Did you happen to notice that Christians are speaking out against this? There are crazy and thoughtless people in all realms of life. Condemning a very large and diverse group of people due to the actions of a few is fairly ignorant (and no, I'm not Christian).
Also, serious question, how is this any worse than Grand Theft Auto?
Ummm...are you kidding? Those icons haven't been updated in years. They're still using the old crappy Gnome 1.x foot logo when the new artwork for 2.x has been around for, let's see, 4 years?!!
I'm sorry, I AM a K-12 admin for a fairly large school system (10,000 desktops) and we use Windows for several non-linux bashing reasons: Exchange, AD, compatibility with other districts, and price/support to staying the course as opposed to rebuilding everything.
Yeah, those are good reasons, the same ones you will encounter everywhere. There's no reason to switch from a working system to a different system unless it is going to be of some real benefit (price, features, performance, etc). If it were my network, I would probably choose to run servers and such with linux (Windows for Exchange, but Samba can do AD), and have mixed labs with linux/bsd/macos/windows. I think you get more value out of your network if the infrastructure can support more diversity. The problem with Windows-only networks is that it is really difficult to hook up a linux box to do certain things, should you want to. If you plan accordingly, though, say by extending your AD schema to support unix-like uids and gids, you can easily hook up a few linux workstations with domain logins to support those who might want to use linux. Once it becomes possible for people to switch, then you will start to see people switch, gradually. A sudden dump from Windows-only to linux-only, though, is almost always going to be a bad idea and a difficult transition.
But, then microsoft would just starve the BIOS and MOBO makers of marketing dollars. Sure, make it compatible/supportive of Vista, but them mshaft will just make the OS kernel check the BIOS maker ID-- IDs assigned by mshaft. As much as I'd love to see the hardware more Linux-friendly, I have no doubt it'll still be some time off.
Wait, I missed the part where you explained why exactly MS would want to do this? Microsoft isn't in the bios business. Why would they care who writes the bios?
In general, it would add some credibility to your position if you understood what the people writing the code actually care about.
I understand that some of the kernel developers don't care about the freedom of free software in the same way that the FSF does. If they just want to code and release it for free, fine. There is nothing wrong with the BSD license, and they certainly have the right to choose whatever license they want. All I'm trying to point out is that a) they do care about freedom to some extent because they initially chose the GPL and not the BSD license; b) there seems to be an irrational opposition to GPLv3 that probably has something to do with dislike for Richard Stallman and leftover bad feelings from the BitKeeper incident; c) that GPLv3 is intended to protect the freedom in free software that they obviously do care about to some extent (see debates over NVidia's binary kernel driver for an example); and d) I have yet to read/hear any kind of rational and logical reason to not adopt GPLv3 if you are already using GPLv2.
Calling them "ignorant wankers" sure sounds like an ad-hominem attack to me.
Then maybe you should pay attention to the context. I am claiming that they are ignorant of the issues they are trying to address with their statement, which still appears to be true, and they are blowing a lot of hot air without first trying to understand the purpose of the object of their irrational dislike. It's pretty simple, if I don't think they know what they are talking about, I'm not going to respect their opinion. Likewise, a kernel developer is not going to respect criticism of their code from someone who has never contributed a patch. That's not ad hominem, it's dismissing the peanut gallery.
So you acknowledge that your point about "DRM and patent issues" is "not useful discourse"...? Fine.
Well, I don't recall making a point about "DRM and patent issues." I also don't recall claiming to be having discourse. I'm simply stating that the reasons given for opposing the GPLv3 are nonsense. I don't really want to pick through and respond to each one because it has already been done numerous times elsewhere. Nice try, though.
The GPLv3 license has not been finalized, but the proposed no-DRM clauses would certainly appear to forbid the combination of GPLv3 code with the OpenSSL library, which is used for PKI, OpenSSH's authentication, x.509 certs used by SSL webservers, and so forth.
Come again? I'm not aware of DRM implementations in any of the things you have mentioned. I believe you are confusing a couple of different things. Certificate signing for public key authentication purposes is not even addressed by the GPL and has nothing to do with it. Code-signing is addressed by the GPL, but the anti-DRM clause only comes into effect if it is impossible to run unsigned code (referred to as tivoisation). Seriously, do you really think RMS would add a clause to the GPL that would prevent the use of cryptography? Richard Stallman may be a lot of things, but he isn't stupid. There is, apparently, an incompatibility between the GPL and the OpenSSL license (note: not functions used in the OpenSSL library). This is resolved, not exacerbated, in GPLv3 via the additional restrictions clause.
Just how much of your code is in the Linux kernel today? If the answer is zero, well, you might want to contribute something more useful than a broad ad-hominem attack against the people who actually are writing Linux kernel code.
What does my contributing to the linux kernel have to do with a debate about licenses? The kernel developers are brilliant when it comes to writing kernel subsystems, and I wouldn't argue with them about the best way to write a scheduler algorithm...that doesn't mean they are equally knowledgable about everything else. This isn't an ad-hominem attack. The reasons they have given for their position are nonsensical and indicate they are largely ignorant of corporate politics, intellectual property, and the freedom in free software that the FSF is trying to protect.
You bet. Obviously the issue is so controversial that wide agreement has not been reached, and may not ever be reached.
Sure, but rehashing the same arguments over and over that have already been addressed in other contexts is not useful discourse.
For the GPLv3 to be usable for general-purpose operating systems, it needs to not forbid mechanisms like/etc/passwd and/bin/login from checking whether a user has a valid password, and it needs to not prevent systems like OpenSSH or OpenSSL from being used with GPLv3 code.
Ummm, what are you trying to say here? That a GPLv3 networking library can't link to OpenSSL? Nonsense. I think you are being deliberately vague because you don't know what you are talking about. How about providing a specific example of a case where GPLv3 code can't be used with common mechanisms like/etc/passwd. I can't agree with or refute vague assertions.
There are plenty of companies who like the GPL just fine; try asking MySQL
Completely irrelevant. Yeah, Red Hat likes the GPL too, but we aren't talking about open source companies. We are talking about embedded companies, mainframe companies, even proprietary software companies...companies that use things like the linux kernel, or the GNU toolchain, or the OpenSSL libraries because they can customize it as they wish and it saves them in developer time and money. These companies adhere to the GPL because they have to (for both legal and good karma reasons). I'm quite sure they would be happy without the provisions that require them to publish their changes. In fact, companies that use the linux kernel get nailed for trying to weasel out of it quite often (at least once a year). Having a stricter license that closes loopholes may drive a few companies away, but it is not going to cut off corporate support of open source development.
Agreed. This position statement seems to be, "We don't like politics and GPLv3, so here are a bunch of bs reasons for opposing the license change." Not liking politics is fine, but if you are going to draft a position statement you need to make sensible arguments and not just sound like ignorant wankers. The DRM and patent issues have been addressed by the FSF a number of times. This is not about a war against DRM and patents; it is about free software being free. Oh, and corporations may not like it, but since when do corporations like any version of the GPL? I'm sure they would really prefer the BSD license, so should we just switch to that to keep our corporate friends happy. The thing is, corporations want to use linux. They like linux, and they will put up with the license to use linux because, in the long run, it saves them money, even if they have to give something back to the community in return.
No, the contributors still maintain copyright, they just have to grant an unlimited, unrevokable, license to the project (or company) that is compatible with the planned distribution model. So if, say, a company wanted to distribute a GPL free version and a binary non-free version to paying customers, contributers would have to grant a license to the company to distribute the work in those ways. They don't have to transfer ownership rights, they just have to grant redistribution rights. It is no different really than contributing to something like the Linux kernel where you have to agree to use a license that is compatible with the rest of the project (GPL v2) if you want to get your code in the main tree. ReiserFS is still owned by Namesys even though it is in the main kernel tree and has been patched/maintained by a number of other people since its inclusion.
With transfer of ownership rights, Digium could choose to eliminate the GPL version and require everybody to purchase the non-free version. Not saying Digium would actually do that, just that it is possible. With the Linux kernel, it is impossible for Linus to change from the GPL license to, say, the MPL license. He would have to get the agreement of all of the contributors or replace their code.
Well, there is the MPL, although the FSF considers it to be a "weak copyleft" license. I can't think of any legal reason a project would have to take ownership rights of contributed code in order to distribute under a dual-license. The FSF requests contributions to GNU projects be transferred to them for license enforcement reasons, but they don't require it.
Not sure about mysql and trolltech (I think they are mostly developed in house, actually), but Apache uses the Apache license which allows for non-free distribution of the code. The contributors have to license their contributions properly to get them accepted into the main code base, but they don't have to give up their ownership rights.
although that does bring up another potential thing to trip up new users... too much choice.
Sure, but the default Ubuntu installation does Gnome, so there is no choice. You have to specifically download Kubuntu if you want the KDE desktop.
about the system volume...
I read both of your comments...I still have no idea what you are talking about. What do you mean by "doesn't remember the system volume?" If it doesn't know where the root partition is, that's a bug in the installer that needs to be fixed. The only time I have ever had boot loader issues is when I am installing on SCSI drives. This is really a GRUB issue. I know other people have the same problem, so hopefully it will be fixed soon. But I don't think that is what you are talking about because Macs haven't had SCSI drives for quite a while.
about the overly complex system settings- I'm mostly talking about KDE's system settings.
Yes, this is my #1 complaint about KDE, which is why I don't use it. Some people seem to like it that way, so let them keep it. As long as I have a simple, uncluttered Gnome interface, I am happy.
the installer really should have a streamlined install.
I agree with having a streamlined install, but I think the Ubuntu install is pretty good. I don't see how it could really get any easier. I was able to go from a Windows installation to a dual-boot configuration in about 30 minutes with a friend's laptop (auto-resizing of partitions and everything). That's pretty good. The advanced options are left in because some people, like me, don't want the installer mucking around with the partitions. Most people can select the "automatically partition free space" option. This could still confuse some people, but it is easy enough that they could follow somebody's clear instructions.
So, uh...yeah. With all of the talk about the War on Terror these days, what is being done about this kind of crap. Isn't the instillment of fear with threats of violence to obtain political victories terrorism? I don't see GW talking about terrorism that happens in our own country (you know, by Americans) in the news....
I have no problem with closed hardware, or software. The problem comes when closed hardware (or software) companies use open source software to save on development time and then don't give back to the community. Sorry, letting me look at code I can't run may be following the letter of the license, but it doesn't follow the spirit. Companies, being amoral entities, only care about the letter, though, which is why the license needs to be modified to close loopholes.
But the vendor must still publish the source of any changes he makes to the code. So the vendor is giving back.
No, it's not. The device driver code for an obscure chip on, say, a wireless router is completely useless if I can't run the code on said hardware. Think about the Linksys WRT54G. It was just running a linux kernel modified to run on the router with a set of minimal networking utilities. When the source was released, people were able to add all sorts of stuff: better firewalls, servers, ssh utilities, more efficient and capable routing.... They wouldn't have been able to do any of this stuff if Linksys had tossed in a TPM chip and made it so only their signed modified linux would run. The changes they made were also useless without the hardware (yay, so now we know how to run linux on a Linksys router, to bad we can't actually do it).
So, yes, a consumer could just go buy a different wireless router, but that's not the point. The point is that Linksys would have been using GPL'd software to reduce their development costs without giving a useful share back to the community. Developers who don't mind that sort of thing use the BSD license. Developers who do use the GPL. For the GPL to continue to be relevant, it has to be modified to close loopholes that didn't exist 10 years ago.
The way Linus sees it is from the "developer" viewpoint. The code is still free from this viewpoint, since all modifications are published. You can modify it and run it on a DRM-free machine.
That's the short-sightedness of Linus' argument (the same short-sightedness that let him get trapped by the Bitkeeper fiasco). There are DRM-free machines now, but that doesn't mean there will be in the future. If the media companies have their way, every desktop computer will have a TPM chip in it, and if you want to view things like HD-content, it has to be enabled and running. So a company can take a project, like say mplayer, make a version that plays their video format, decrypting the stream via the TPM hardware, and then sign the binary and then sell it. Congratulations, a company has just saved themselves a couple of years of development time to make a video player to help sell their video files, and you can't modify it at all. If you modify it, it reverts to just plain old mplayer without the ability to use the code that was added. That defeats the purpose of the GPL. Linus really needs to wake up here. Yes, proprietary software has a right to exist, but pretending a company won't take advantage of free software to reduce their development costs (without giving anything back, if they can) is stupid. The GPL allows commercial use of free software as long as you give a fair share back to the community. It is not some fiendish scheme to force all software to be free as some people would say. The GPL as it is has worked fine for the last decade, but now it needs to change or it will no longer serve its purpose.
I'm not so sure they'll be able to pull this off, though. Limewire is a Gnutella client. Everything is distributed. It isn't like Napster where they had a centralized search engine to find pirated music. The only thing Limewire does is implement the Gnutella protocol, which means you can use it to search for any shared files. Even if the Limewire people wanted to block copyrighted material, there isn't a feasible way for them to do it. Block by filename and people will just change it to something else. Prevent searching for certain search terms and people will start to use secret keywords. There's just no way for a blacklist to keep up with attempts to circumvent it. So then you could sample individual files looking for signatures, but that a) would slow things to a crawl, b) probably wouldn't be resistant to changes in encoding/bitrate. So in addition to just mp3 signatures, you would have to check ogg/wav/flac/aac.... And if you ever took it that far, people would just start hiding things in archives or using encryption. Blocking files or searches just isn't going to be possible.
As for legitimate uses of Limewire, there are plenty: shareware, freeware, Creative Commons licensed stuff, samples/demos/free trials, not copyrighted stuff...now you can argue that Limewire isn't used for these things 90% of the time, which is probably true. But Limewire has a pretty clear defense: software was built to promote legal filesharing, software accomplishes this goal (see random statistic showing legally shared files), yes illegal filesharing happens, but there is nothing we can do about that above and beyond what we are already doing (putting strong warnings everywhere and allowing content owners to submit a SHA1 hash of files they don't want distributed).
Don't be so sure. Simple convenience features like fast servers, good bandwidth, quality encodings, choice of encodings, easy searching, etc...can go a long way, and people will pay for it even if they could get the same thing for free (see just about every commercial linux distribution). They wouldn't be able to get away with charging $1/song without copyright/drm, but a lot of people (me included) are willing to pay $0.20/song or so, or maybe even a flat fee of $10/month.
Yes, that is exactly what he is saying. Distributors have a fundamental right to include copyright protection, and consumers have a fundamental right to break that copyright protection on media they have bought. That doesn't mean they have a right to then break copyright law by giving copies to all their friends, but if, say, copyright law allowed somebody to make additional copies for their own personal use (*cough* fair use *cough cough*), they would first have to break any copyright protection before they could do that. The DMCA makes circumventing copyright protection illegal, which basically allows distributors to define their own copyright law anyway they please (because any unauthorized use or manipulation of the content is illegal).
Is it killing off bacteria we want to keep around? Maybe. Is it creating an environmental pressure for the evolution of Triclosan-resistant bacteria? Definitely.
I have to argue No and No to these questions, unless you are talking about a hospital or something. By the time the already low concentration of Triclosan in soap reaches the sink, it is diluted past its efficacy. Add to that that it takes time before it can do its job, as another replyer noted, and the Triclosan in antibacterial soap is completely useless for killing off bacteria. The detergents in soap are already doing a much better job.
There is also a fairly common misunderstanding about how antibacterial-resistance develops. Most bacteria replicate quickly and can therefore mutate quickly to adapt to adverse conditions (ex: an antibacterial agent). However, bacteria will revert just as quickly when mutant phenotypes are no longer necessary. Since most mutations that confer antibiotic or antiseptic resistance involve the expression of additional genes and consequently a slower dividing time, bacteria will revert in the absence of a selective pressure. People who work with antibiotic-resistant strains of bacteria in labs observe this all the time. That is not to say that antibiotic-resistance isn't a problem...it is. But squirting a little antiseptic-containing soap down your drain every once in a while is not likely to be contributing to the completely unrelated problem of antibiotic resistance (ex: vancomycin resistant S. aureus in hospitals).
3. Society in general and the marketing departments of various household goods: all of the various soaps & cleansers which promise to kill bugs when you use them. You're only supposed to use soap to clean your hands off - remove the stuff which doesn't belong there - remove as in get it off of your hands, not kill some of the bugs and leave a small number of immune ones in place. Eugenics takes over and we begin breeding superbugs.
Antibacterial soaps are a marketing ploy and nothing more...all soaps are antibacterial. How well a given soap removes bacteria from your hands is directly proportional to how well you clean your hands (i.e: do you just get them wet, or do you really soap up and scrub them down). Bacteria aren't some magical things that can survive the same conditions that will remove dirt, grime, oil, protein, salt, and metals from your hands. Some companies throw in a little bit of antiseptic to get people to buy their soap, but it is no more or less effective than regular soap, and at the concentrations present, it is highly unlikely to cause resistant bacterial strains to develop.
As a current and past employee of several companies that make wireless transceivers subject to FCC licensing, I can tell you that there is no cost effective way to limit a device to FCC restrictions purely in hardware.
Um, sure. But I don't think Theo cares about it being "pure hardware" or a hardware/firmware mixture. As long as it isn't in the software driver, wifi hardware companies can't claim that releasing source is against FCC rules. And, as the parent was saying, it probably isn't anyway.
I wasn't going here for support, I was making the observation that ext2 was fragile, and that I had lost data from it. And immediately, it was all my fault.
I was saying, and rightly so, that Linux wasn't ready for business yet, and wouldn't be until it had a better filesystem. This was approximately like pouring blood in the water; the sharks showed up minutes later.
After ext3 and Reiser went mainstream, Linux was finally ready for primetime... and then everyone agreed that ext2 was really awful.
But that's just it, ext2 isn't awful. And no I'm not a linux zealot (although I am pro-linux). ext2 lacks one critical feature for high availability...journaling. Everything else about ext2 is great: it is fast, efficient, stable, extensible, portable, and it auto-defragments. You couldn't say the same thing about reiserfs until later versions. So, yeah, reiserfs had journaling, but it was still slow and unstable and nobody wanted to touch it. In the early years, a business could easily make use of linux with ext2 if they acknowledged the risks and took the right precautions (ex: a UPS).
If people like a particular operating system, they will tend to overlook its weaknesses in favor of its strengths, but that doesn't make them zealots. They have just decided that the strengths make the OS good in spite of its weaknesses. And, naturally, when another weakness is checked off the list, they can look back and go "Wow, we've come a long ways. Yay linux!"
I like how nobody in this thread, including the parent actually tried to investigate the problem. mod_security is gpl'd and available in Debian as libapache2-mod-security, so wtf are you talking about? Next time try to actually look for the packages before claiming that they don't exist.
In general, though, Debian already does what you ask for. It has a main distribution, for software that meets the Debian Free Software Guidelines, and a non-free distribution for other software. If Debian doesn't distribute something (officially), there is probably a serious legal hindrance, and other distributions like Red Hat probably don't distribute it either. However, also like Red Hat, there are unofficial repositories you can use to get these packages if you don't care about the ethical/legal issues. I don't really see why this should or how this could be easier. Anything more automagical, like/etc/allow-packages, would be a legal problem for Debian to support officially.
Slashdot Mirror
Did you happen to notice that Christians are speaking out against this? There are crazy and thoughtless people in all realms of life. Condemning a very large and diverse group of people due to the actions of a few is fairly ignorant (and no, I'm not Christian).
Also, serious question, how is this any worse than Grand Theft Auto?
Whoops, I take that back. It appears they did finally update the logo....
0 7
http://slashdot.org/article.pl?sid=06/09/07/02402
Ummm...are you kidding? Those icons haven't been updated in years. They're still using the old crappy Gnome 1.x foot logo when the new artwork for 2.x has been around for, let's see, 4 years?!!
I'm sorry, I AM a K-12 admin for a fairly large school system (10,000 desktops) and we use Windows for several non-linux bashing reasons: Exchange, AD, compatibility with other districts, and price/support to staying the course as opposed to rebuilding everything.
Yeah, those are good reasons, the same ones you will encounter everywhere. There's no reason to switch from a working system to a different system unless it is going to be of some real benefit (price, features, performance, etc). If it were my network, I would probably choose to run servers and such with linux (Windows for Exchange, but Samba can do AD), and have mixed labs with linux/bsd/macos/windows. I think you get more value out of your network if the infrastructure can support more diversity. The problem with Windows-only networks is that it is really difficult to hook up a linux box to do certain things, should you want to. If you plan accordingly, though, say by extending your AD schema to support unix-like uids and gids, you can easily hook up a few linux workstations with domain logins to support those who might want to use linux. Once it becomes possible for people to switch, then you will start to see people switch, gradually. A sudden dump from Windows-only to linux-only, though, is almost always going to be a bad idea and a difficult transition.
But, then microsoft would just starve the BIOS and MOBO makers of marketing dollars. Sure, make it compatible/supportive of Vista, but them mshaft will just make the OS kernel check the BIOS maker ID-- IDs assigned by mshaft. As much as I'd love to see the hardware more Linux-friendly, I have no doubt it'll still be some time off.
Wait, I missed the part where you explained why exactly MS would want to do this? Microsoft isn't in the bios business. Why would they care who writes the bios?
In general, it would add some credibility to your position if you understood what the people writing the code actually care about.
I understand that some of the kernel developers don't care about the freedom of free software in the same way that the FSF does. If they just want to code and release it for free, fine. There is nothing wrong with the BSD license, and they certainly have the right to choose whatever license they want. All I'm trying to point out is that a) they do care about freedom to some extent because they initially chose the GPL and not the BSD license; b) there seems to be an irrational opposition to GPLv3 that probably has something to do with dislike for Richard Stallman and leftover bad feelings from the BitKeeper incident; c) that GPLv3 is intended to protect the freedom in free software that they obviously do care about to some extent (see debates over NVidia's binary kernel driver for an example); and d) I have yet to read/hear any kind of rational and logical reason to not adopt GPLv3 if you are already using GPLv2.
Calling them "ignorant wankers" sure sounds like an ad-hominem attack to me.
Then maybe you should pay attention to the context. I am claiming that they are ignorant of the issues they are trying to address with their statement, which still appears to be true, and they are blowing a lot of hot air without first trying to understand the purpose of the object of their irrational dislike. It's pretty simple, if I don't think they know what they are talking about, I'm not going to respect their opinion. Likewise, a kernel developer is not going to respect criticism of their code from someone who has never contributed a patch. That's not ad hominem, it's dismissing the peanut gallery.
So you acknowledge that your point about "DRM and patent issues" is "not useful discourse"...? Fine.
Well, I don't recall making a point about "DRM and patent issues." I also don't recall claiming to be having discourse. I'm simply stating that the reasons given for opposing the GPLv3 are nonsense. I don't really want to pick through and respond to each one because it has already been done numerous times elsewhere. Nice try, though.
The GPLv3 license has not been finalized, but the proposed no-DRM clauses would certainly appear to forbid the combination of GPLv3 code with the OpenSSL library, which is used for PKI, OpenSSH's authentication, x.509 certs used by SSL webservers, and so forth.
Come again? I'm not aware of DRM implementations in any of the things you have mentioned. I believe you are confusing a couple of different things. Certificate signing for public key authentication purposes is not even addressed by the GPL and has nothing to do with it. Code-signing is addressed by the GPL, but the anti-DRM clause only comes into effect if it is impossible to run unsigned code (referred to as tivoisation). Seriously, do you really think RMS would add a clause to the GPL that would prevent the use of cryptography? Richard Stallman may be a lot of things, but he isn't stupid. There is, apparently, an incompatibility between the GPL and the OpenSSL license (note: not functions used in the OpenSSL library). This is resolved, not exacerbated, in GPLv3 via the additional restrictions clause.
Just how much of your code is in the Linux kernel today? If the answer is zero, well, you might want to contribute something more useful than a broad ad-hominem attack against the people who actually are writing Linux kernel code.
/etc/passwd and /bin/login from checking whether a user has a valid password, and it needs to not prevent systems like OpenSSH or OpenSSL from being used with GPLv3 code.
/etc/passwd. I can't agree with or refute vague assertions.
What does my contributing to the linux kernel have to do with a debate about licenses? The kernel developers are brilliant when it comes to writing kernel subsystems, and I wouldn't argue with them about the best way to write a scheduler algorithm...that doesn't mean they are equally knowledgable about everything else. This isn't an ad-hominem attack. The reasons they have given for their position are nonsensical and indicate they are largely ignorant of corporate politics, intellectual property, and the freedom in free software that the FSF is trying to protect.
You bet. Obviously the issue is so controversial that wide agreement has not been reached, and may not ever be reached.
Sure, but rehashing the same arguments over and over that have already been addressed in other contexts is not useful discourse.
For the GPLv3 to be usable for general-purpose operating systems, it needs to not forbid mechanisms like
Ummm, what are you trying to say here? That a GPLv3 networking library can't link to OpenSSL? Nonsense. I think you are being deliberately vague because you don't know what you are talking about. How about providing a specific example of a case where GPLv3 code can't be used with common mechanisms like
There are plenty of companies who like the GPL just fine; try asking MySQL
Completely irrelevant. Yeah, Red Hat likes the GPL too, but we aren't talking about open source companies. We are talking about embedded companies, mainframe companies, even proprietary software companies...companies that use things like the linux kernel, or the GNU toolchain, or the OpenSSL libraries because they can customize it as they wish and it saves them in developer time and money. These companies adhere to the GPL because they have to (for both legal and good karma reasons). I'm quite sure they would be happy without the provisions that require them to publish their changes. In fact, companies that use the linux kernel get nailed for trying to weasel out of it quite often (at least once a year). Having a stricter license that closes loopholes may drive a few companies away, but it is not going to cut off corporate support of open source development.
Agreed. This position statement seems to be, "We don't like politics and GPLv3, so here are a bunch of bs reasons for opposing the license change." Not liking politics is fine, but if you are going to draft a position statement you need to make sensible arguments and not just sound like ignorant wankers. The DRM and patent issues have been addressed by the FSF a number of times. This is not about a war against DRM and patents; it is about free software being free. Oh, and corporations may not like it, but since when do corporations like any version of the GPL? I'm sure they would really prefer the BSD license, so should we just switch to that to keep our corporate friends happy. The thing is, corporations want to use linux. They like linux, and they will put up with the license to use linux because, in the long run, it saves them money, even if they have to give something back to the community in return.
No, the contributors still maintain copyright, they just have to grant an unlimited, unrevokable, license to the project (or company) that is compatible with the planned distribution model. So if, say, a company wanted to distribute a GPL free version and a binary non-free version to paying customers, contributers would have to grant a license to the company to distribute the work in those ways. They don't have to transfer ownership rights, they just have to grant redistribution rights. It is no different really than contributing to something like the Linux kernel where you have to agree to use a license that is compatible with the rest of the project (GPL v2) if you want to get your code in the main tree. ReiserFS is still owned by Namesys even though it is in the main kernel tree and has been patched/maintained by a number of other people since its inclusion.
With transfer of ownership rights, Digium could choose to eliminate the GPL version and require everybody to purchase the non-free version. Not saying Digium would actually do that, just that it is possible. With the Linux kernel, it is impossible for Linus to change from the GPL license to, say, the MPL license. He would have to get the agreement of all of the contributors or replace their code.
Well, there is the MPL, although the FSF considers it to be a "weak copyleft" license. I can't think of any legal reason a project would have to take ownership rights of contributed code in order to distribute under a dual-license. The FSF requests contributions to GNU projects be transferred to them for license enforcement reasons, but they don't require it.
Not sure about mysql and trolltech (I think they are mostly developed in house, actually), but Apache uses the Apache license which allows for non-free distribution of the code. The contributors have to license their contributions properly to get them accepted into the main code base, but they don't have to give up their ownership rights.
although that does bring up another potential thing to trip up new users... too much choice.
Sure, but the default Ubuntu installation does Gnome, so there is no choice. You have to specifically download Kubuntu if you want the KDE desktop.
about the system volume...
I read both of your comments...I still have no idea what you are talking about. What do you mean by "doesn't remember the system volume?" If it doesn't know where the root partition is, that's a bug in the installer that needs to be fixed. The only time I have ever had boot loader issues is when I am installing on SCSI drives. This is really a GRUB issue. I know other people have the same problem, so hopefully it will be fixed soon. But I don't think that is what you are talking about because Macs haven't had SCSI drives for quite a while.
about the overly complex system settings- I'm mostly talking about KDE's system settings.
Yes, this is my #1 complaint about KDE, which is why I don't use it. Some people seem to like it that way, so let them keep it. As long as I have a simple, uncluttered Gnome interface, I am happy.
the installer really should have a streamlined install.
I agree with having a streamlined install, but I think the Ubuntu install is pretty good. I don't see how it could really get any easier. I was able to go from a Windows installation to a dual-boot configuration in about 30 minutes with a friend's laptop (auto-resizing of partitions and everything). That's pretty good. The advanced options are left in because some people, like me, don't want the installer mucking around with the partitions. Most people can select the "automatically partition free space" option. This could still confuse some people, but it is easy enough that they could follow somebody's clear instructions.
Other than that, I don't understand why the --enable-compositor compile-time option isn't included by default.
Well, it will be eventually. This is currently experimental code, so they don't want it to be in Gnome by default yet.
So, uh...yeah. With all of the talk about the War on Terror these days, what is being done about this kind of crap. Isn't the instillment of fear with threats of violence to obtain political victories terrorism? I don't see GW talking about terrorism that happens in our own country (you know, by Americans) in the news....
I have no problem with closed hardware, or software. The problem comes when closed hardware (or software) companies use open source software to save on development time and then don't give back to the community. Sorry, letting me look at code I can't run may be following the letter of the license, but it doesn't follow the spirit. Companies, being amoral entities, only care about the letter, though, which is why the license needs to be modified to close loopholes.
But the vendor must still publish the source of any changes he makes to the code. So the vendor is giving back.
No, it's not. The device driver code for an obscure chip on, say, a wireless router is completely useless if I can't run the code on said hardware. Think about the Linksys WRT54G. It was just running a linux kernel modified to run on the router with a set of minimal networking utilities. When the source was released, people were able to add all sorts of stuff: better firewalls, servers, ssh utilities, more efficient and capable routing.... They wouldn't have been able to do any of this stuff if Linksys had tossed in a TPM chip and made it so only their signed modified linux would run. The changes they made were also useless without the hardware (yay, so now we know how to run linux on a Linksys router, to bad we can't actually do it).
So, yes, a consumer could just go buy a different wireless router, but that's not the point. The point is that Linksys would have been using GPL'd software to reduce their development costs without giving a useful share back to the community. Developers who don't mind that sort of thing use the BSD license. Developers who do use the GPL. For the GPL to continue to be relevant, it has to be modified to close loopholes that didn't exist 10 years ago.
The way Linus sees it is from the "developer" viewpoint. The code is still free from this viewpoint, since all modifications are published. You can modify it and run it on a DRM-free machine.
That's the short-sightedness of Linus' argument (the same short-sightedness that let him get trapped by the Bitkeeper fiasco). There are DRM-free machines now, but that doesn't mean there will be in the future. If the media companies have their way, every desktop computer will have a TPM chip in it, and if you want to view things like HD-content, it has to be enabled and running. So a company can take a project, like say mplayer, make a version that plays their video format, decrypting the stream via the TPM hardware, and then sign the binary and then sell it. Congratulations, a company has just saved themselves a couple of years of development time to make a video player to help sell their video files, and you can't modify it at all. If you modify it, it reverts to just plain old mplayer without the ability to use the code that was added. That defeats the purpose of the GPL. Linus really needs to wake up here. Yes, proprietary software has a right to exist, but pretending a company won't take advantage of free software to reduce their development costs (without giving anything back, if they can) is stupid. The GPL allows commercial use of free software as long as you give a fair share back to the community. It is not some fiendish scheme to force all software to be free as some people would say. The GPL as it is has worked fine for the last decade, but now it needs to change or it will no longer serve its purpose.
I'm not so sure they'll be able to pull this off, though. Limewire is a Gnutella client. Everything is distributed. It isn't like Napster where they had a centralized search engine to find pirated music. The only thing Limewire does is implement the Gnutella protocol, which means you can use it to search for any shared files. Even if the Limewire people wanted to block copyrighted material, there isn't a feasible way for them to do it. Block by filename and people will just change it to something else. Prevent searching for certain search terms and people will start to use secret keywords. There's just no way for a blacklist to keep up with attempts to circumvent it. So then you could sample individual files looking for signatures, but that a) would slow things to a crawl, b) probably wouldn't be resistant to changes in encoding/bitrate. So in addition to just mp3 signatures, you would have to check ogg/wav/flac/aac.... And if you ever took it that far, people would just start hiding things in archives or using encryption. Blocking files or searches just isn't going to be possible.
As for legitimate uses of Limewire, there are plenty: shareware, freeware, Creative Commons licensed stuff, samples/demos/free trials, not copyrighted stuff...now you can argue that Limewire isn't used for these things 90% of the time, which is probably true. But Limewire has a pretty clear defense: software was built to promote legal filesharing, software accomplishes this goal (see random statistic showing legally shared files), yes illegal filesharing happens, but there is nothing we can do about that above and beyond what we are already doing (putting strong warnings everywhere and allowing content owners to submit a SHA1 hash of files they don't want distributed).
Don't be so sure. Simple convenience features like fast servers, good bandwidth, quality encodings, choice of encodings, easy searching, etc...can go a long way, and people will pay for it even if they could get the same thing for free (see just about every commercial linux distribution). They wouldn't be able to get away with charging $1/song without copyright/drm, but a lot of people (me included) are willing to pay $0.20/song or so, or maybe even a flat fee of $10/month.
Yes, that is exactly what he is saying. Distributors have a fundamental right to include copyright protection, and consumers have a fundamental right to break that copyright protection on media they have bought. That doesn't mean they have a right to then break copyright law by giving copies to all their friends, but if, say, copyright law allowed somebody to make additional copies for their own personal use (*cough* fair use *cough cough*), they would first have to break any copyright protection before they could do that. The DMCA makes circumventing copyright protection illegal, which basically allows distributors to define their own copyright law anyway they please (because any unauthorized use or manipulation of the content is illegal).
Is it killing off bacteria we want to keep around? Maybe. Is it creating an environmental pressure for the evolution of Triclosan-resistant bacteria? Definitely.
I have to argue No and No to these questions, unless you are talking about a hospital or something. By the time the already low concentration of Triclosan in soap reaches the sink, it is diluted past its efficacy. Add to that that it takes time before it can do its job, as another replyer noted, and the Triclosan in antibacterial soap is completely useless for killing off bacteria. The detergents in soap are already doing a much better job.
There is also a fairly common misunderstanding about how antibacterial-resistance develops. Most bacteria replicate quickly and can therefore mutate quickly to adapt to adverse conditions (ex: an antibacterial agent). However, bacteria will revert just as quickly when mutant phenotypes are no longer necessary. Since most mutations that confer antibiotic or antiseptic resistance involve the expression of additional genes and consequently a slower dividing time, bacteria will revert in the absence of a selective pressure. People who work with antibiotic-resistant strains of bacteria in labs observe this all the time. That is not to say that antibiotic-resistance isn't a problem...it is. But squirting a little antiseptic-containing soap down your drain every once in a while is not likely to be contributing to the completely unrelated problem of antibiotic resistance (ex: vancomycin resistant S. aureus in hospitals).
Everything you say is good up until here.
3. Society in general and the marketing departments of various household goods: all of the various soaps & cleansers which promise to kill bugs when you use them. You're only supposed to use soap to clean your hands off - remove the stuff which doesn't belong there - remove as in get it off of your hands, not kill some of the bugs and leave a small number of immune ones in place. Eugenics takes over and we begin breeding superbugs.
Antibacterial soaps are a marketing ploy and nothing more...all soaps are antibacterial. How well a given soap removes bacteria from your hands is directly proportional to how well you clean your hands (i.e: do you just get them wet, or do you really soap up and scrub them down). Bacteria aren't some magical things that can survive the same conditions that will remove dirt, grime, oil, protein, salt, and metals from your hands. Some companies throw in a little bit of antiseptic to get people to buy their soap, but it is no more or less effective than regular soap, and at the concentrations present, it is highly unlikely to cause resistant bacterial strains to develop.
As a current and past employee of several companies that make wireless transceivers subject to FCC licensing, I can tell you that there is no cost effective way to limit a device to FCC restrictions purely in hardware.
Um, sure. But I don't think Theo cares about it being "pure hardware" or a hardware/firmware mixture. As long as it isn't in the software driver, wifi hardware companies can't claim that releasing source is against FCC rules. And, as the parent was saying, it probably isn't anyway.
I wasn't going here for support, I was making the observation that ext2 was fragile, and that I had lost data from it. And immediately, it was all my fault.
I was saying, and rightly so, that Linux wasn't ready for business yet, and wouldn't be until it had a better filesystem. This was approximately like pouring blood in the water; the sharks showed up minutes later.
After ext3 and Reiser went mainstream, Linux was finally ready for primetime... and then everyone agreed that ext2 was really awful.
But that's just it, ext2 isn't awful. And no I'm not a linux zealot (although I am pro-linux). ext2 lacks one critical feature for high availability...journaling. Everything else about ext2 is great: it is fast, efficient, stable, extensible, portable, and it auto-defragments. You couldn't say the same thing about reiserfs until later versions. So, yeah, reiserfs had journaling, but it was still slow and unstable and nobody wanted to touch it. In the early years, a business could easily make use of linux with ext2 if they acknowledged the risks and took the right precautions (ex: a UPS).
If people like a particular operating system, they will tend to overlook its weaknesses in favor of its strengths, but that doesn't make them zealots. They have just decided that the strengths make the OS good in spite of its weaknesses. And, naturally, when another weakness is checked off the list, they can look back and go "Wow, we've come a long ways. Yay linux!"
I like how nobody in this thread, including the parent actually tried to investigate the problem. mod_security is gpl'd and available in Debian as libapache2-mod-security, so wtf are you talking about? Next time try to actually look for the packages before claiming that they don't exist.
/etc/allow-packages, would be a legal problem for Debian to support officially.
In general, though, Debian already does what you ask for. It has a main distribution, for software that meets the Debian Free Software Guidelines, and a non-free distribution for other software. If Debian doesn't distribute something (officially), there is probably a serious legal hindrance, and other distributions like Red Hat probably don't distribute it either. However, also like Red Hat, there are unofficial repositories you can use to get these packages if you don't care about the ethical/legal issues. I don't really see why this should or how this could be easier. Anything more automagical, like