For those of you not familiar with that situation, the author borrowed and modified some open source code for a terminal app into a Finder alternative. A bunch of GPL zealots then started a flame war with many actually demanding that he release the entire source code to the world under the GPL! What was worse was that the author had already helped the community by releasing many of the classes he had developed!
Hrm. I hadn't heard of this PathFinder GPL violation before, but a few minutes with Google paints an entirely different picture to the "GPL Zealots vs the Virtuous PathFinder Guy" that your story painted.
From here it seems that rather than "many actually demanding" a GPL release, it was very few people even hinting at a GPL release, and no actual demands were made. In any event, only the iTerm authors can make demands and even then they can't demand a GPL release.
For the most part, people were just exploring the possibilities in a mostly civil manner. If I was forced to polarise the discussion I would say the most significant minority of comments were anti-GPL trolls; typically saying things like "GPL BAD BSD GOOD" and other nonsense.
Later, the author of PathFinder apologises and admits it was an honest mistake. Many subsequent comments are then repeating that the mistake was honest so they should try and find a solution where everybody walks away happy. Nobody wanted to crucify the PathFinder guy... at least, not that I saw.
So I don't know where you got this entirely negative opinion of "GPL Zealots" from. My view is that the mistake was honest, the mistake was admitted, the iTerm authors seemed content to find an equitable solution, there were the typical anti-GPL trolls, and no actual demands for a "GPL release" were made.
Now contrast this with intentional violations of the GPL where the violator refuses to comply. For example, easyRDP.
What will happen is a lot of local culture will disappear in place of the golden arches and starbucks. All so McDonalds can claim to have served 4 trillion instead of 3 trillion.
If the local people want to eat McDonalds and drink Coca-Cola instead of eating what their forefathers ate then what gives you the right to tell them to do otherwise?
Is culture so important that society must be retarded by it? At which point did what somebody wants today become less important than what their ancestors did 100s of years ago?
Let's make the question even more general: why are you trying to stop people in other countries from making their own damn decision.
My understanding is that the ACT Government represents the ACT (strange that)... an underfunded town that is smaller and less influential than Munich.
Except the ACT has the highest concentration of public service in the entire country. There is a huge IT industry in Canberra to support the public service. Any decisions in ACT will have a fairly significant knock-on effect across the rest of the country.
What is the actual benefit of voting electronically? Many countries use the tried and true method of voting using paper and pen -- just mark your X in the square next to the name. Volunteers tally up the votes at the end of the voting day and, within hours of closing, you get your results.
It's something everybody understands. The paper waste is minimal compared to the paper output of election-related things -- government paperwork, campaign signs, and flyers in your mailbox and everywhere else. You absolutely don't get hanging chads, broken levers, or some other malfunctioning convoluted contraption. Recounts and verifications are simple -- get those same volunteers to count 'em again.
Geek factor aside, where's the benefit of going electronic?
The cost of tallying. My brother took a part time job as an election tallyer. It's a slow and expensive process with dozens of people working for many hours after the voting booths are sealed. Also hand-tallying is not very accurate. I forget the exact details (going on memory of the stories he told me) but I think he was allowed 2% errors. There was a process whereby stacks of votes would be recounted by different staff to try and detect fradulent tallying and other inaccuracies but it was not comprehensive.
Of course, the current Diebold system sounds even more expensive and less accurate than hand counting. It's even less accountable because at least the hand-tallying process could be double checked. The Diebold electronic system sounds as bad as it could possibly be.
The Australian govt. didn't send this guy there, he's not representing Australia so his nationality shouldn't play any part in this. Would be great if it really worked that way though, I could pull out my old Australian passport next time I go though US customs and they'll let me go to the front of the line in front of the Chinese, French and Germans. That's not really why you guys went to Iraq, is it? To buy a 'special treatment for all our citizens/preferred ally' card?
As much as I'd like to say that wasn't the reason, my cynical opinion is that's exactly the reason why our government assisted the USA.
Though bear in mind, when you say "you guys", this country had it's biggest protest march in history against going to war, with every major city having significant anti-war protests and demonstrations, with anti-war speeches by some of our most noted academic, religious and social representatives. Our government basically ignored us. The prime minister later said something to the effect "I know the facts better than the entire Australian population though I'm not going to show you the evidence and I don't care what you want, we're going to war anyway". Thus proving once and for all that democracy in this country is a sham.
Afterwards it was revealed that the evidence our government used to justify its decision was fabricated and high-level government officials knew it was fabricated. Do you think anybody lost their jobs? Or went to gaol? Hell, no. The prime minister that was involved in this corrupt farce is running in the next election!
Putting aside the source code issue, one of SCO's complaints is that IBM has released "UNIX technology" for free (as in beer) and this has undercut SCO's profit margins from UNIX. This is in addition to SCO's complaints over copyright infringement and trade secrets being leaked (both of which are on shaky factual and legal grounds). SCO mentions their eroding market share and their lost profits multiple times in their submissions.
But now Sun is releasing the very same "UNIX technology" for free (as in beer). So what's the difference?
SCO might say that the difference is one of trade secrets. But end-users can't be held liable for trade secrets leaked by IBM.
SCO might say that the difference is one of improper contribution: Sun has a license to put "UNIX technology" into Solaris, and IBM has a license to put "UNIX technology" into AIX, but IBM doesn't have a license to put "UNIX technology" into Linux. But that's an argument that still needs to be decided in court (plus the facts and the law are heavily against SCO).
But in terms of eroding SCO's market share, Free Solaris/x86 is exactly the same as Free Linux. There is no difference. Both products are superior to UnixWare and both are available at no appreciable cost.
So I'd like to see how SCO reacts to this. If they don't complain then what they're realling saying is that they don't mind their core product (UnixWare) being undercut by a far superior UNIX (Solaris/x86). What they really care about is that the product killing their market is Linux. And that's suspicious. Why should they only care that it's Linux?
Since the kernel module is included as a part of the 2.4 kernel, and that I have the Slackware Glide source, it is my belief that all the libraries and modules necessary for 3DFX support are open source.
Correct. The entire tdfx driver-chain is open source. From Mesa (high level OpenGL) to the DRI (direct rendering) to XFree86 (resource manager) through to Linux (tdfx kernel module) and Glide3 (basic hardware interface from client-side libGL direct-rendering library).
In fact, I consider the Voodoo 5 a great example of how well open source drivers can work. OpenGL support ended at version 1.1 under Windows, but has continued to progress allong with the Mesa project under Linux.
Also correct. 3DFX is dead but their drivers live on. The latest improvements to Mesa improved tdfx as well. Future improvements to Mesa will also improve tdfx. Glide3 is still maintained and probably will be until the very last 3DFX card dies.
I see no point in releasing any of my code as that would lower my income, because none of the manufacturers would need to buy my code anymore...
Yes they would. Copyright doesn't evaporate once somebody sees the code. You still own the copyright. They have to license it from you. If you want to charge them money for the license then they must pay. Them's the rules. I don't know where this crazy meme "software must remain secret or the copyright disappears" came from, but it's plain wrong. Copyright is enforcable whether the source code is open or closed.
If you want to keep the code a secret because you think all the manufacturers are unscrupulous thieves who will steal your code and not pay you as copyright requires, then that says more about the rampant dishonesty in the software industry than anything else.
The IDE Raid controller you're talking about uses some functionality which I have a patent on.
Bear in mind that patents can NOT be kept secret. Patents MUST be on the public record or the patent will not be granted. This is another crazy meme "if it is patented I'm obliged to keep it secret". Wrong! You are obliged to tell everybody exactly how it works. That's the purpose of a patent; you must disseminate the idea far and wide to receive the time-limited exclusive monopoly.
However, patents do have a significance in this discussion because they are incompatible with the GPL.
No, they are not. The existence of RCU (an IBM owned patented technique) in Linux is proof of that. In the case of RCU, Linus demanded and got a royalty-free perpetual license for RCU before he permitted it into the kernel. You even acknowledge this possibility:
ATi in the above case, would have to license their patents for everyone's free use, as GPL requires, and would not be able to charge royalties for its use.
So I don't see that there is any dispute here. Nvidia doesn't release the source because they don't want to. Licenses, patents, agreements, trade secrets, these are all solvable problems. Nvidia simply doesn't want to solve them. They are happy with the current situation because the vast majority of their Linux customers don't care. Apathy is our own worst enemy.
And it was only because of binary-only drivers that Linus was able to see his screen and therefore code Linux.
Minix is not and never was "binary-only". It and all the drivers came with full source code though the license was far too restrictive to be considered free software.
In certain cases, manufacturers can provide open source modules, but many times, there simply is no viable way. Complex hardware or hardware/software combination is usually covered by multiple patents,...
Red Herring. Patents have to be on the public record. That's a requirement of getting the patent.
Software vendors bury their patents inside copyrighted code because they're greedy. They get the patent and they don't have to reveal the implementation. It's definitely against the spirit of patents even if it does fall within the letter of the law. But open-sourcing the module only has relevance to copyright licensing and trade secrets and agreements; the patents are still valid and they can charge whatever royalties they like.
"I'd prefer there wasn't a closed source driver" is an opinion you are clearly applying only to yourself. "Having a closed source driver is worse than no driver at all" is a blanket statement that will be taken as applying to everyone. If that's not how you meant it you should have been more precise.
I think you are splitting hairs. "Having a closed source driver is worse than no driver at all" is very clearly an opinion. You would have to struggle to interpret it any other way.
Unless you are under the mistaken impression that I'm the release manager for nvidia and that I have the power to stop them releasing drivers? Or that I have the power to stop you from using them?
If companies can't use proprietary, binary modules to protect their/others IP, then Linux will never be a truly "first class" OS.
Why? Many people make this claim but what is the rationale?
My definition of "first class" OS is one that is reliable, scalable, manageable, supportable, high-performance, feature-rich, low-cost, etc. None of that requires closed-source binary-only modules.
The only possible argument is that certain IP (patents, copyrights, trade secrets) needs to remain closed-source binary-only before they can be used in Linux. Of course, patents cannot be closed-source by definition (the method must be public knowledge) and RCU proves that patented IP can be used in Linux as open-source code. Trade secrets and copyrights? The thing about trade secrets and copyrights is that if we independently create them, they are ours forever. We don't need closed-source binary-only implementations, because we can write our own unencumbered versions. It might take longer but since when do we have to be impatient in our quest for world domination?
Therefore I claim that you are wrong. I assert that closed-source binary-only drivers are not required for Linux to be a "first class" OS.
Without nVidia's closed source drivers I wouldn't be using Linux at all.
I honestly wouldn't have a problem with that. Don't ever make the mistake of thinking I care what software you run. I don't even care if you use the binary-only nvidia driver, though it seems to be a common mistake on this thread for people to think that I do. I stated my preference that the closed source binary-only driver didn't exist but, if the driver does exist, it's your choice whether you use it or not.
You seem to be capable of writing a video driver, so why don't you reverse-engineer the binary nVidia driver and roll your own open-source driver?
Effectively, we did. Not myself personally but the Utah-GLX project had a working nvidia driver, originally contributed from nvidia, no less. It wasn't very fast but all Linux drivers begin their life "sub-par" and rapidly improve. It's an expectation to have wobbly beginnings and a solid finish. The Utah-GLX driver was rapidly improving; in the few months I worked on it the performance doubled, bugs were disappearing faster than they were being found, and patches were pouring in from dozens of enthusiastic hackers.
Then nvidia released their binary-only driver.
The entire user base disappeared. The source of all feedback and patches was gone. The problem was that the nvidia drivers were good and they didn't cost a cent. That is the worst environment possible for the development of a free driver. If the nvidia drivers had cost $50 or were crappy then the free driver would have bloomed. Instead, the entire Utah-GLX/nvidia userbase disappeared practically overnight. No more bug reports. No more feedback. No more patches. No more interest. That's exactly what kills an open source project. It was very demoralising.
You might be aware that there have always been closed-source X drivers. Back in the early days it was quite common to install the non-free Metro-X software. Their drivers were excellent. In some cases 4-5x faster than their XFree86 equivalents, if XFree86 even supported the same hardware! Red Hat bundled Metro-X to increase their supported cards. But that was all OK because Metro-X wasn't cheap. There was still an incentive to create the open-source equivalents in XFree86. And that was good because although it hurt initially we came out stronger because of it. Nobody bothers with Metro-X anymore; the combination of closed-source and a high price-tag forced the community to produce decent open-source equivalents.
The nvidia driver is different. It's not only good but it also comes at no cost. That's the worst possible combination. That's that will harm the growth of Linux. Not patents. Not lawsuits. Apathy and complacency are our enemies. The nvidia drivers encourage apathy because they're "good enough" and most people are willing to compromise their principles for the immediate gratification of 3D graphics.
And the fact that of all the responses I have had, not a single one has agreed with me, shows that there are very few of us who recognise that closed-source drivers are bad. As I said, I have no desire to stop you doing what you want. If you want to use closed-source drivers then you go right ahead. But I think it will slow the spread of Linux, not hasten it.
Oh yes, I agree, which is why I know the cost of Solaris is prohibitive. You need to get the Ultrasparc hardware and all the optional non-free software pacakges (Forte) to make it worth your while.
I actually have Solaris at home running on an ancient Ultrasparc 2. My day job involves Solaris admin work. I don't need any advice on what is and isn't good about Solaris. I've some small amount of experience in this area.
So, in short, you would rather the company that makes video drivers for my card not, so that I can't use them to their full extent (until someone reverse engineers the drivers, which won't be a small job.) Hence, me losing some capabilities on my computer.
Yes, I would rather that the drivers did not exist, but I don't see how you manage to make the leap to "rather I didn't have the freedom to do what I want". I have no desire to impede your freedoms. I simply wish that the drivers did not exist, because then we would be forced to do something about it instead of slumming in this horrible limbo of complacency.
I'd bet that only a very small minority run Linux due to mainly ideological reasons. You seem to be one of them
Absolute nonsense. I mainly use Linux because of cost. I'm a cheap bastard. I wanted a UNIX and I'm not willing to pay for a Solaris environment (the only other UNIX I would consider if I could not use Linux).
But the more I use Linux the more I appreciate that the reasons I like Linux stem from the ideology. Source code. Zero cost. No license daemons. Community and kinship. Best of breed. I realised after many years of reflection that all the benefits of Linux are because of the free as in freedom aspects of Linux. It is something you can only properly appreciate after you have drowned in Linux for years and then stepped back.
But the rest of us care more about getting the most out of hardware we paid for.
And that's your prerogative. I don't know why you (and others) have got the impression that I want to stop you. Do whatever you want. I'm in no position to tell you otherwise and nor do I have any desire to stop you.
Perhaps you think that because I said that closed-source nvidia drivers are more harmful than no drivers at all - and that's a statement I'm not backing down from - that you think I'm working actively to destroy nvidia! Don't be silly. I have no control over nvidia. They will release their drivers and you will continue to use them. Doesn't bother me. Never has. Never will.
But I stand by what I said. I strongly believe that in the long-run, closed-source nvidia drivers are harming Linux more than they are helping. They are preventing the development of open-source drivers. Any brilliant ideas in nvidia's drivers can not be reused in the ati drivers. Similarly, any brilliants aspects of the DRI cannot be used within the nvidia drivers because nvidia doesn't use the DRI. The closed-source nvidia drivers are buggy and we can't fix that; we are beholden to nvidia to fix the bugs for us. These are all detrimental aspects of the closed-source binary-only nvidia drivers. You'll hopefully notice that they are all pragmatic reasons, not ideological reasons.
And next time, please leave out the better-than-you-since-I've-used-linux-longer pissing contest thing. It added nothing to your otherwise intelligent post.
I made the remark to highlight that Linux came from very humble beginnings - no graphics, no LVM, no RAID, no SCSI! - and that it didn't need binary only closed-source drivers to escape that mire of feature-less and driver-less poverty. It escaped solely because of the freedom aspects of Linux; because somebody with unsupported hardware could download the kernel source, write a driver for their particular hardware, and send the patch back to Linus. If Linux wasn't free as in freedom then it would never have grown beyond Linus's 386 with an IDE drive and text-mode only interface.
And cost. Never forget cost. The Real UNIX(tm) I was using before Linux cost an absolute mint. Linux was faster, stabler, had more features, came with source code and was cheaper. It really was no-contest.
It would seem that you would rather I didn't have the freedom to do what I want with the OS on my computer.
I don't know why you would think that. I don't have any desire to tell you what you can do with your computer and your software. I said what I thought and what I would do. That was all.
What I find odd are people who put words into my mouth and attribute me with opinions that I don't agree with, and then attack these phantoms as if that would invalidate my real opinions.
I'll run my binary modules, unload them if I think I have a bug to test it, and get my very nice hardware acceleration out of a card that I payed $200 for.
That's nice. I paid $20 for a card that worked with the default Debian install. But you do whatever rocks your world.
Sacrificed? It doesn't go away because of a videodriver. I don't know what choices nvidia have. Maybe they could release enough to make it possible to write a good driver? And if so, they should. And we should definitely keep bugging them for not doing it. But to say that chosing a binary only driver sacrifices the whole operating system including the effort of "1000s of volunteers" is a bit over the top.
Yes, that is over the top, but it's not what I said. Read it again. I expressly said "Not for the same code but it's the same principles". It's the principles that you sacrifice, not the whole operating system. If you don't appreciate that the principles are important then you probably won't understand.
The main problem I still see is that you need to draw a line somewhere and allow closed source drivers or else you impose a limit on what Linux user can do. That is the ultimate consequence of what you're saying. Nvidia cards are mainstream and someone is bound to create an open source driver for them if nvidia refuses to. But if you use hardware that isn't mainstream and are incapable of developing your own driver, are closed source drivers still a bad idea? Or should people who can't write their own drivers stick to Windows? I am just curious, how far do you want to take this?
It's up to the individual to make their own decision. If someone wants to use closed-source binary-only drivers then I won't stop them. It is their freedom that they are throwing away.
How far do I personally take this? Where do I draw the line? For the record, if I have hardware and there is no open-source driver then I simply don't use the hardware. Or I sell it. Or I write the driver. I have stuck to these principles ever since switching to Linux.
For example, I bought an nvidia TNT2 several years ago. I was fooled by nvidia's claims that they were going to support Linux. They told the truth - they did support Linux - and the mistake was mine because I incorrectly assumed nvidia was going to release open source drivers. But I still felt cheated. I joined the Utah GLX project to write drivers for the TNT2 and I think the result was successful enough. I later sold the card and bought something that was properly supported by open source XFree86 drivers.
I don't have a problem with "imposing limits" on Linux users. I remember using Linux when there were no supported video cards at all (before X386 was ported). You had text mode, and that was it. Linux grew beyond those days without requiring binary-only drivers. We wrote open-source drivers as required and Linux grew from basically nothing on the strength of open source. Why do we need to change the winning recipe now? What has changed so that we suddenly need binary-only closed-source drivers, or Linux will fail? I don't see it. Open source got us this far. It'll take us the whole way. It may take slightly longer, or limit our choices of hardware, but so what? I see no reason to change strategy now. I see no reason to compromise the ideals to fulfil a crass gratification.
Wow. It's just amazing to me how the Linux community supports the wholesale theft of SCO's intellectual property,
Hard to believe you're not a troll but just in case...
The Linux community wants SCO's "intellectual property" out of Linux. The GPL doesn't permit their IP to stay in there. It must be removed; the license and the law demands it.
But first they have to tell us WTF their IP actually is. We can't remove it because we don't know WTF they're talking about! Even IBM doesn't know and the judge couldn't figure it out either, which is why SCO has been compelled to tell the court WTF they think they own in 30 days time.
Linux developers are very respectful of IP rights, even if sometimes the users (*cough*MP3 thieves*cough*) are not. This is why we wrote our own operating system, from scratch, rather than subject ourselves to binary-only licensing.
Slashdot Mirror
Hrm. I hadn't heard of this PathFinder GPL violation before, but a few minutes with Google paints an entirely different picture to the "GPL Zealots vs the Virtuous PathFinder Guy" that your story painted.
From here it seems that rather than "many actually demanding" a GPL release, it was very few people even hinting at a GPL release, and no actual demands were made. In any event, only the iTerm authors can make demands and even then they can't demand a GPL release.
For the most part, people were just exploring the possibilities in a mostly civil manner. If I was forced to polarise the discussion I would say the most significant minority of comments were anti-GPL trolls; typically saying things like "GPL BAD BSD GOOD" and other nonsense.
Later, the author of PathFinder apologises and admits it was an honest mistake. Many subsequent comments are then repeating that the mistake was honest so they should try and find a solution where everybody walks away happy. Nobody wanted to crucify the PathFinder guy... at least, not that I saw.
So I don't know where you got this entirely negative opinion of "GPL Zealots" from. My view is that the mistake was honest, the mistake was admitted, the iTerm authors seemed content to find an equitable solution, there were the typical anti-GPL trolls, and no actual demands for a "GPL release" were made.
Now contrast this with intentional violations of the GPL where the violator refuses to comply. For example, easyRDP.
If the local people want to eat McDonalds and drink Coca-Cola instead of eating what their forefathers ate then what gives you the right to tell them to do otherwise?
Is culture so important that society must be retarded by it? At which point did what somebody wants today become less important than what their ancestors did 100s of years ago?
Let's make the question even more general: why are you trying to stop people in other countries from making their own damn decision.
Except the ACT has the highest concentration of public service in the entire country. There is a huge IT industry in Canberra to support the public service. Any decisions in ACT will have a fairly significant knock-on effect across the rest of the country.
The cost of tallying. My brother took a part time job as an election tallyer. It's a slow and expensive process with dozens of people working for many hours after the voting booths are sealed. Also hand-tallying is not very accurate. I forget the exact details (going on memory of the stories he told me) but I think he was allowed 2% errors. There was a process whereby stacks of votes would be recounted by different staff to try and detect fradulent tallying and other inaccuracies but it was not comprehensive.
Of course, the current Diebold system sounds even more expensive and less accurate than hand counting. It's even less accountable because at least the hand-tallying process could be double checked. The Diebold electronic system sounds as bad as it could possibly be.
As much as I'd like to say that wasn't the reason, my cynical opinion is that's exactly the reason why our government assisted the USA.
Though bear in mind, when you say "you guys", this country had it's biggest protest march in history against going to war, with every major city having significant anti-war protests and demonstrations, with anti-war speeches by some of our most noted academic, religious and social representatives. Our government basically ignored us. The prime minister later said something to the effect "I know the facts better than the entire Australian population though I'm not going to show you the evidence and I don't care what you want, we're going to war anyway". Thus proving once and for all that democracy in this country is a sham.
Afterwards it was revealed that the evidence our government used to justify its decision was fabricated and high-level government officials knew it was fabricated. Do you think anybody lost their jobs? Or went to gaol? Hell, no. The prime minister that was involved in this corrupt farce is running in the next election!
Putting aside the source code issue, one of SCO's complaints is that IBM has released "UNIX technology" for free (as in beer) and this has undercut SCO's profit margins from UNIX. This is in addition to SCO's complaints over copyright infringement and trade secrets being leaked (both of which are on shaky factual and legal grounds). SCO mentions their eroding market share and their lost profits multiple times in their submissions.
But now Sun is releasing the very same "UNIX technology" for free (as in beer). So what's the difference?
SCO might say that the difference is one of trade secrets. But end-users can't be held liable for trade secrets leaked by IBM.
SCO might say that the difference is one of improper contribution: Sun has a license to put "UNIX technology" into Solaris, and IBM has a license to put "UNIX technology" into AIX, but IBM doesn't have a license to put "UNIX technology" into Linux. But that's an argument that still needs to be decided in court (plus the facts and the law are heavily against SCO).
But in terms of eroding SCO's market share, Free Solaris/x86 is exactly the same as Free Linux. There is no difference. Both products are superior to UnixWare and both are available at no appreciable cost.
So I'd like to see how SCO reacts to this. If they don't complain then what they're realling saying is that they don't mind their core product (UnixWare) being undercut by a far superior UNIX (Solaris/x86). What they really care about is that the product killing their market is Linux. And that's suspicious. Why should they only care that it's Linux?
An e-mail address! Quick, send him an Outlook virus!
Dear Software/Hardware Engineers,
Stop reading Slashdot and get back to work or you're both fired.
Sincerely,
UR Manager.
Correct. The entire tdfx driver-chain is open source. From Mesa (high level OpenGL) to the DRI (direct rendering) to XFree86 (resource manager) through to Linux (tdfx kernel module) and Glide3 (basic hardware interface from client-side libGL direct-rendering library).
Also correct. 3DFX is dead but their drivers live on. The latest improvements to Mesa improved tdfx as well. Future improvements to Mesa will also improve tdfx. Glide3 is still maintained and probably will be until the very last 3DFX card dies.
Yes they would. Copyright doesn't evaporate once somebody sees the code. You still own the copyright. They have to license it from you. If you want to charge them money for the license then they must pay. Them's the rules. I don't know where this crazy meme "software must remain secret or the copyright disappears" came from, but it's plain wrong. Copyright is enforcable whether the source code is open or closed.
If you want to keep the code a secret because you think all the manufacturers are unscrupulous thieves who will steal your code and not pay you as copyright requires, then that says more about the rampant dishonesty in the software industry than anything else.
Bear in mind that patents can NOT be kept secret. Patents MUST be on the public record or the patent will not be granted. This is another crazy meme "if it is patented I'm obliged to keep it secret". Wrong! You are obliged to tell everybody exactly how it works. That's the purpose of a patent; you must disseminate the idea far and wide to receive the time-limited exclusive monopoly.
No, they are not. The existence of RCU (an IBM owned patented technique) in Linux is proof of that. In the case of RCU, Linus demanded and got a royalty-free perpetual license for RCU before he permitted it into the kernel. You even acknowledge this possibility:
So I don't see that there is any dispute here. Nvidia doesn't release the source because they don't want to. Licenses, patents, agreements, trade secrets, these are all solvable problems. Nvidia simply doesn't want to solve them. They are happy with the current situation because the vast majority of their Linux customers don't care. Apathy is our own worst enemy.
Minix is not and never was "binary-only". It and all the drivers came with full source code though the license was far too restrictive to be considered free software.
Red Herring. Patents have to be on the public record. That's a requirement of getting the patent.
Software vendors bury their patents inside copyrighted code because they're greedy. They get the patent and they don't have to reveal the implementation. It's definitely against the spirit of patents even if it does fall within the letter of the law. But open-sourcing the module only has relevance to copyright licensing and trade secrets and agreements; the patents are still valid and they can charge whatever royalties they like.
I think you are splitting hairs. "Having a closed source driver is worse than no driver at all" is very clearly an opinion. You would have to struggle to interpret it any other way.
Unless you are under the mistaken impression that I'm the release manager for nvidia and that I have the power to stop them releasing drivers? Or that I have the power to stop you from using them?
Why? Many people make this claim but what is the rationale?
My definition of "first class" OS is one that is reliable, scalable, manageable, supportable, high-performance, feature-rich, low-cost, etc. None of that requires closed-source binary-only modules.
The only possible argument is that certain IP (patents, copyrights, trade secrets) needs to remain closed-source binary-only before they can be used in Linux. Of course, patents cannot be closed-source by definition (the method must be public knowledge) and RCU proves that patented IP can be used in Linux as open-source code. Trade secrets and copyrights? The thing about trade secrets and copyrights is that if we independently create them, they are ours forever. We don't need closed-source binary-only implementations, because we can write our own unencumbered versions. It might take longer but since when do we have to be impatient in our quest for world domination?
Therefore I claim that you are wrong. I assert that closed-source binary-only drivers are not required for Linux to be a "first class" OS.
That is correct.
Really? To me they mean the same thing. Tell me, what do you think the difference is?
I honestly wouldn't have a problem with that. Don't ever make the mistake of thinking I care what software you run. I don't even care if you use the binary-only nvidia driver, though it seems to be a common mistake on this thread for people to think that I do. I stated my preference that the closed source binary-only driver didn't exist but, if the driver does exist, it's your choice whether you use it or not.
Effectively, we did. Not myself personally but the Utah-GLX project had a working nvidia driver, originally contributed from nvidia, no less. It wasn't very fast but all Linux drivers begin their life "sub-par" and rapidly improve. It's an expectation to have wobbly beginnings and a solid finish. The Utah-GLX driver was rapidly improving; in the few months I worked on it the performance doubled, bugs were disappearing faster than they were being found, and patches were pouring in from dozens of enthusiastic hackers.
Then nvidia released their binary-only driver.
The entire user base disappeared. The source of all feedback and patches was gone. The problem was that the nvidia drivers were good and they didn't cost a cent. That is the worst environment possible for the development of a free driver. If the nvidia drivers had cost $50 or were crappy then the free driver would have bloomed. Instead, the entire Utah-GLX/nvidia userbase disappeared practically overnight. No more bug reports. No more feedback. No more patches. No more interest. That's exactly what kills an open source project. It was very demoralising.
You might be aware that there have always been closed-source X drivers. Back in the early days it was quite common to install the non-free Metro-X software. Their drivers were excellent. In some cases 4-5x faster than their XFree86 equivalents, if XFree86 even supported the same hardware! Red Hat bundled Metro-X to increase their supported cards. But that was all OK because Metro-X wasn't cheap. There was still an incentive to create the open-source equivalents in XFree86. And that was good because although it hurt initially we came out stronger because of it. Nobody bothers with Metro-X anymore; the combination of closed-source and a high price-tag forced the community to produce decent open-source equivalents.
The nvidia driver is different. It's not only good but it also comes at no cost. That's the worst possible combination. That's that will harm the growth of Linux. Not patents. Not lawsuits. Apathy and complacency are our enemies. The nvidia drivers encourage apathy because they're "good enough" and most people are willing to compromise their principles for the immediate gratification of 3D graphics.
And the fact that of all the responses I have had, not a single one has agreed with me, shows that there are very few of us who recognise that closed-source drivers are bad. As I said, I have no desire to stop you doing what you want. If you want to use closed-source drivers then you go right ahead. But I think it will slow the spread of Linux, not hasten it.
Oh yes, I agree, which is why I know the cost of Solaris is prohibitive. You need to get the Ultrasparc hardware and all the optional non-free software pacakges (Forte) to make it worth your while.
I actually have Solaris at home running on an ancient Ultrasparc 2. My day job involves Solaris admin work. I don't need any advice on what is and isn't good about Solaris. I've some small amount of experience in this area.
Yes, I would rather that the drivers did not exist, but I don't see how you manage to make the leap to "rather I didn't have the freedom to do what I want". I have no desire to impede your freedoms. I simply wish that the drivers did not exist, because then we would be forced to do something about it instead of slumming in this horrible limbo of complacency.
Absolute nonsense. I mainly use Linux because of cost. I'm a cheap bastard. I wanted a UNIX and I'm not willing to pay for a Solaris environment (the only other UNIX I would consider if I could not use Linux).
But the more I use Linux the more I appreciate that the reasons I like Linux stem from the ideology. Source code. Zero cost. No license daemons. Community and kinship. Best of breed. I realised after many years of reflection that all the benefits of Linux are because of the free as in freedom aspects of Linux. It is something you can only properly appreciate after you have drowned in Linux for years and then stepped back.
And that's your prerogative. I don't know why you (and others) have got the impression that I want to stop you. Do whatever you want. I'm in no position to tell you otherwise and nor do I have any desire to stop you.
Perhaps you think that because I said that closed-source nvidia drivers are more harmful than no drivers at all - and that's a statement I'm not backing down from - that you think I'm working actively to destroy nvidia! Don't be silly. I have no control over nvidia. They will release their drivers and you will continue to use them. Doesn't bother me. Never has. Never will.
But I stand by what I said. I strongly believe that in the long-run, closed-source nvidia drivers are harming Linux more than they are helping. They are preventing the development of open-source drivers. Any brilliant ideas in nvidia's drivers can not be reused in the ati drivers. Similarly, any brilliants aspects of the DRI cannot be used within the nvidia drivers because nvidia doesn't use the DRI. The closed-source nvidia drivers are buggy and we can't fix that; we are beholden to nvidia to fix the bugs for us. These are all detrimental aspects of the closed-source binary-only nvidia drivers. You'll hopefully notice that they are all pragmatic reasons, not ideological reasons.
I made the remark to highlight that Linux came from very humble beginnings - no graphics, no LVM, no RAID, no SCSI! - and that it didn't need binary only closed-source drivers to escape that mire of feature-less and driver-less poverty. It escaped solely because of the freedom aspects of Linux; because somebody with unsupported hardware could download the kernel source, write a driver for their particular hardware, and send the patch back to Linus. If Linux wasn't free as in freedom then it would never have grown beyond Linus's 386 with an IDE drive and text-mode only interface.
And cost. Never forget cost. The Real UNIX(tm) I was using before Linux cost an absolute mint. Linux was faster, stabler, had more features, came with source code and was cheaper. It really was no-contest.
I don't know why you would think that. I don't have any desire to tell you what you can do with your computer and your software. I said what I thought and what I would do. That was all.
What I find odd are people who put words into my mouth and attribute me with opinions that I don't agree with, and then attack these phantoms as if that would invalidate my real opinions.
That's nice. I paid $20 for a card that worked with the default Debian install. But you do whatever rocks your world.
Yes, that is over the top, but it's not what I said. Read it again. I expressly said "Not for the same code but it's the same principles". It's the principles that you sacrifice, not the whole operating system. If you don't appreciate that the principles are important then you probably won't understand.
It's up to the individual to make their own decision. If someone wants to use closed-source binary-only drivers then I won't stop them. It is their freedom that they are throwing away.
How far do I personally take this? Where do I draw the line? For the record, if I have hardware and there is no open-source driver then I simply don't use the hardware. Or I sell it. Or I write the driver. I have stuck to these principles ever since switching to Linux.
For example, I bought an nvidia TNT2 several years ago. I was fooled by nvidia's claims that they were going to support Linux. They told the truth - they did support Linux - and the mistake was mine because I incorrectly assumed nvidia was going to release open source drivers. But I still felt cheated. I joined the Utah GLX project to write drivers for the TNT2 and I think the result was successful enough. I later sold the card and bought something that was properly supported by open source XFree86 drivers.
I don't have a problem with "imposing limits" on Linux users. I remember using Linux when there were no supported video cards at all (before X386 was ported). You had text mode, and that was it. Linux grew beyond those days without requiring binary-only drivers. We wrote open-source drivers as required and Linux grew from basically nothing on the strength of open source. Why do we need to change the winning recipe now? What has changed so that we suddenly need binary-only closed-source drivers, or Linux will fail? I don't see it. Open source got us this far. It'll take us the whole way. It may take slightly longer, or limit our choices of hardware, but so what? I see no reason to change strategy now. I see no reason to compromise the ideals to fulfil a crass gratification.
Hard to believe you're not a troll but just in case...
The Linux community wants SCO's "intellectual property" out of Linux. The GPL doesn't permit their IP to stay in there. It must be removed; the license and the law demands it.
But first they have to tell us WTF their IP actually is. We can't remove it because we don't know WTF they're talking about! Even IBM doesn't know and the judge couldn't figure it out either, which is why SCO has been compelled to tell the court WTF they think they own in 30 days time.
Linux developers are very respectful of IP rights, even if sometimes the users (*cough*MP3 thieves*cough*) are not. This is why we wrote our own operating system, from scratch, rather than subject ourselves to binary-only licensing.