Castle Denies GPL Breach

Anonymous Coward writes "Castle Technology, who were accused of breaching the GPL in RISC OS 5, have made a press release denying the allegations. This story has been covered on The Iconbar RISC OS news and resource site." We've given Castle some loving here on slashdot recently. Looks like this one isn't going away quietly.

Re:Where did the accusation come from..

[Anonymous Coward]

It was originally posted on the Linux Kernel Mailing List (lkml) by Russel King, here [theaimsgroup.com].

Re:Where did the accusation come from..

[Dr. Ion]

I've always been under the impression that borrowing code from a GPL based package was acceptable, as long as credit is given where credit is due.

Uh, no. You need to go read your license agreement, usually called "COPYING" in the source tree. It's quite readable, and very clear about what you must do to "borrow code". It's more than just giving credit.

After all, that GPL code you're borrowing.. I'll bet you find it handy that you have the entire source code, and not just a useless footnote giving credit to some author. You are expected to pass on the same freedom with your software, not a 'credit'.

Re:Where did the accusation come from..

[mobiGeek]

But I've always been under the impression that borrowing code from a GPL based package was acceptable, as long as credit is given where credit is due

Your thinking about another license altogether. With the GPL, you can't "borrow code" and keep it locked up inside of your code base.

GPL "opens" code. It does not allow people to "close up" the code. In a nutshell, if you use GPL code, you must make your source code freely available to whoever receives your binary. (There are a number of subtle points beyond this, but this is the GPL in a nutshell).

If you aren't going to make your source code available, then don't include GPL code with your code. It's simply a matter of choice.

Re:Will this be the first GPL test case?

[will_die]

Except that the GPL gives you rights that you would not have, under normal laws dealing with the code, the copywrite laws.
So if the GPL is ruled invalid the sourcecode would fall under the copywrite and castle or microsoft would have the same rights the that code that you do to microsoft code.

Re:Confusion

[Anonymous Coward]

Actually, that is perfectly alright and compliant with section

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,

The problem is not that they would use floppy disks (which are pretty much the most frequent medium for software interchange, still), but that they apparently conveniently forgot to include that written offer!

In addition, it would seem that we have forgotten another few phrases:

a) You must cause the modified files to carry prominent notices
stating that you changed the files and the date of any change.

b) You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any
part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.

c) If the modified program normally reads commands interactively
when run, you must cause it, when started running for such
interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a
notice that there is no warranty (or else, saying that you provide
a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this
License. (Exception: if the Program itself is interactive but
does not normally print such an announcement, your work based on
the Program is not required to print an announcement.)

Re: Confusion

[markov_chain]

Read the press release more closely: "based in part on the following functions" could mean that they just looked at Linux code and then wrote their own from scratch.

Re:Well, denies breach, but...

[Christopher Craig]

Yes, it is a breach of at least two sections of the license:

1. The GPL requires that any work derived from GPLed code be released under the GPL.
2. The GPL requires that any copy of GPLed code must be distributed either with the source code, or with written notice of a system good for three years from the date of distribution of distributing said source code at no cost greater than the distribution cost. (I presume, of course, that they did not in fact distribute such a notice)

I'll readily admit I know nothing about this case, but most people who knowingly choose to release GPL code do so specifically because they intend code release to be the cost of deriving from their work. To the people who will inevitably whine about how the GPL restricts author's rights: The GPL contains a provision that you aren't even required to accept the GPL, but the author only agrees to let you modify, redistribute or derive from his code if you do.

Re:Confusion

[Target Drone]

No, they are saying that they did not use any GPL code.

The RISC OS ... kernel did not contain work taken from or derived from the ARM-Linux or Linux kernel."

If you mail them a floppy you get a copy of components source code that allegedly violates the GPL so that you can see for yourself that it's all-legit. We'll just have to wait and see if anybody gets a copy of the code mailed back to them and if it violates the GPL or not. Personally I find it fishy that they just don't post it on the web right now to clear their name. I suspect the mail in a floppy is a stall tactic.

Re:Seems reasonable

[Anonymous Coward]

"They say its not used GPL code in some old editions" - They say GPL code has not been used in RISC OS 5.00, 5.01 or 5.02. Those are the only versions of RISC OS where this GPL code could possibly have been used because there are no other versions of RISC OS which include the PCI manager (and it is the PCI manager which was alleged to contain GPL code).

"Its not clear if there is some release they did." Yes it is. They mention all the releases in question and clearly state that GPL code has been used in none of them.

"They don't say they havem't done it with current code." Yes they do. RISC OS 5.02 is the current code and they say they have not used GPL code in it.

Liars. Google cache proof.

[Anonymous Coward]

The Google cache of the (now removed) page http://www.iyonix.com/32bit/PCI_API.shtml
is at http://216.239.53.100/search?q=cache:mf1nlduliL4C: www.iyonix.com/32bit/PCI_API.shtml+&hl=no&ie=UTF-8

Note that the source code for many of the Linux PCI device drivers is publicly available on the Internet and may be useful in developing the corresponding RISC OS device driver.

So, if they had clean conscious, why would they remove that page?

I don't buy into this.

Re:Will this be the first GPL test case?

[Christopher Craig]

Without a license, you cannot use copyrighted material. If you use copyrighted material, without a license, you are in violation of that copyright. The only matters before the court would be "did you use the software" and "are you licensed to do so".

This isn't technically true. You are allowed to use copyrighted materials without a license, but you aren't allowed to copy, distribute, modify or derive from copyrighted material without a license. The DMCA has restricted this a little more so that now a company can require you give up some of the rights you would have had without a license in order to use their product (which is one of the huge issues with the DMCA), but the GPL doesn't rely on this.

The GPL specifically provides that you are not required to accept the license, but without accepting it the author gives you no rights above the standard ones provided by copyright law (pretty much reading the source code and running the software)

Re:Confusion

[ucblockhead]

It's the second part that's important...from their press release, they seem to think that they can just give you the source for the routines they copied. You can't even link to GPL'd code without GPLing your code. (The LGPL is a different story, though the Linux kernel isn't LGPLed, so that's moot.)

Calm down...

[zjbs14]

They're not releasing that piece of source as part of some GPL requirements, they're releasing to show that it's not covered by the GPL. From the press release (empahsis mine):

For the avoidance of doubt, the hardware abstraction layer (roughly analogous to a PC's BIOS) has it's PCI allocation and bridge setup based in part on the following functions from the Linux kernel sources

I admit that it could probably be worded better, but it sounds like they could have took the function names/possibly signatures and wrote their own code. Get the source and find out. However, if the experts in this matter can still show that the object form is too close to the GPL output, then there may be something to worry about.

Re:Will this be the first GPL test case?

[kyz]

No, this will be a plain old copyright test case.

Over and above the rights you have to the Linux kernel under copyright law, the GPL offers you further rights (such as the right to make a derived work) provided you agree to and follow the GPL license.

If you do not follow the license, you are not entitled to those further rights, and you may find yourself in breach of copyright law.

Castle's choices in this case are:

• Castle can argue that they were complying with the GNU GPL, which allowed them to take kernel source code and use it in RISC OS (i.e. create a derived work from the kernel PCI code). This, however, is clearly not true, as they are not distributing their entire derived work (RISC OS) under the GPL. It would be different if the kernel PCI code was packaged by itself and available under the LGPL, or they had secured permission from the copyright owner of the PCI co de to use this code without following the GPL, but they didn't.
  
  
• Castle can argue that they had made "fair use" of the Linux kernel source code, and therefore they are not breaking copyright law. However, if the PCI code in the RISC OS ROMs "substantially" copies from the Linux PCI code, then this is not true.
  
  
• Castle can argue that they did not make "substantial" copying of the kernel PCI source. Given they have admitted that their PCI code is "based on" the kernel PCI source, that makes them look bad. What is "substantial" has to be decided by the court, but in previous cases I've read about, substantial would be anything more than a few lines of code.
  

Re:Where did the accusation come from..

[Anonymous Coward]

also it is here [mcgroarty.net].

is it true it would be the first gpl court test?

Re:Liars. Google cache proof.

[Anonymous Coward]

Check again - it's still there, including the legitimate line about using Linux driver sources to develop RISC OS drivers (which may be separate components and may be GPL'd):

http://www.iyonix.com/32bit/PCI_API.shtml

Re:Will this be the first GPL test case?

[ChaosDiscord]

The GPL is a license to software. Plain and simple.

Without a license, you cannot use copyrighted material. If you use copyrighted material, without a license, you are in violation of that copyright. The only matters before the court would be "did you use the software" and "are you licensed to do so".

You are operating on a popular but completely incorrect belief. This incorrect belief grants copyright holders far more power than the law really gives them. Copyright industries want to encourage this erroneous belief but we need to fight back.

You do not need a license to use material protected by copyright. If I buy a book, a DVD, or a CD I'm free to take it hope and read it, watch it, listen to it, loan it out to a friend, destroy it, give it away, or sell it. No license is needed or granted. The particular item that I purchased is mine, the copyright holder no longer has any claim to it. What the copyright holder does have claim to is the exclusive right to make and distribute copies. (Well, the right to perform publically is also in there, and there are lots of complex exceptions, but that's the gist of it.)

This is important and many people seem to have missed it: You do not need a license to personally use (read, watch, run, listen to, whatever) a copyrighted work you purchased.

Given this, the GPL is not a license to use the software. You're free to use software under GPL without ever reading or agreeing to it (but you should probably note the "NO WARRANTEE" clause). You can refuse to agree to the GPL and use the software. The GPL only seriously comes into play if you want to distribute copies. Normally under copyright law you cannot ever distribute copies. The GPL is an open offer to let you distribute copies (granting you more freedom than copyright law normally allows), in exchange for certain behavior on your part.

Normal software licenses attempt to change your purchase of a particular thing restricted by copyright into license of something you don't own. This is completely alien to the United States copyright system. In a similar case much earlier (around 1900 if I remember correctly) a publisher tried to put a license on an actual book. It was soundly defeated in court. The legal precedent for software End User License Agreements is pretty shaky, primarily resting on a single case at a lower court level (district?) that rather insanely decided that copying a program into memory to run was an infringing copy and as such required a license. It could yet be overturned. If it does get overturned traditional software will revert back to the same rules books, CDs, tapes, and DVDs live with and do fine under. The GPL will continue to work fine because it already assumes you have every right under copyright law but offers you a license to do more than copyright law allows.

Re:Will this be the first GPL test case?

[ctid]

You do not understand what you are saying.

If there was no GPL on the code, you would be able to do less with the code than the GPL allows you to do. Applying the GPL to a work removes restrictions. For example, without the GPL, you would have zero rights to copy it.

Re:Come on mods...

[zjbs14]

Yes, and so does the page which is still at http://www.iyonix.com/32bit/PCI_API.shtml [iyonix.com]

The point is that the page has not been removed/altered/etc. and that before modding accusatory messages up, people should take 5 seconds and check the link.

Re:Only 5.0, 5.01, 5.02 and beyond mentioned

[Anonymous Coward]

The bulk of it predates Linux, so you could argue that RISC OS 5.XX (even if GPL code HAD been used in the HAL) predates Linux and therefore can't have been "derived" from GPL code.

Doesn't matter. The moment you incorporate GPL code into it, it's GPL-derived. If you want to use GPL code without the licensing hassle, you must abstract it out and link to it externally. Can't handle that? Then write your own fucking code.

In short, no

[Kjella]

The people trying to enforce the GPL are in fact very kind. They usually leave you two choices

1) Release it under GPL
2) Admit to breaching it (by accident or intentional), settle for a full press release and removal of the infringing source code

If they really wanted to be assholes, they could simply file charges for copyright infringement, Usually, if they not only copied the code, but presented it as their own work they could be sued with fraud too. And (if found guilty) the company would have to pay damages. Complying with the licence after the crime would not free them from any liability.

So you see, releasing it under GPL is a settlement offer from the copyright holders. They don't have to make the offer, and the company doesn't have to accept.

Kjella

Re:Hold On.

[Corydon76]

can somebody just say what the evidence is?

You missed the second part of the allegation. Not only were Linux kernel function names found in the binary, but after making several insignificant changes to the Linux source, the resulting binaries were identical to Castle's ROM code.

There's more than one way to skin a cat, but when you get binary-equal results, it's highly likely that somebody copied.

Re:Will this be the first GPL test case?

[KDan]

Yup, but in the paid GPL case, the GPL actually extends the rights you would expect to have from buying a box with software in it, whereas if you buy a usual proprietary software, the licence decreases the rights you could expect to have on that software (eg the right to resell it).

A good comparison is a book, which has existed for so long that the rights you expect to get from buying a book are well established. You buy a book, you expect to be allowed to lend it, you expect to be allowed to sell it second hand. You buy a GPL'ed software, you have the extra rights that you can alter the book, reprint it, redistribute it, etc. You buy a proprietary software, you lose the right to lend it, and in some cases you even lose the right to resell it second hand!

Hence the decrease/increase perspective.

Daniel

Re:Will this be the first GPL test case?

[dark_panda]

Section 5 of the GPL: "5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License."

So, yeah, you can use linux or any GPL'd software without accepting the terms of the GPL. You just can't modify or redistribute it in binary or source unless you follow the terms of the GPL.

J

Re:Fuckin hell. that's some shit

[Anonymous Coward]

Why not actually RESEARCH the claim? Oh, wait, you arn't too bright are ya?

http://216.239.37.100/search?q=cache:618Ah-5N0gk C: web.gnu.walfield.org/mail-archive/linux-kernel-dig est/2000-July/0112.html+gpl+violation+webplayer&hl =en&ie=UTF-8

Thanks for showing you are clueless.