I have no problem with being reality-checked and approve of the "see for ourselves" spirit in your comment. The text of the applicable portion of the agreement is in the comments following the Technocrat article. You might also ask some of the beta-testers.
I called the Corel switchboard when I first found out about this issue, and was put through to the Linux marketing person, who answered on the first ring and took charge of the problem right away. I subsequently spoke with the Linux product manager and with one of the engineers, who is also prominent in their local LUG. All of the Corel staff I spoke with were extremely polite and cooperative. Besides talking with me, they also corresponded with Debian and KDE folks, and they read your opinions on Slashdot.
Corel is a huge company, they claim to be right behind Microsoft in size in the market sector they address, and they are the largest software company in Canada. For a company of that size to act on an issue like this in 4 days is excellent. That's how long it took from when they first became aware of the problem to when they made a policy-change announcement. I got the impression that they sincerely cared about what the Free Software community wanted.
I'm also pleased that they based their system on Debian. It's nice to see that Debian is finally being appreciated for the fine system it is, by Corel, by SGI/VA/O'Reilly, by Storm Linux, etc.
Pixar's specialty is not the absolute leading-edge of computer graphics, although they certainly do innovate and try to be on the leading edge. Their specialty is story. They could probably have done Toy Story II with Toy Story I technology and people would still be watching it 20 years after FF8 is forgotten, for reasons that have nothing to do with computer graphics and everything to do with the story. That doesn't mean that you might not like FF8 much better than TS2. FF8 is an "adrenaline flick". There's a time in young men's lives when nothing can beat an adrenaline flick.
But this is not to say anything, positive or negative, about the graphics in FF8 vs. Pixar. I really haven't looked.
I'm afraid you're confused. Film production, for the most part, doesn't involve tasks that would need real-time services from the operating system. The only ones I can think of are data-collection from rotoscoping devices (rotoscoping was anathema at Pixar, though), film recording, and video recording. At the NYIT CGL (Pixar's predecessor) we had some software that would look at the horizontal sync and timecode and tell VCRs when to record - that was before you could get a single-frame recorder from the factory, so we had to hack the hardware. We used PDP-11 V6 Unix for that job, and did the real-time programming at interrupt level in the device driver. If you do that, you can get real-time service from Unix or Linux, but it's probably not as easy to program as QNX. I did that VCR control software for years, but the original hack was by Bruce Laskin and Karre Christian, I think.
The render-farm only runs Renderman and other Pixar-internal software. No interactive software there, so no concern about the modeling programs. There is no reason they couldn't use Linux if they wanted to, but there's not much reason for them to use Linux, either because once they have a Unix-like system supporting their batch-processing software what Unix it is doesn't matter. Sun made that deal very attractive for them, I doubt they paid software license fees. And Steve wants to sell Macs, not Linux. In the several discussions we had about Linux before I left Pixar, Steve hadn't "gotten it": he still thought it took a mega-corporation to make a good desktop and that Linux had no hope of ever getting a good GUI.
I find it amusing to watch SGI go for Linux and Debian, because it might end up that my old software group at Pixar will use my hobby project as their interactive operating system. Some of them were quite resistant to Linux when I left.
Gee, O'Reilly's actually coming out with an Open Source book (I'm assuming none of the options are checked on the Open Publication License) and they're giving profits from the book/CD package back to Debian. It sounds like a very positive change.
As a free software author, what reason do I have to facilitate non-free software? Sometimes, I wonder if the people who ask for BSD licensing really only want other people to use BSD licensing, and never intend to BSD-license their own stuff.
The GPL is more balanced. It says "You want to use my software? Then give me the same license terms I gave you." That says quid-pro-quo to me. The BSD license, in contrast, says "use me and don't give anything back". That's hardly a quid-pro-quo.
Well, I sure wouldn't think of suing anyone before giving them a chance to make it right. And I agree that it's probably a simple mistake. The fact is, I've never come close to suing someone over a GPL violation, they always fix the violation. But I think I might need a translator to talk with these guys. Any volunteers?
You've got to show me evidence that they distributed the binary first. Or just show me the binary, or a machine containing the binary. If they are just "publicly performing" but not distributing, that is not enough.
I'm still not sure "distribution" applies to people who are clearly your employees, and in any case I don't think even RMS wants to apply it this way. But I'd be willing to discuss it further with an attorney.
i) Who is the complainant in the trial? As you point out, the kernel is copyright by thousands of people and the scope of this copyright is very blurred (does someone really own the one line of code they changed). Does this mean that the entire group of copyright owners have to sue 'en masse'?
At least one author would be the complaintant, and would claim infringement of a specific part of the kernel that he clearly wrote. A single function would be enough.
ii) Does this mean that you have to give back the binary because they had no right to distribute in the first place (ie you are doing yourself harm by complaining)?
I don't think that under the GPL a distributor can compel someone to whom the product was distributed to destroy the product. But the distributor has infringed even if the product is later destroyed.
iii) What is 'reasonable cost'? Could a company legitamately claim that the cost of production of a one-off CDROM or DVDROM is of the order of $100,000 and charge that for the source? After all, commercially pressed CDs are the norm for software distribution.
$100,000 is not a reasonable cost. At the most you could charge for the media and an hour's labor by an operator to write the medium.
I'm not wrong about that. Read the GPL around this part:
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
That applies if you don't distribute source code with the product.
First, the tactic he outlined would itself be illegal, and to willfully do it would probably be fraud, which can get you jail time. Second, nobody who does business in the Linux community wants the bad P.R. that comes with being a GPL-violator, so we never get to court on these things, they fix them first.
If they give someone outside of their company a beta-test version, and that beta-test version doesn't have the source code or an offer for the source code (which is fulfilled promptly), they would be violating the GPL.
A lot of people have trouble understanding that beta-tests are distribution. But when you transfer a copyrighted work between two different legal entities, that's distribution. So, the first time beta code goes to someone who doesn't work for your company, you must fulfill your GPL obligations. Your intent to distribute source later doesn't matter, it's a violation now.
When the GPL is violated, the person who would bring suit is the copyright holder. In the case of the kernel, this would be several hundred people and organizations - but you just need one of them. Not that I am holding forth any hope about suing in China, but suing a Chinese company that does business in other nations is certainly possible.
But again, we don't know yet if these people are just confused about the GPL as Corel was, and think they can wait for the end of their Beta tests to release source or something.
You don't understand patents. You could patent a particular algorithm for sorting (in the U.S. where software patents are allowed) but you could not patent the "sort" program itself. You could also not successfully patent something that has alredy been published.
Regarding copyright infringement lawsuits, given all of the money behind Linux these days, one would think that we would be able to find significant financial support for enforcing our copyrights, as we did in correcting the Linux trademark issue. There is also some chance that we can bring a criminal, not civil, prosecution in some forms of copyright piracy.
But the sad fact is that it may simply not be possible to enforce the GPL in some nations, China included. You certainly can enforce it, though, if the Chinese company decides to sell its product in another nation that has conventional copyright laws.
But we still haven't seen if the company is simply confused about the GPL.
If you get the binary product and it does not come with an offer for source code, that's a GPL violation right there.
Of course, I have not seen the product and will need to examine it along with someone who can read Chinese before I can say for sure that the GPL's being violated.
First, get a copy of the binary software. Preferrably you should get it directly from the manufacturer, and in a nation where you have a hope of bringing a lawsuit to enforce the GPL. The United States would be best for this.
After you have gotten a copy of the binary software, you must see if there is source code for all GPL software, or if there is a clear written offer for the source code distributed along with the binary software (perhaps on the CD itself). If neither of these exist, the GPL is being violated.
Then, you should contact the manufacturer and ask for source code (for the kernel modifications, or whatever). Then if you don't get the source code, that's a GPL violation too.
Don't listen to anyone who calls it "whining" to protest a GPL violation. Remember that the Linux kernel and applications are copyrighted property of some thousands of contributors, and they have a right to enforce their license, which requires that source code be distributed upon request.
Under the GPL, if source code is not distributed with the product, the distributor of the binary code must give source code to anyone who requests it, not just people who have received the binary code. But we must see a copy of the binary product to determine conclusively whether or not source code is distributed.
Well, I could suggest you walk a mile in my shoes...
I've done some stupid stuff in my time - although I don't know what particular things you are referring to, maybe it's something I'm still proud of:-) . You never have done anything stupid? Or perhaps you were lucky enough to not be in the public eye, and thus when you did stupid stuff fewer people knew about it. Then, you're lucky. I'm lucky too, because I've actually learned from some of these things.
Gee, maybe if you understood my reasons you wouldn't think my actions were insane.
Eric has put forth some good reasons why Open Source works. He's also put out a good deal of pseudo-anthropology that could be disproved without casting any doubt upon why Open Source works, and that's where his paper is most easily attacked. Debates like this are good recreation but they don't say much about the practical issues of Open Source.
Constructive solid geometry, spline meshes, etc., have been used by Pixar's Renderman and other similar products for about 10 years, and Renderman definitely shades surfaces better if you use them rather than a polygonal model.
If there is any innovation in this article at all, it's moving this technology down to the level of a video game from the non-real-time applications where it's been used for decades.
Slashdot Mirror
Bruce
Thanks
Bruce
Corel is a huge company, they claim to be right behind Microsoft in size in the market sector they address, and they are the largest software company in Canada. For a company of that size to act on an issue like this in 4 days is excellent. That's how long it took from when they first became aware of the problem to when they made a policy-change announcement. I got the impression that they sincerely cared about what the Free Software community wanted.
I'm also pleased that they based their system on Debian. It's nice to see that Debian is finally being appreciated for the fine system it is, by Corel, by SGI/VA/O'Reilly, by Storm Linux, etc.
Thanks
Bruce Perens
Thanks
Bruce
But this is not to say anything, positive or negative, about the graphics in FF8 vs. Pixar. I really haven't looked.
Bruce
Bruce
I find it amusing to watch SGI go for Linux and Debian, because it might end up that my old software group at Pixar will use my hobby project as their interactive operating system. Some of them were quite resistant to Linux when I left.
Bruce
Gee, O'Reilly's actually coming out with an Open Source book (I'm assuming none of the options are checked on the Open Publication License) and they're giving profits from the book/CD package back to Debian. It sounds like a very positive change.
Thanks
Bruce
The GPL is more balanced. It says "You want to use my software? Then give me the same license terms I gave you." That says quid-pro-quo to me. The BSD license, in contrast, says "use me and don't give anything back". That's hardly a quid-pro-quo.
Thanks
Bruce
Bruce
Thanks
Bruce
Thanks
Bruce
Just delete the "But" and read the paragraph without it.
Thanks
Bruce
Thanks
Bruce
At least one author would be the complaintant, and would claim infringement of a specific part of the kernel that he clearly wrote. A single function would be enough.
ii) Does this mean that you have to give back the binary because they had no right to distribute in the first place (ie you are doing yourself harm by complaining)?
I don't think that under the GPL a distributor can compel someone to whom the product was distributed to destroy the product. But the distributor has infringed even if the product is later destroyed.
iii) What is 'reasonable cost'? Could a company legitamately claim that the cost of production of a one-off CDROM or DVDROM is of the order of $100,000 and charge that for the source? After all, commercially pressed CDs are the norm for software distribution.
$100,000 is not a reasonable cost. At the most you could charge for the media and an hour's labor by an operator to write the medium.
Thanks
Bruce
That applies if you don't distribute source code with the product.
Bruce
Thanks
Bruce
A lot of people have trouble understanding that beta-tests are distribution. But when you transfer a copyrighted work between two different legal entities, that's distribution. So, the first time beta code goes to someone who doesn't work for your company, you must fulfill your GPL obligations. Your intent to distribute source later doesn't matter, it's a violation now.
Thanks
Bruce
But again, we don't know yet if these people are just confused about the GPL as Corel was, and think they can wait for the end of their Beta tests to release source or something.
Thanks
Bruce
Regarding copyright infringement lawsuits, given all of the money behind Linux these days, one would think that we would be able to find significant financial support for enforcing our copyrights, as we did in correcting the Linux trademark issue. There is also some chance that we can bring a criminal, not civil, prosecution in some forms of copyright piracy.
But the sad fact is that it may simply not be possible to enforce the GPL in some nations, China included. You certainly can enforce it, though, if the Chinese company decides to sell its product in another nation that has conventional copyright laws.
But we still haven't seen if the company is simply confused about the GPL.
Thanks
Bruce
Of course, I have not seen the product and will need to examine it along with someone who can read Chinese before I can say for sure that the GPL's being violated.
Bruce
After you have gotten a copy of the binary software, you must see if there is source code for all GPL software, or if there is a clear written offer for the source code distributed along with the binary software (perhaps on the CD itself). If neither of these exist, the GPL is being violated.
Then, you should contact the manufacturer and ask for source code (for the kernel modifications, or whatever). Then if you don't get the source code, that's a GPL violation too.
Don't listen to anyone who calls it "whining" to protest a GPL violation. Remember that the Linux kernel and applications are copyrighted property of some thousands of contributors, and they have a right to enforce their license, which requires that source code be distributed upon request.
Under the GPL, if source code is not distributed with the product, the distributor of the binary code must give source code to anyone who requests it, not just people who have received the binary code. But we must see a copy of the binary product to determine conclusively whether or not source code is distributed.
Thanks
Bruce Perens
I've done some stupid stuff in my time - although I don't know what particular things you are referring to, maybe it's something I'm still proud of :-) . You never have done anything stupid? Or perhaps you were lucky enough to not be in the public eye, and thus when you did stupid stuff fewer people knew about it. Then, you're lucky. I'm lucky too, because I've actually learned from some of these things.
Gee, maybe if you understood my reasons you wouldn't think my actions were insane.
Bruce
Thanks
Bruce
If there is any innovation in this article at all, it's moving this technology down to the level of a video game from the non-real-time applications where it's been used for decades.
Bruce