what.
I know a new printer config tool is in the works, but as far as I can tell it has been delayed through the last two releases, and it doesn't look like it will make it into Gutsy either. It seems there have been some hangups.
I'm not so sure it is wise to jump to such a hasty conclusion. The thing is, despite the marketing hype, hi-def just isn't that big yet. There isn't enough content, there aren't enough people buying hdtvs, and there just isn't a lot of interest. The PS3 hasn't been out for very long. Most likely the people who have bought it up until now (despite the lack of interesting games which won't be out until next year), didn't do it because they wanted the blu-ray player. Combine that with the fact that the current selection of blu-ray movies is fairly mediocre, and it isn't a surprise that number of players doesn't correlate well with movie sales. In a few years, when there is a larger selection of content, prices are lower, and there is generally more interest, you can be sure that the more ubiquitous player (don't know if it is going to end up being blu-ray or hd-dvd at this point) is going to have a huge advantage. And if you already have a PS3, why would you buy a hd-dvd player?
Uhhh, not necessarily agreeing with the GPs post, but I think your numbers are fairly exaggerated...a million songs to support one member of a band? I don't think so.
Let's take this from a different angle. A song on iTunes costs $0.99. Let's assume for the sake of argument that the royalty rate to the band is about the same as for cds (around 10% unless you are popular enough to negotiate higher). That is $0.10/song. Also, remember that the band has to pay back the label for the overhead costs, so they are getting even less than that, to split up between all of the band members. The other $0.89 doesn't go to overhead costs, it goes right into the pockets of the labels. So when the GP is saying he will pay $0.10/song, I'm sure (or at least I hope) he means $0.10 directly to the artists, which is comparable to what a lot of them get now (maybe even a little better). In other words, cut out the middle man and there is no need for prices to be that high.
The problem I have is that like many here on slashdot I do happen to know a thing or two about computers.
But, do you know a thing or two about science? Exactly what are you looking for when you say you want more substantial evidence. A scientific model is a scientific model. The validity is going to depend on the assumptions, the measurements to check those assumptions, and the data that is fed into the model. Whether or not it is done on a computer has nothing to do with its validity, although computers are arguably more capable of dealing with things that require massive amounts of computation, like solving the complex differential equations associated with climate models.
What I do see very clearly is that all suggested solutions seek to punish the united states while leaving some other areas of the world untouched.
Ok, that's a nice political slant to put on things. I'm not sure what you are specifically referring to here, but if you are concerned about carbon emission caps and such, the United States (and Canada) consumes significantly more energy per capita than other comparably developed nations around the world. That contributes to significantly larger carbon emissions. To reach sustainable levels, the United States has to reduce more because it is producing more. You can say that is "punishing the United States", but that is really irrelevant. If the United States had a more conservative energy usage policy to begin with, it wouldn't be in this situation. However, for many reasons including the heavy influence of the oil and coal lobby, that is not the case, and so here we are.
That was never it's purpose. The purpose of DRM was to curb copying,
The problem with that statement, and I'm surprised nobody has pointed it out yet because a lot of other comments are saying the same thing, is that DRM does nothing to prevent copying, not even a little bit. I can freely copy encrypted dvds and encrypted music files to my computer, to my friend's computer, to my portable music/video player, anywhere. The only thing DRM does is require an "authorized player" to watch the video/play the music. As long as I have an authorized player (ex: PowerDVD), I can play my copied and encrypted dvd. The authorized player can, of course, try to enforce it's own copying restrictions (ex: iTunes) by refusing to decrypt content it doesn't think you own or have the rights to, but this is the player and not the DRM. Massively distributed players that need to be able to play "random dvd off the shelf" can't enforce any type of copying restrictions, so you are left with this elaborate and expensive encryption scheme that ultimately doesn't do what it is supposedly designed to do, which is prevent copying. But, of course, that's not what DRM was really designed to do...I'll leave the rest of that thought to the reader.
And creates ambiguity between GNU Projects and Linux
I think the slash fairly clearly indicates that it is GNU slash linux (ie: a mixture of two projects), and not GNU Linux. The Debian doc you have linked to is over 10 years old. If you look at the front page of their website, you will see that all references to the distribution are denoted as Debian GNU/Linux.
I assume you mean: "adopted the project and copyright as a GNU project".
No, what I mean is the FSF starts a GNU project, say GCC. They have written a large chunk of code at that point and own the copyright on it. Now if the project is managed like the linux kernel, anybody who contributes code has to license it under the GPL, but they retain their copyright to the code they have written. The FSF, however, asks all contributors to transfer their copyright to the FSF if they want to be a part of the project. Other projects are known to do this as well: CUPS, MySQL, OpenOffice.org, and Apache are the few that come to mind right away.
Once you sign over the copyright, its not your project anymore
You do realize you can transfer your copyright on some code and not other code, right? Also, the terms of a copyright transfer will vary, but the original developer almost always retains unrestricted rights to his code. He just ends up granting unrestricted rights to the third-party as well.
As for needing the copyright to enforce the GPL, after using XEmacs code in GNU Emacs under the GPL they're perfectly capable of enforcing the GPL on that code because the derived works are covered by the GPL v2.0 section 2 - b; unless they don't believe that clause is enforcible.
You cannot prosecute GPL violations without money and legal expertise. The FSF has both. The FSF is more than willing to prosecute on your behalf, but they can't do that if they don't own the copyright.
That credit doesn't extend to kowtowing to RMS's ego and renaming software because he says so.
RMS hasn't said anything about renaming the linux kernel project. He just wants mixtures of Linux and GNU software to be referred to as GNU/Linux. That does not seem unreasonable to me. It doesn't hurt anyone in any way to do that.
is attempts to rebrand Linux as GNU/Linux are effectively trying to brand Linux as a GNU package
No, I think you are letting a dislike of RMS distort your opinion. He is not rebranding the linux kernel. He is not trying to take over anything. He just wants people to know, from the name, that a linux distribution contains software in addition to the linux kernel that is completely separate and has nothing to do with the brainchild of Linus Torvalds. If I wanted to be pedantic, I could say that using the name Linux for the distribution is an attempt to rebrand the GNU software that is included and make it appear like it all came from the same source when it didn't. The GNU project existed long before Linux, after all. But, I think the argument is stupid, so I won't.
Transfer of copyright is transfer of ownership or "take over".
I think you missed (or just ignored) the part where I said why they did this and that other projects do the same thing. If I start a project and you want to contribute code to it, I can ask you transfer your copyright to me. You can, of course, refuse and and in that way I would be excluding a lot of potential developers from the project, but if I want to make that requirement it is within my rights. It's not "taking over" another project because it was my project to begin with. RMS has only asked for transfer of copyright on contributions to GNU Emacs, not XEmacs. Now the naming issue mentioned in the FAQ, I do agree is silly. If XEmacs is indeed largely independent code (I don't know because I haven't looked), RMS doesn't have a claim to the name. Also, it seems RMS has backed off because the emacs FAQ references XEmacs as XEmacs, not GNU XEmacs. Other projects on the GNU website do as well.
Its there, its GPL, nothing prevents you from doing so.
Well, I asked whether it was ok, not whether it was legal. I know it is legal, but I'm not sure I am ethically ok with taking large chunks of another project and not at least mentioning that my project is derived from a different project. I suppose a better question is, would you take the linux kernel and credit Alan Cox for his contributions, but not Andrew Morton?
What happened with GNU and XEmacs says something completely different.
Not sure what you are referring to here. If you are referring to the debate over copyright assignment, the FSF asks all GNU projects to transfer their copyright so they are better able to prosecute GPL violations. The FSF is certainly not the only organization to do this. Easy Software Products and the CUPS project is a recent example that has received attention lately due to the buyout from Apple. The contention between XEmacs and GNU Emacs is that XEmacs occasionally wants to contribute code back to GNU Emacs, but the FSF wants the copyright transferred. It's a conflict between the two projects, yes, but I think suggesting that the FSF is trying to "take over" XEmacs is a deliberate misconstruing of the actual situation. And what does this have to do with the linux kernel? Where is the posting on LKML where RMS asks all of the kernel developers to transfer over their copyrights to the FSF?
So, if I took the linux kernel and rewrote the SCSI subsystem, and then rebranded and distributed it as Rutulian's superkernel, that would be completely ok with you?
What about the fact that branding it GNU/Linux is effectively an attempt to make it part of GNU?
One could perfectly well have a Linux system without Gnu tools, and no, it wouldn't be useless. Embedded systems often run a single application and this could be done with tools other than Gnu's.
Yes, and in that case you wouldn't call it GNU/Linux. In fact, they don't really call it anything most of the time because most people don't really care about the operating system of their embedded devices as long as it works. The point isn't that Linux can or can't be packaged with other things. The point is that the linux kernel by itself is not an operating system (neither is GNU for that matter).
Linux uses Gnu for the simple reason that Gnu existed, if it didn't exist it wouldn't be very hard to write. A lot of work, yes, but not as difficult as writing an OS kernel, the hurd is there to prove that.
No, I don't think that is true at all. Writing a kernel is hard, but writing a compiler and toolchain isn't easy either. The Hurd has been slow to get off the ground for a lot of reasons. The difficulty of the task is really only one of them.
The Gnu project needs Linux more than Linux needs Gnu.
That's a pretty stupid statement to make. It's like saying strawberry shortcake needs the shortcake more than the strawberries. No, they both need each other equally. You can package Linux with BSD or GNU with Solaris, but very few people do. Most people package GNU with Linux. Seriously, why do people fight this. It's just a name. RMS wants credit for his part in the creation of the modern linux distribution. Why is that so difficult for people to accept?
For the record, I never refer to my operating system as GNU/Linux because it is tedious, and I prefer to just shorten it to Linux. But it is understandable for RMS to request that people refer to it as GNU/Linux. And, if I was ever writing or speaking about it in some sort of official capacity, I would probably extend him that courtesy.
Yeah, sure, I suppose that is a legitimate fear...although Red Hat is actively working on their own management tools at the moment, and it will probably be a mute point in a year or so. Now that I think about it, though, I don't think I have ever known Red Hat to take another distribution's management or config utility and incorporate it into their own. Usually it is the other way around. Red Hat uses the Gnome System Tools package, but their printing and all of their server config tools are their own. Suse has had Yast and Mandrake has had the Drak tools for quite a while, but Red Hat hasn't touched them, even though some of them are superior to the ones Red Hat uses.
Also, what Xen fears they can also use to their advantage. Take Red Hat's distribution and all of their work patching the kernel, customize it to be a drop in hypervisor with easy config and management tools, and sell it with support options. I'm sure there are people out there looking for turn-key virtualization solutions. Remember, open source allows the free exchange of ideas, which means ideas can flow both ways. Red Hat can take from Xen, but Xen can also take from Red Hat.
What I called hypocritical was to do so in the name of "freedom", while you're at the same time trying to argue that I don't have the "freedom" to make my own choice.
Yeah, well, if they compile my code and sign the binary and make their hardware run only the signed binary, they have restricted my freedom to do what I want with my code on hardware I have purchased.
If I put my code under the gplv3, thereby requiring them to distribute their signing keys with my code, I have preserved my own freedom, but impinged on their freedom to prevent modified code from running on hardware they have created (unless, of course, they pay me or somebody else for that freedom).
So who's freedom is more important?
I lean toward my own freedom because I have taken the time to write the code and am offering it up for consumption by the community. Companies should have freedom too, but they don't get free labor. There is the freedom of ideas, but there is also economics. Nothing in the business world is free. If they don't want to pay me by adhering to the terms of my license, they need to pay me in cash, or they need to pay somebody else to develop for them.
Tivo's video-processing code can still be used in other applications, after all.
That's a good point, but what if I want to add a new dithering algorithm and run it on my Tivo? Actually, let's take a different tack. I write general networking stack. Company makes a networking appliance and uses general networking stack to run it. I realize later that there is an exploitable bug in my code and write a patch. Now, can I use my patched software with the networking appliance I purchased, or do I have to wait for the company to get around to compiling the new version and distributing signed binaries?
Furthermore, doesn't the GPLv3 prohibit *anybody* from writing GPLv3 code that runs on a Tivo
No. I'm not sure where you got that idea. I suggest actually reading the GPL. Specifically the part in the preamble that states the terms of the GPL apply only to those who distribute the software. If you own the copyright to the code, you can do whatever you want with it.
Linus is, of course, entitled to his opinion. But, seriously, the new provisions of the gpl don't hurt anyone. You have to remember, companies that wish to distribute gpl'd software are still getting a lot of work done for free (which saves them a lot in development money). Giving back modifications and the right to use modifications free from patent and hardware restrictions is a fair trade. It's not like they are requiring everybody to contribute $5 to the FSF legal fund for every binary distributed. If a company doesn't like the terms of the gpl, they can do as they have always been able to: find a differently licensed project with similar functionality, write their own code, or pay the developers for the right to distribute the code under an alternate license.
What limit on hardware manufacturer's? There is no limit on hardware manufacturer's. The only requirement is that if your hardware will only run signed code, the means to reproduce that signed code must be distributed with the software (which is distributed with the hardware). You can do whatever you want with your hardware, but if you use and distribute (i.e: make money selling) my software with your hardware, you can't prevent me from modifying (ex: with an newer version that adds extra features) said software and running it on your hardware (or rather, the hardware can prevent it, but you have to provide me with the means to make my software conform to what the hardware will run). It seems like a fair trade to me, and is not entirely unreasonable. The only people who see it as unreasonable are the hardware manufacturer's who want to use free code and not give anything back to the original developers.
Well, there's not a lot of context in the article, but piecing things together, I'd guess that what has Linus ticked off is not injecting personal morality into the license, but forcing others to accept that license and by extension their personal morality.
What are you talking about? Nobody is being forced to accept the license. If you are writing your own code, you can use any license you want. If you are distributing gpl'd code, you have to adhere to the gpl. If you don't like the gpl, don't distribute gpl'd software. It's fairly simple, and everybody has the option to avoid the gpl if they so choose.
Microsoft tried to exploit a loophole in the gpl to play their little patent game. They fully deserve anything that might be coming to them. They knew gplv3 was being drafted and what it might imply well in advance of their patent deal. Also, the rules haven't changed for the particular versions of the software that they made the deal with (at least those without the "gplv2 or later" clause). Only new versions of the software licensed under gplv3 will cause problems for them. Quite frankly, the changes with gplv3 are a lot "fairer" than Microsoft has ever been with license changes in its own software.
I think you have hit the nail on the head. Vista may offer a few niceties, but XP is good enough for most people and they don't have a compelling reason to upgrade. It is the same problem linux advocates have been having for years. Linux offers some niceties over XP, but XP is good enough. Only the adventurous are going to run out and try a new OS just for fun. The majority are going to stick with what works until it no longer works.
First there are all of the linux distributions which, by the way, contribute a lot to free and open source software: Redhat, Novell, Mandriva, Ubuntu, etc... Then there are the MySQL, Apache, Samba, and Openoffice.org projects which have a lot of paid developers. Speaking of Openoffice.org, did you know Java and Solaris are now open source? Ok, let's see, there is CUPS, Xen, Cedega, and the QT toolkit. That's just off the top of my head. Yes, a lot of this is commercial software, or has a commercial offering, or is incorporated in some way into a commercial product...and yet it is still open source. Seriously, where have you been? Sure, there is a lot of closed source software, and there always will be, but a lot of companies incorporate open source software into their business models, pay developers, and make money off of it (granted in the latter case they aren't usually selling the software itself).
There are also organizations like OSDL that fund open source projects, like the Linux kernel. OpenBSD gets external funding to support sub-projects like OpenSSH. Many of the major Gnome developers are sponsored by companies. In the case of Redhat and Sun, they are paying for development of the desktop they use in their products, but there are other players as well. And there are companies like Dreamworks and Pixar that contribute developers to open source software because they actively use it to produce the product they actually sell, in their case movies.
In short, can you develop open source software and pay the bills? Yes. You may have to look around a little, but there are plenty of opportunities available. Not everybody wants to, so they don't seek those jobs, but that doesn't mean they don't exist.
Also, I can't find it right now, but there used to be a blurb on the GNU page talking specifically about this. Basically, without copyright law, there would be no need for the gpl to exist. It wouldn't be possible for software released under the public domain to be made proprietary. I know the original article made some sort of reference to using drm, but I can't really see how this would work, unless it is tied to a piece of hardware.
If I fail to protect my password, I cannot ethically or legally force others to stop spreading it.
Sure you can. Well, maybe not a password because those tend to be more volatile anyway, but how about your checking account number? I haven't studied law, but I'm sure you could make an argument in court against the malicious redistribution of your account number by somebody who managed to get a hold of it. Ethically, I'm not sure where the question lies. If you manage to get somebody's account number, and you know you shouldn't have it, and you willfully distribute it to all of your friends, you are being a jerk, plain and simple. Determining whether or not something is ethical can be subjective, but I think a few general guidelines apply. If you are willfully amusing or profiting yourself at the expense of another, you are acting unethically. You certainly have the right, in many cases, to choose to be unethical, but your behavior is still unethical.
Are you kidding me? All encryption keys attempt to prevent the "free exchange of ideas" to unauthorized third parties. Is OpenSSH now censorship software? Is copyright law censorship? In case you were wondering, the answer is no.
Ummm...it just isn't. This has nothing to do with suppression of the free exchange of ideas. The "seemingly random collection of hexadecimal digits", so quaintly put, is an encryption key. It is fairly disingenuous to claim that it is just an alphanumeric string like any other, passed around in the context of an ordinary conversation. Now, since we at Slashdot hate the MPAA and all of their DRM, we want to construe this as a freedom of speech issue, but it isn't. If you are going to argue that the unauthorized distribution of private encryption keys falls under your first amendment rights, then you probably won't mind if I just post your credit card number up here on Slashdot...I mean it's just an ordinary 16-digit number, right?
Just for the record, I am not a fan of DRM and I break it all of the time to watch dvds on my linux box, but this isn't censorship. The MPAA has every right to try to suppress distribution of this key. Is the effort pointless and futile? Yes. But they still have the right to waste their time and money trying.
Slashdot Mirror
What is with all of these UPS analogies?
The best analogies come from ninjas!
"No, I'm sorry, you can't integrate my small component into your giant proprietary and ITAR-restricted satellite system unless you agree to give away the entire code to your giant ITAR-restricted satellite system to anybody who wants it. But hey, RMS says that freedom is good, so you can do that, right?"
So, you're another one of those "I don't like the GPL, but I haven't actually read the license" type of people, eh?
Well, to save you the trouble of actually reading the license, I will point you to a quick summary on Wikipedia.
They are the ones who decided on this policy, and they claim a moral (yes, moral) high-ground because of it. But really, they only mean this within the context of their own ideological community. It's really a difference of attitude.
Nonsense. The GPL is a very pragmatic license. People who write and distribute their code decide how they want to allow it to be redistributed. There is no moral claim there at all. Don't confuse this with Richard Stallman's claim of the morality of Free Software, which is a completely different subject.
Anyway, the claim a lot of people are making here is that advocates of the GPL should contribute back to BSD-projects under a BSD-license because otherwise they are being hypocritical. I think that is fairly silly, because then you are basically saying, this is a BSD-project, this code can't be used by GPL-projects, and that is contrary to the BSD spirit. Either you care about how your code is redistributed or you don't. You can't say, "go ahead and use our code, we don't care what you do with it" and then a little bit later say, "except we don't want our code used in a GPL-project."
Of course, this whole discussion is a non-issue in the first place. The original commit was a mistake and has been fixed. The code derived from was dual-licensed under BSD and GPLv2, which explicitly allows it to be redistributed under GPLv2. The mistake was that the commit omitted the dual-licensing part, which a number of people on LKML noted, and was fixed.
Actually, that's coming too.
https://wiki.ubuntu.com/DisplayConfigGTK
Well, I'm not sure what the GP's printer is, but many of the Brother laser printers are Postscript printers. The problem isn't the printer, or CUPS, it is the way that printing is setup in Ubuntu. The config tools just don't setup the printer profiles correctly. I've found that if I bypass the printer config tool and just use the CUPS web-based tool, the printer works perfectly. Meanwhile, the last time I was playing around with Fedora, their printer config tool worked perfectly. This is really just an Ubuntu problem, although I'm not sure what exactly the problem is because I'm using the same Postscript driver with the same PPD for my printer no matter what.
I know a new printer config tool is in the works, but as far as I can tell it has been delayed through the last two releases, and it doesn't look like it will make it into Gutsy either. It seems there have been some hangups.
I'm not so sure it is wise to jump to such a hasty conclusion. The thing is, despite the marketing hype, hi-def just isn't that big yet. There isn't enough content, there aren't enough people buying hdtvs, and there just isn't a lot of interest. The PS3 hasn't been out for very long. Most likely the people who have bought it up until now (despite the lack of interesting games which won't be out until next year), didn't do it because they wanted the blu-ray player. Combine that with the fact that the current selection of blu-ray movies is fairly mediocre, and it isn't a surprise that number of players doesn't correlate well with movie sales. In a few years, when there is a larger selection of content, prices are lower, and there is generally more interest, you can be sure that the more ubiquitous player (don't know if it is going to end up being blu-ray or hd-dvd at this point) is going to have a huge advantage. And if you already have a PS3, why would you buy a hd-dvd player?
How many bands sell millions of songs per year?
Uhhh, not necessarily agreeing with the GPs post, but I think your numbers are fairly exaggerated...a million songs to support one member of a band? I don't think so.
Let's take this from a different angle. A song on iTunes costs $0.99. Let's assume for the sake of argument that the royalty rate to the band is about the same as for cds (around 10% unless you are popular enough to negotiate higher). That is $0.10/song. Also, remember that the band has to pay back the label for the overhead costs, so they are getting even less than that, to split up between all of the band members. The other $0.89 doesn't go to overhead costs, it goes right into the pockets of the labels. So when the GP is saying he will pay $0.10/song, I'm sure (or at least I hope) he means $0.10 directly to the artists, which is comparable to what a lot of them get now (maybe even a little better). In other words, cut out the middle man and there is no need for prices to be that high.
The problem I have is that like many here on slashdot I do happen to know a thing or two about computers.
But, do you know a thing or two about science? Exactly what are you looking for when you say you want more substantial evidence. A scientific model is a scientific model. The validity is going to depend on the assumptions, the measurements to check those assumptions, and the data that is fed into the model. Whether or not it is done on a computer has nothing to do with its validity, although computers are arguably more capable of dealing with things that require massive amounts of computation, like solving the complex differential equations associated with climate models.
What I do see very clearly is that all suggested solutions seek to punish the united states while leaving some other areas of the world untouched.
Ok, that's a nice political slant to put on things. I'm not sure what you are specifically referring to here, but if you are concerned about carbon emission caps and such, the United States (and Canada) consumes significantly more energy per capita than other comparably developed nations around the world. That contributes to significantly larger carbon emissions. To reach sustainable levels, the United States has to reduce more because it is producing more. You can say that is "punishing the United States", but that is really irrelevant. If the United States had a more conservative energy usage policy to begin with, it wouldn't be in this situation. However, for many reasons including the heavy influence of the oil and coal lobby, that is not the case, and so here we are.
That was never it's purpose. The purpose of DRM was to curb copying,
The problem with that statement, and I'm surprised nobody has pointed it out yet because a lot of other comments are saying the same thing, is that DRM does nothing to prevent copying, not even a little bit. I can freely copy encrypted dvds and encrypted music files to my computer, to my friend's computer, to my portable music/video player, anywhere. The only thing DRM does is require an "authorized player" to watch the video/play the music. As long as I have an authorized player (ex: PowerDVD), I can play my copied and encrypted dvd. The authorized player can, of course, try to enforce it's own copying restrictions (ex: iTunes) by refusing to decrypt content it doesn't think you own or have the rights to, but this is the player and not the DRM. Massively distributed players that need to be able to play "random dvd off the shelf" can't enforce any type of copying restrictions, so you are left with this elaborate and expensive encryption scheme that ultimately doesn't do what it is supposedly designed to do, which is prevent copying. But, of course, that's not what DRM was really designed to do...I'll leave the rest of that thought to the reader.
And creates ambiguity between GNU Projects and Linux
I think the slash fairly clearly indicates that it is GNU slash linux (ie: a mixture of two projects), and not GNU Linux. The Debian doc you have linked to is over 10 years old. If you look at the front page of their website, you will see that all references to the distribution are denoted as Debian GNU/Linux.
I assume you mean: "adopted the project and copyright as a GNU project".
No, what I mean is the FSF starts a GNU project, say GCC. They have written a large chunk of code at that point and own the copyright on it. Now if the project is managed like the linux kernel, anybody who contributes code has to license it under the GPL, but they retain their copyright to the code they have written. The FSF, however, asks all contributors to transfer their copyright to the FSF if they want to be a part of the project. Other projects are known to do this as well: CUPS, MySQL, OpenOffice.org, and Apache are the few that come to mind right away.
Once you sign over the copyright, its not your project anymore
You do realize you can transfer your copyright on some code and not other code, right? Also, the terms of a copyright transfer will vary, but the original developer almost always retains unrestricted rights to his code. He just ends up granting unrestricted rights to the third-party as well.
As for needing the copyright to enforce the GPL, after using XEmacs code in GNU Emacs under the GPL they're perfectly capable of enforcing the GPL on that code because the derived works are covered by the GPL v2.0 section 2 - b; unless they don't believe that clause is enforcible.
You cannot prosecute GPL violations without money and legal expertise. The FSF has both. The FSF is more than willing to prosecute on your behalf, but they can't do that if they don't own the copyright.
Ok, one more try...
That credit doesn't extend to kowtowing to RMS's ego and renaming software because he says so.
RMS hasn't said anything about renaming the linux kernel project. He just wants mixtures of Linux and GNU software to be referred to as GNU/Linux. That does not seem unreasonable to me. It doesn't hurt anyone in any way to do that.
is attempts to rebrand Linux as GNU/Linux are effectively trying to brand Linux as a GNU package
No, I think you are letting a dislike of RMS distort your opinion. He is not rebranding the linux kernel. He is not trying to take over anything. He just wants people to know, from the name, that a linux distribution contains software in addition to the linux kernel that is completely separate and has nothing to do with the brainchild of Linus Torvalds. If I wanted to be pedantic, I could say that using the name Linux for the distribution is an attempt to rebrand the GNU software that is included and make it appear like it all came from the same source when it didn't. The GNU project existed long before Linux, after all. But, I think the argument is stupid, so I won't.
Transfer of copyright is transfer of ownership or "take over".
I think you missed (or just ignored) the part where I said why they did this and that other projects do the same thing. If I start a project and you want to contribute code to it, I can ask you transfer your copyright to me. You can, of course, refuse and and in that way I would be excluding a lot of potential developers from the project, but if I want to make that requirement it is within my rights. It's not "taking over" another project because it was my project to begin with. RMS has only asked for transfer of copyright on contributions to GNU Emacs, not XEmacs. Now the naming issue mentioned in the FAQ, I do agree is silly. If XEmacs is indeed largely independent code (I don't know because I haven't looked), RMS doesn't have a claim to the name. Also, it seems RMS has backed off because the emacs FAQ references XEmacs as XEmacs, not GNU XEmacs. Other projects on the GNU website do as well.
Its there, its GPL, nothing prevents you from doing so.
Well, I asked whether it was ok, not whether it was legal. I know it is legal, but I'm not sure I am ethically ok with taking large chunks of another project and not at least mentioning that my project is derived from a different project. I suppose a better question is, would you take the linux kernel and credit Alan Cox for his contributions, but not Andrew Morton?
What happened with GNU and XEmacs says something completely different.
Not sure what you are referring to here. If you are referring to the debate over copyright assignment, the FSF asks all GNU projects to transfer their copyright so they are better able to prosecute GPL violations. The FSF is certainly not the only organization to do this. Easy Software Products and the CUPS project is a recent example that has received attention lately due to the buyout from Apple. The contention between XEmacs and GNU Emacs is that XEmacs occasionally wants to contribute code back to GNU Emacs, but the FSF wants the copyright transferred. It's a conflict between the two projects, yes, but I think suggesting that the FSF is trying to "take over" XEmacs is a deliberate misconstruing of the actual situation. And what does this have to do with the linux kernel? Where is the posting on LKML where RMS asks all of the kernel developers to transfer over their copyrights to the FSF?
So? It doesn't matter what his motive is. He did the work, he can rightly demand credit for it.
So, if I took the linux kernel and rewrote the SCSI subsystem, and then rebranded and distributed it as Rutulian's superkernel, that would be completely ok with you?
What about the fact that branding it GNU/Linux is effectively an attempt to make it part of GNU?
No it's not. 'nuff said
One could perfectly well have a Linux system without Gnu tools, and no, it wouldn't be useless. Embedded systems often run a single application and this could be done with tools other than Gnu's.
Yes, and in that case you wouldn't call it GNU/Linux. In fact, they don't really call it anything most of the time because most people don't really care about the operating system of their embedded devices as long as it works. The point isn't that Linux can or can't be packaged with other things. The point is that the linux kernel by itself is not an operating system (neither is GNU for that matter).
Linux uses Gnu for the simple reason that Gnu existed, if it didn't exist it wouldn't be very hard to write. A lot of work, yes, but not as difficult as writing an OS kernel, the hurd is there to prove that.
No, I don't think that is true at all. Writing a kernel is hard, but writing a compiler and toolchain isn't easy either. The Hurd has been slow to get off the ground for a lot of reasons. The difficulty of the task is really only one of them.
The Gnu project needs Linux more than Linux needs Gnu.
That's a pretty stupid statement to make. It's like saying strawberry shortcake needs the shortcake more than the strawberries. No, they both need each other equally. You can package Linux with BSD or GNU with Solaris, but very few people do. Most people package GNU with Linux. Seriously, why do people fight this. It's just a name. RMS wants credit for his part in the creation of the modern linux distribution. Why is that so difficult for people to accept?
For the record, I never refer to my operating system as GNU/Linux because it is tedious, and I prefer to just shorten it to Linux. But it is understandable for RMS to request that people refer to it as GNU/Linux. And, if I was ever writing or speaking about it in some sort of official capacity, I would probably extend him that courtesy.
Yeah, sure, I suppose that is a legitimate fear...although Red Hat is actively working on their own management tools at the moment, and it will probably be a mute point in a year or so. Now that I think about it, though, I don't think I have ever known Red Hat to take another distribution's management or config utility and incorporate it into their own. Usually it is the other way around. Red Hat uses the Gnome System Tools package, but their printing and all of their server config tools are their own. Suse has had Yast and Mandrake has had the Drak tools for quite a while, but Red Hat hasn't touched them, even though some of them are superior to the ones Red Hat uses.
Also, what Xen fears they can also use to their advantage. Take Red Hat's distribution and all of their work patching the kernel, customize it to be a drop in hypervisor with easy config and management tools, and sell it with support options. I'm sure there are people out there looking for turn-key virtualization solutions. Remember, open source allows the free exchange of ideas, which means ideas can flow both ways. Red Hat can take from Xen, but Xen can also take from Red Hat.
What I called hypocritical was to do so in the name of "freedom", while you're at the same time trying to argue that I don't have the "freedom" to make my own choice.
Yeah, well, if they compile my code and sign the binary and make their hardware run only the signed binary, they have restricted my freedom to do what I want with my code on hardware I have purchased.
If I put my code under the gplv3, thereby requiring them to distribute their signing keys with my code, I have preserved my own freedom, but impinged on their freedom to prevent modified code from running on hardware they have created (unless, of course, they pay me or somebody else for that freedom).
So who's freedom is more important?
I lean toward my own freedom because I have taken the time to write the code and am offering it up for consumption by the community. Companies should have freedom too, but they don't get free labor. There is the freedom of ideas, but there is also economics. Nothing in the business world is free. If they don't want to pay me by adhering to the terms of my license, they need to pay me in cash, or they need to pay somebody else to develop for them.
Tivo's video-processing code can still be used in other applications, after all.
That's a good point, but what if I want to add a new dithering algorithm and run it on my Tivo? Actually, let's take a different tack. I write general networking stack. Company makes a networking appliance and uses general networking stack to run it. I realize later that there is an exploitable bug in my code and write a patch. Now, can I use my patched software with the networking appliance I purchased, or do I have to wait for the company to get around to compiling the new version and distributing signed binaries?
Furthermore, doesn't the GPLv3 prohibit *anybody* from writing GPLv3 code that runs on a Tivo
No. I'm not sure where you got that idea. I suggest actually reading the GPL. Specifically the part in the preamble that states the terms of the GPL apply only to those who distribute the software. If you own the copyright to the code, you can do whatever you want with it.
Linus is, of course, entitled to his opinion. But, seriously, the new provisions of the gpl don't hurt anyone. You have to remember, companies that wish to distribute gpl'd software are still getting a lot of work done for free (which saves them a lot in development money). Giving back modifications and the right to use modifications free from patent and hardware restrictions is a fair trade. It's not like they are requiring everybody to contribute $5 to the FSF legal fund for every binary distributed. If a company doesn't like the terms of the gpl, they can do as they have always been able to: find a differently licensed project with similar functionality, write their own code, or pay the developers for the right to distribute the code under an alternate license.
What limit on hardware manufacturer's? There is no limit on hardware manufacturer's. The only requirement is that if your hardware will only run signed code, the means to reproduce that signed code must be distributed with the software (which is distributed with the hardware). You can do whatever you want with your hardware, but if you use and distribute (i.e: make money selling) my software with your hardware, you can't prevent me from modifying (ex: with an newer version that adds extra features) said software and running it on your hardware (or rather, the hardware can prevent it, but you have to provide me with the means to make my software conform to what the hardware will run). It seems like a fair trade to me, and is not entirely unreasonable. The only people who see it as unreasonable are the hardware manufacturer's who want to use free code and not give anything back to the original developers.
Well, there's not a lot of context in the article, but piecing things together, I'd guess that what has Linus ticked off is not injecting personal morality into the license, but forcing others to accept that license and by extension their personal morality.
What are you talking about? Nobody is being forced to accept the license. If you are writing your own code, you can use any license you want. If you are distributing gpl'd code, you have to adhere to the gpl. If you don't like the gpl, don't distribute gpl'd software. It's fairly simple, and everybody has the option to avoid the gpl if they so choose.
Microsoft tried to exploit a loophole in the gpl to play their little patent game. They fully deserve anything that might be coming to them. They knew gplv3 was being drafted and what it might imply well in advance of their patent deal. Also, the rules haven't changed for the particular versions of the software that they made the deal with (at least those without the "gplv2 or later" clause). Only new versions of the software licensed under gplv3 will cause problems for them. Quite frankly, the changes with gplv3 are a lot "fairer" than Microsoft has ever been with license changes in its own software.
I think you have hit the nail on the head. Vista may offer a few niceties, but XP is good enough for most people and they don't have a compelling reason to upgrade. It is the same problem linux advocates have been having for years. Linux offers some niceties over XP, but XP is good enough. Only the adventurous are going to run out and try a new OS just for fun. The majority are going to stick with what works until it no longer works.
Are you kidding?
First there are all of the linux distributions which, by the way, contribute a lot to free and open source software: Redhat, Novell, Mandriva, Ubuntu, etc... Then there are the MySQL, Apache, Samba, and Openoffice.org projects which have a lot of paid developers. Speaking of Openoffice.org, did you know Java and Solaris are now open source? Ok, let's see, there is CUPS, Xen, Cedega, and the QT toolkit. That's just off the top of my head. Yes, a lot of this is commercial software, or has a commercial offering, or is incorporated in some way into a commercial product...and yet it is still open source. Seriously, where have you been? Sure, there is a lot of closed source software, and there always will be, but a lot of companies incorporate open source software into their business models, pay developers, and make money off of it (granted in the latter case they aren't usually selling the software itself).
There are also organizations like OSDL that fund open source projects, like the Linux kernel. OpenBSD gets external funding to support sub-projects like OpenSSH. Many of the major Gnome developers are sponsored by companies. In the case of Redhat and Sun, they are paying for development of the desktop they use in their products, but there are other players as well. And there are companies like Dreamworks and Pixar that contribute developers to open source software because they actively use it to produce the product they actually sell, in their case movies.
In short, can you develop open source software and pay the bills? Yes. You may have to look around a little, but there are plenty of opportunities available. Not everybody wants to, so they don't seek those jobs, but that doesn't mean they don't exist.
Also, I can't find it right now, but there used to be a blurb on the GNU page talking specifically about this. Basically, without copyright law, there would be no need for the gpl to exist. It wouldn't be possible for software released under the public domain to be made proprietary. I know the original article made some sort of reference to using drm, but I can't really see how this would work, unless it is tied to a piece of hardware.
If I fail to protect my password, I cannot ethically or legally force others to stop spreading it.
Sure you can. Well, maybe not a password because those tend to be more volatile anyway, but how about your checking account number? I haven't studied law, but I'm sure you could make an argument in court against the malicious redistribution of your account number by somebody who managed to get a hold of it. Ethically, I'm not sure where the question lies. If you manage to get somebody's account number, and you know you shouldn't have it, and you willfully distribute it to all of your friends, you are being a jerk, plain and simple. Determining whether or not something is ethical can be subjective, but I think a few general guidelines apply. If you are willfully amusing or profiting yourself at the expense of another, you are acting unethically. You certainly have the right, in many cases, to choose to be unethical, but your behavior is still unethical.
Are you kidding me? All encryption keys attempt to prevent the "free exchange of ideas" to unauthorized third parties. Is OpenSSH now censorship software? Is copyright law censorship? In case you were wondering, the answer is no.
How is that not censorship?
Ummm...it just isn't. This has nothing to do with suppression of the free exchange of ideas. The "seemingly random collection of hexadecimal digits", so quaintly put, is an encryption key. It is fairly disingenuous to claim that it is just an alphanumeric string like any other, passed around in the context of an ordinary conversation. Now, since we at Slashdot hate the MPAA and all of their DRM, we want to construe this as a freedom of speech issue, but it isn't. If you are going to argue that the unauthorized distribution of private encryption keys falls under your first amendment rights, then you probably won't mind if I just post your credit card number up here on Slashdot...I mean it's just an ordinary 16-digit number, right?
Just for the record, I am not a fan of DRM and I break it all of the time to watch dvds on my linux box, but this isn't censorship. The MPAA has every right to try to suppress distribution of this key. Is the effort pointless and futile? Yes. But they still have the right to waste their time and money trying.