Today's SCO News

joebeone writes "Linus has commented on the SCO v. IBM suit saying "SCO is playing it like the Raelians" and that he will withhold his judgement until the code in question is shown in court. He has also recommended that former slashdot editor, Chris DiBona, be appointed to a panel offered by SCO to examine the evidence." Businessweek has an interview with SCO's CEO. The Open Group would like to remind everyone that SCO is only one of many in the Unix world.

Comments

Something Mismatches

[tanveer1979]

Yea the SCO is firing on all counts against Linux. But there are certain strange anomalies as far as their India Division is concerned. SCO India is apprently still pusihing linux!

The May Issue of Linux for You India has interview of SCO India Head in which that guy is pushing linux and says linux is the key focus of SCO with they wanting to contribute to the Linux Community by way of more software. Isnt that a bit odd!

SCO's own goal

[bazik]

Anyone else noticed that SCO continued to sell their Linux distribution for two months _AFTER_ they sued IBM? They even had a kernel source code on their servers available for download >:)


For more information click here.

Re:considered the father of Linux?

[aborchers]

This SCO guy raises one really disturbing question. What if someone has, in the past or future, malicously or accidentally, injected proprietary code w/ copyright or patent entanglements, into core Linux systems? What are the implications for users who have no way of recognizing the code in violation? Is this really a serious flaw in the open source model?

SCO's Ineffable Fallacy

[codefool]

"If trade secrets are the issue, it wouldn't be hard for the Linux community to recode the offending software."

This is precicely why SCO does not divulge exactly what's in question: it would be too easy for IBM et al to say "Oh. So sorry. Many regretti." and recode it, thus deflating any hope they have for the Home Run.

All SCO can be after is money - QED.

Re:show us the CODE!

[DailyGrind]

Actually the executives are going somewhere.... have you seen the stock price.

Unfortunately in the U$ the end justifies the means and the executive options for SCO are now making them rich.,.

It is not important if they win as long as they can cash out before the outcome....

Re:show us the CODE!

[Mr2cents]

Well, the article states it is:
- it is added in the last 18 months
- it is added by IBM
- they added 'a whole program'

I think this narrows things down a bit, shouldn't it be possible to make a list of code added by IBM in the last 18 months?
Maybe we could all put a snippet of this code on our website, and the one who gets sued by SCO has the right part :)

Re:considered the father of Linux?

[aborchers]

No, I mean in general, not just in this SCO case. Anyone contributing code could theoretically copy and paste unattributed commercial code into the system. Since it's proprietary, noone (say, on the kernel group) reviewing it would recognize it and it might get approved and passed into the kernel. The person who did the paste would certainly be guilty of copyright infringement, but is that liability passed on to every user of the infringing derivative work?

Even if SCO shows the code

[mocm]

how can they prove that they did not take the linux code and incorporate into their code.
The dates and times of code inclusion into linux are pretty well documented, but how can you do that with closed source.

How come...

[Anonymous Coward]

it is ok for the open source crowd to simply remove the offending code, but not for those who have GPL violations, like OpenTV to simply remove their offending code?

Violation, you scream. You must pay the consequences, no matter how the code got in there.

But, if it is the open source crowd that put the bad code in, then just remove it, it's all good.

More open source double-standards.

There's lots of BSD code in AT&T Unix

[fstanchina]

The article from Newsforge had this (anonymous) comment attached to it. I didn't think of this, but it looks like he's right.

[disclaimer: as stated above, this comment was written by an Anonymous Reader -- I'm just pasting; any positive moderation doesn't belong to me, except a "+1 informative" if you will]

"As people may recall from the original settlement of the BSD lawsuit, three files had to be removed from BSD that represented things in SysV source. What is often forgotten, though, is that AT&T itself was in a far greater bind because while there was some SysV code in BSD, there was a LOT of "borrowed" and misattributed BSD code found to be in AT&T SysV. BSD permits this, but the license at the time required the advertising clause, and AT&T fraudulently ignored this. The actual settlement said that AT&T would no longer sue the BSD people, and that the University of California would also agree to hold AT&T harmless for misappropriating BSD code. Hence, much of the code that SCO owns is actually misattributed BSD code for which UC permitted AT&T (and it's decendents) to use."

"Now much of Linux also shares code derived from ancestrial BSD sources or people who have worked in common on both, and I am sure many of the same ancestrial routines still found today at the core of SysV are in fact also BSD derived. Hence, where common code may exist, it's code that AT&T originally misappropriated, and that SCO is free to use and relicense from the AT&T/BSD settlement, but in point of law neither AT&T nor the current SysV owner has actual legal copyright over. Perhaps the regents of UC could hall these SCO scum back into court, as they are in fact in material breach of the AT&T/BSD settlement if SCO now claims copyright "ownership" of that originally misappropriated code since the settlement gave AT&T no such rights."

Re:Key quote. My question: how to remedy?

[egoff]

Another thing that quote brings up is what program? Can anyone who has a SCO UNIX license go program by program through the code in Linux and find any possible culprits? If SCO won't show the code, perhaps we can find (or not find) anything based on these clues. Anyone with a copy of SCO's UNIX should be able to do this, even if you can't program you can at least be able to tell, "This block of mumbo jumbo is the same as that block". The results of a study like this would be very valuable

Re:show us the CODE!

[jgerman]

Are those statements true? I kind of hope so. Linux is just the kernel (name flames aside, technically it's true). So it's not possible that anyone can add a whole program to a monolithic kernel. That would indicate that they are actually talking about something included as a program/utility/whatever in the OS by the distributor (I'm assuming this is IBM). If it's even true that a whole progrm was released under GPL when it wasn't supposed to be, it's easy to excise it from any distributions that are including it. It's certainly not a core piece of the OS, GNU and Linux are enough to create a useable OS/Linux distribution.

FSF's Bradley Kuhn comments on SCO case

[vinsci]

In an e-mail interview with the Sydney Morning Herald, FSF's executive director Bradley Kuhn says several interesting things, for example:

"Indeed, FSF holds documents from SCO regarding some of this code. SCO has disclaimed copyright on changes that were submitted and assigned by their employees to key GNU operating system components."

and earlier:

"SCO was not merely a distributor of the kernel named Linux; they were the distributor off the entire GNU/Linux system, which includes Linux as well as the core components of the GNU operating system, such as glibc, GCC, GDB, etc.

"Most of the core GNU components are all copyrighted by the Free Software Foundation and distributed under our auspices under GPL. SCO's right to redistribute them, and Linux too, is the GNU GPL and only the GNU GPL."

[...]

"For nearly two decades, the FSF has carefully and arduously collected copyright assignments on each contribution to the GPL'ed programs on which we hold copyright. We carry out due diligence to ask contributors if they have any reason to believe that trade secrets, patents, or other copyright claims cover their work before they submit it to us. We then collect a copyright assignment from the contributor (and a copyright disclaimer from their employer when necessary) to ensure that we hold proper title to the software on which we place our copyright notice and license freely under GPL or LGPL.

"Individuals and companies using FSF copyrighted programs know as much as one can know that the software has been examined carefully, that its authors certify that the work is their own, and that the authors have no knowledge of other claims conflicting with its licensing under GPL or LGPL."

For several other interesting quotes, see the whole interview.

Names dropped by SCO...

[Cranst0n]

Has anyone else noticed that in all the reports, press releases, and other things related to this frivolous lawsuit, two major playres are not mentioned by anyone. Mandrake and Debian don't appear in anything SCO has released. Is this just a coincidence or is it intentional? And either way.. why not warn them and their users also, why only talk about Red HAt and SuSe? Just a thought....

Re:show us the CODE!

[Eunuchswear]

They've said they will, soon.

SCO ship a bunch of Linux stuff with their "LKP", (Linux Kernel Personality) add-on for UnixWare, and according to The Inquirer they've written to say:

Secondly we [SCO] have begun examining the Linux RPM CD which we ship with the UnixWare Media Kits to expunge any material which is thought to have any IP issues...

Within a few weeks we expect to have completed the purging of the Linux Kernel RPM and will be remastering a new CD which will allow customers to begin receiving UnixWare 7.1.3 Media Kits with the needed LKP materials.

So, within a few weeks we'll be able to compare the old "Linux RPM CD" with the new one and find out what SCO/Caldera think was stolen.

Re:considered the father of Linux?

[cygnus]

What if someone has, in the past or future, malicously or accidentally, injected proprietary code w/ copyright or patent entanglements, into core Linux systems?

There are no implications for the user, period. If someone uses code he is not allowed to, it's his problem and nobody else's. And this applies to all licenses and all development models equally.

i believe the question was asked because there are parties that might see fit to sabatoge Linux and other Open Source projects. Not to say that the Open Source/Free Software methodologies a priori are on shakier legal ground, but that they're possibly more vulnerable to sabotage given malicious third parties.

Re:Can the paranoia and defensiveness

[Surak]

My question wasn't meant to be whether users would be criminally liable, but whether the product they were using would be considered illegal and subject to any kind of claim by the copyright owner. In property law, one who purchases a stolen item isn't entitled to keep it once it is found to be hot, even if they bought it unaware.

Hmmm... say you own a legally purchased copy of the song 'My Sweet Lord' by George Harrison. This lawsuit comes up. Are you now required to return your copy of 'My Sweet Lord' because it is now considered 'stolen property'?

The answer is 'no, of course not.' Because copyright violation and theft, despite what the BSA wants you to think, are not the treated the same under the law. Copyright violation is not theft per se because it is a 'theft' of an expression of an idea, not the theft of physical property -- basically it's plagiarism. The plagiarizer is the liable party in this case. Copyright violation is also (generally) a civil matter, rather than a criminal matter. It's (literally) an infringement upon somone else's exclusive right to copy. Basically the party that causes damages to the copyright holder is the liable party.

I hope this makes sense to you because I feel like I'm babbling. ;)

Kernel developers and FSF file lawsuit against SCO

[small_dick]

Why not? Isn't SCO's action libelous against the core developers of Linux? There appear to be several derisive comments about Linux in the SCO complaint.

The kernel developers have carefully crafted Linux over many years. It seems to me they would be willing to protect *their* image and *their* product from these kind of attacks.

Especially the people who developed the capabilities that SCO claims did not exist in Linux until IBMs intervention. The SCO action claims these developers are lying, that they never did the work!

If EFF files a lawsuit on behalf of named developers of the Linux kernel, I'd pledge $100 over paypal to such an effort.

Reason Online about SCO, Linux and Redmond

[Anonymous Coward]

Reason Online has this to say about Microsoft paying off^H^H^H^H^H^H^H^H^H^Hfunding^H^H^H^H^H^H^H licensing Unix from SCO:

2. Penguin Patrol

Someone in Redmond has a wicked sense of humor -- and timing. Just after the SCO Group got done sending letters to 1,350 users of the Linux operating system warning them that "Linux infringes on our Unix intellectual property and other rights" and that "legal liability that may arise from the Linux development process may also rest with the end user," Microsoft decided, gee, let's license some SCO Group software. In one delicious move, Microsoft has in effect hired its old antitrust nemesis, David Boies, to go to work beating Linux competitors off its Windows server installs.

The SCO Group had selected Boies to research an intellectual property case that many in the software industry felt was bogus -- nothing more than a last-ditch attempt by a failing software company to shake some cash from the trees. Sure enough, the SCO Group sued IBM in March, charging unfair competition and breach of contract.

Now, with Redmond lining up behind SCO and Boies, IBM must be just thrilled. For a tiny investment Microsoft gets to make potential corporate buyers of Linux, or IBM's operating system AIX, worry about getting dragged in court over the software running in their back office. The message is clear: Buy Windows, stay out of court.

Of course, this is why it was a horribly bad idea for Netscape and friends to run to the feds and the state attorneys general over Microsoft Internet Explorer way back in the day. Make politicians, lawyers, and judges the arbiters of good software, and you'll get software as buggy as the law.

Linux *can* effectively die if we aren't careful

[lordcorusa]

Contrary to a popular hacker belief, Linux (for purposes of this post, Linux refers to the kernel, GNU, assorted tools and environments, etc) can effectively die if we, the community, aren't careful.

In order to stay "alive" Linux must continue to provide innovative and useful tools AND support for modern Internet standards and protocols on on modern hardware. Without support for new Internet standards/protocols and new hardware, fewer people will be inclined to try Linux, and some Linux users will be inclined to switch away. This process is a negative feedback loop, which is sometimes known as a death spiral. Each generation that a negative feedback loop is allowed to continue makes it more likely that it will continue for another generation. If allowed to continue for too long, you end up with an extremely small and insular group of coders developing software that is utterly useless to anyone but themselves.

To avoid this negative feedback loop, we must maintain and expand our current user base. This is an inherently positive process, as it means that new coders will join our ranks to help make better software. However, while we continue to gain new coders every year, the overwhelming majority of people who are new Linux users are not writing new code. As cliched as it may be, these kind of people expect Linux to just work with their (often new) hardware and with modern standards, protocols, and file formats, and if it doesn't they are unlikely to return.

The obvious response is that Free Software coders will code in support for new hardware, like they have done in the past. However, a few points bear mentioning. One is that much of the device support in the kernel is the direct result of contributions (code, money, extensive documentation) by hardware manufacturers. Another is that as devices become more advanced and complicated, it becomes exponentially more difficult to write adequate drivers for them without extensive documentation provided by hardware manufacturers. In both of these cases, hardware manufacturers may be unwilling to give their assistance unless Linux has sufficient market share, or the prospects of a larger market share in the future.

In a related vein, Internet standards and protocols, as well as office file formats, are under constant threat of patents and proprietary trickery. The only way to stop or prevent this is to have a large user population which can react negatively to any such proprietary encroachment. For example, the community just barely averted disaster with W3C patent policy, and it still was not a complete victory for us. This was just the tip of the iceberg, and we may not be so lucky in the future unless we continue to grow.

So failure to maintain and expand the Linux user base threatens our ability to support new hardware and standards/protocols, and failure to support new hardware and standards/protocols threatens our maintenance of the user base. A classic feedback loop. Now that we know why expanding the user base is important, the question remains, which potential converts do we target? The choices are individual users or companies.

Targetting individual users is difficult and manpower intensive (LUGs). Most end-users do not perceive immediate benefits of switching, and will thus resist switching. And even when they do switch, there is little benefit to the catalyst of the switch, other than a vague sense of accomplishment. Not that this isn't worth doing, but it's not the optimal strategy.

On the other hand, corporate switching has a much higher payoff. A single corporate switch can affect hundreds or thousands of users. Benefits of switching to corporations are quickly noticed (ie: cost savings, reliability, etc). And benefits to the catalyst (ie: money) are also palpable.

In conclusion we see that maintaining and growing the Linux user base is necessary for its long term viability. And we see that the best way to expand Linux's user base is to start by targetting businesses. Thus, anything that threatens that strategy threatens the long term viability and survival of Linux as a whole.