See Full Story
The other day I was asked what the odds were that SCO would win against IBM. On the basis of how the two parties were behaving, I offered a range of between 55 and 65 percent. One of the strongest indications I have that SCO might win is that several of the financial analyst groups who work with me, after reviewing the SCO submissions, have concluded that SCO might actually have the advantage. These firms are relatively unbiased and, generally, if they do have bias, it would typically favor a company like IBM with which they have substantial interest.
But if what your saying is true about having ownership of an idea, then couldn't IBM sue clone pc makers for making pc's since it was IBM's idea to begin with?
Is it just me, or is there a clear difference between the writing styles in the "pro-SCO" and "pro-IBM" camps? I've noticed every time someone writes an article that seems to support SCO and Darl McBride, they use a lot of analogies. Those who support IBM or support Linux in general are left quoting specifics and offering concrete data. As examples of the former camp ("pro-SCO") I cite:
-- This article. "Have you ever watched a divorce..."
-- McBride. "There is no Free Lunch...",
-- Howard Strauss at Princeton. "Free, 0% APR, Better Sex, No Effort Diet", Nigerian Money Scams, and yadda yadda.
In any case, it's refreshing to see such a unbalanced and biased article on the issue of SCO vs. IBM. Oops, did I say that?
Unbiased article? Give me a break... What no one seems to be thinking about is SCO trying to extort licensing fees from Linux customers. Even if SCO was right about every claim, and 25% of Linux kernel code was really copyrighted by them, they could not charge licensing fees. The rest of the code is copyrighted by the developers that released it only under the GNU license, which specifically forbits charging license fees. There is no way that SCO's 1.2 million lines of code could be used without the other 75% of the kernel. In essence they are licensing the whole work. That would be like if they found some of their code in Windows 2003 Server, so they started charging users DIRECTLY for $700 per processor. Not only that, but since Linux has a different software model and licensing structure, it would be like saying you ONLY have to pay SCO the $700, you didn't have to pay Microsoft. That is because with Linux, instead of paying a licensing fee for each processor, you agree to be bound to the terms of the GPL. There is NO LEGAL WAY they can continue to charge licensing fees for Linux. The worst they can do if for some unknown reason it is decided that they don't have to actually show any infringing content so that it can be removed is to deny EVERYONE access to Linux. Then the Linux community would have to re-design Linux from the ground up with everyone solidly attributing copyrights as they check new material in. Of course the only reasonable solution would be for SCO to show the source that is infringing so that it could be removed from the product.
Actually, they only have a legal argument against the CONTRIBUTOR of the infringing code. That's why they're taking IBM to court. In every Linux source file the copyright is clearly stated, so any files that have infringing code would have to be checked and whoever claims the copyright would have to be challenged.
How's this for an analogy, going with someone's gangster analogy.
Let's say that IBM is the big outfit in town and you are a group of small timers that hope to get somewhere. Now, you have just "merged" with another gang, and the new leadership has changed their allignment to some degree. Along with this change, a lot of your top hitmen left, leaving behind what looks to be just slightly more than a skeleton crew.
Now, you can't hope to compete with many of the others, but the leader thinks that if you show your teeth to the big outfit, this outfit might just be impassive and lazy enough to consider giving you all a job, instead of trying to "erase" you(forget all those Hollywood movies, real gangs don't survive by "offing" large numbers of people these days).
There is of course the risk that said big outfit might just decide that they should make a show of power, and you will be the test subjects. But if all ends as it should, your company will have a future, and the leadership will go down as martyrs. The only mistake you did was to alienate those who would have given you most support.
Only something as preposterous, and outright ridiculous as SCO's claims could have made Big Blue and the Open Source community end up on the same team.
Congratulations SCO, your shenanigans have made one of the biggest corporations in the world find itself allied with the "grass roots" open source movement. Maybe you'll try and have Microsoft and Linus team up sometime soon?
Thank you Mr. Enderle for providing an all too rare balanced view of this situation. Some IT people may work for outfits that may not be willing to fight this all the way to the Supreme Court. These people may have career difficulties if their CEO or CFO receives an IP violation letter from somebody's lawyer. Hard questions may follow this letter like, "What made you think that a $10,000 piece of software could be legally purchased for under a $100". "It took how many hours to install that server!", "Waddya mean ____ won't run on it."
There are plenty of good reasons for FUD right now and Mr. Enderle had done as all a favor by illustrating these facts.
'Thank you Mr. Enderle for providing an all too rare balanced view of this situation. Some IT people may work for outfits that may not be willing to fight this all the way to the Supreme Court. These people may have career difficulties if their CEO or CFO receives an IP violation letter from somebody's lawyer. Hard questions may follow this letter like, "What made you think that a $10,000 piece of software could be legally purchased for under a $100". "It took how many hours to install that server!", "Waddya mean ____ won't run on it."'
What IP violation letter? What IP?
If SCO were to send such a letter in Germany, they would be fined a quarter million euros for unfair competition since they failed to prove their case in court. If SCO has an actual case, why are they unable to prove their case in German court?
'There are plenty of good reasons for FUD right now and Mr. Enderle had done as all a favor by illustrating these facts.'
What facts are you talking about?
Until SCO made a link to this article from this main page, every single reply to Enderle criticized him for a flagrant amount of errors he made.
While I agree that there are a number of Open Source proponents out there with blinders on and over zealous in their defense of their position and attack of SCO, there is still plenty to disparage on SCO's part, even if they prove to be right.
I don't think you'd take kindly to me walkiing up to you and saying "you owe me $1000. pay up." "Why?," you say. I say, "the company that built your car is using a piece that I designed and they are infringing my patent. So, you owe me $1000- pay up- now." You say, "how do I know what you are telling me is the truth?" I say,, "because I am telling you it is. I have the designs at home but I can't show them to you now and I am suing the car company, but you still owe me- so pay up now!"
At the very least you are going to think I am a jerk for coming up and telling you that you owe me money in a fairly terse and matter-of-fact way. I haven't proven to you or anyone else that I am owed, and I refuse to prove it while still demanding money. Beyond that, even if we were once friends, you'll tell me to take a flying leap until I can prove to you that you owe me something and your dislike of me is probably only going to grow with the continued unfounded demands.
In the end, you'd probably say "hey, chump, if and when you win your lawsuit and you have some proof I owe you something- then come talk to me. Until then, take a hike."
Beyond that, I don't believe the OS community is talking about changing a few words and variable names to get around the copyright issue. You can patent/copyright a process or method, but not necessarily a concept. You can patent a way to open a file on a computer, but not the act of opening a file. That would be like saying you could patent shoes. Not an individual type or kind, but patent footwear as a whole. This, you can not do. I believe the OS community contains more than enough intelligence to find other ways to code up other processes that achive the same end. If they should need to.
I don't think SCO has a leg to stand on. They have a lawyer who is into increasing his profile with cases against MS, helping Gore, and now this. IBM has more lawyers than SCO has employees and most are experts in IP rights and patent law. By the new ad campaign they are running featuring linux, it really doesn't look like they are sweating a bit, and they have 3 billion on the line!
1. I think your estimate of 55-65% chance of an SCO victory are a bit high, but anything can happen in our legal system.
2. The OS community could, legally, code up functionally equivalent programs to replace any code that actually does belong to SCO.
3. SCO has no legal basis with which to demand payment from anyone and should shut their yaps until they do.
Since http://www.sco.com has linked this story from their main page, somebody coincidentally keeps coming in and posting a short unsubstantiated reply praising Enderle for his lack of bias. This I find rather interesting since every reply before that was quite critical of Enderle for omitting certain facts, like the fact the SCO has already lost a court case in Germany over the allegations it's still making of the Linux community. In fact, SCO was fined as a result:
Also, Rob Enderle's main contention that "Line-by-line replacement won't work in the SCO-vs.-IBM scenario" is also wrong, since mitigation is almost always how IP disputes are handled.
Enderle also neglected to mention this letter:
Which gives IBM the right to use code they have created to use any way they see fit - this is in direct contrast to SCO's claims that RCU, NUMA, and JFS cannot be contributed to Linux because they were also in AIX. All this code was created and copyrighted by IBM. It is this code that SCO is claiming was improperly added to Linux, and the really amusing bit is that the code that was put into Linux was a clean room re-implementation of the code that previously existed in AIX. They are not the same code.
Enderle also didn't mention that SCO's "examples" of infringement were demolished the day they became available to be reviewed:
Rob Enderle has also neglected to mention that SCO's credibility is rather low. SCO, for example, stated they were going to audit AIX customers - something that not only did they fail to do, but something they have no legal right to do.
Enderle has also neglected to mention that SCO is under investigation relating to irregularities of their IPO. You can find more details here:
And SCO may be involved in some questionable practices again, regarding their so-called "purchase" of Vultus. Read about it here:
It surprising what Enderle has missed. From my perspective, it seems that SCO is being run by a gang of unprincipled criminals. I'm confident that the truth will eventually make it to the mainstream media, now that IBM has sent a subpoena to Canopy:
You can find the text version here:
If you're not aware: Canopy is the majority shareholder in SCO, and also, was the VC behind Vultus, which SCO "bought" which allowed Canopy to make back their initial investment in Vultus - a company that was unprofitable, and worthless, and also coincidentally located in the same building SCO is.
This is by far the best unbiased analysis I have seen of the situation so far.
Further more, some of the vitriolic responses to this article, only go to underline the highlighted flawed attitude of many open source advocates on this issue.
"It's about time someone has given a real and logical overview of the SCO Wars. I have had many people say to me that they do not want anything to do with SCO or it's products because 'what it is doing to the Linux world'? What is SCO doing? They are trying to get revenue from a product that looks like, according to the contract between AT&T and IBM in the 80's, their code."
How do you come to that conclusion? Maybe you aren't aware of a few facts:
The code that has become available to the public so far has been proven NOT to be SCO's code.
The first example was created by Dennis Ritchie and Brian Kernighan back in the early 1970's. They included the code in a book, "The C Programming Language", free to be used without restriction. This is the most famous book on programming in C, and EVERY programmer worth anything knows C. Look on pages 180-190 for it. You can buy the book here:
K&R invented C. Dennis Ritchie was one of the main people that invented Unix. Both K&R claim that the code is public domain.
The second example was the BPF code, and was not SCO's code at all. In fact, it was *stolen* from Berkeley, and to make matters worse, the Linux version is a clean room redesign of it.
You can find a more detailed analysis here:
Most of the code that SCO claims is theirs was written and created by IBM. IBM holds the copyrights to them and the patents. This code clearly does not belong to SCO - and in fact, isn't even in SCO Unix. SCO is trying to lay claim to this code, but they (and Rob) ignore this letter:
[quote]2. Regarding Section 2.01, we agree that modifications and derivative works prepared by or for you are owned by you. However, ownership of any portion or portions of SOFTWARE PRODUCTS included in any such modification or derivitave work remains with us.[end quote]
"I'm sick of hearing about 'what SCO is doing', get a life and look at the facts."
Refute any of the facts I posted above. So far, SCO hasn't shown there has been any infringement at all. Perhaps you aren't aware that if it wasn't for Linux, SCO wouldn't even EXIST as a company. They did their IPO when they were named Caldera and back then they were purely a Linux distributor. Caldera later bought, with the money they made from the IPO, the old AT&T codebase and early this year, renamed themselves to "SCO". Without this initial IPO money that Caldera (now SCO) got directly because of Linux, SCO wouldn't even exist and neither would this lawsuit.
SCO is lying about the Linux community, and has stabbed them in the back. It was the Linux community that made SCO's very existence possible. With friends like these, who needs enemies? To call the current management at SCO "scum" is an insult to people who are merely scum. These are vermin. They are trying to live off from other people's work - not their own. They are parasites.
"It is not for us to complain but to wait for the court out come. Thanks Rob for your frank assessment on the SCO Wars and if anyone else has the time, read the contract and see for yourself."
Rob neglected to mention that SCO has already lost a court case in Germany against LinuxTag. LinuxTag sued SCO for basically false advertising, and SCO settled immediately. If SCO has such a strong case - why was SCO unable to demonstrate it in Germany? Why has SCO still failed to demonstrate a SINGLE example of infringement? SCO claims there are a MILLION LINES of Linux copied from them. That is 1/3rd of the kernel. Is it reasonable to believe that nobody noticed such a collasal theft until now? SCO *was* a Linux distributor and a contributor - they worked in the codebase.
Rob also doesn't seem to realize that IBM, HP, Sun, and SGI also have full access to the original AT&T Unix codebase. Any one of these companies could have easily purchased SCO back in January when all this started. They has the full code base to compare, and they know for CERTAIN if there is infringement or not. Considering that SCO is suing for a BILLION DOLLARS, if the case is so solid, why hasn't anybody purchased SCO? Sun is mortal enemies with IBM, and HP isn't too friendly, SGI needs money - yet none of them purchased SCO when all of them have the ability to do that. Why is that?
I wouldn't say Rob has a very "frank assessment" of the situation. He's not even bothered to mention that the Linux community has from day one offered to mitigate and SCO has prevented it. This will seriously hamper SCO's ability to collect damages even if IBM is found to be guilty which is unlikely in the extreme. Rob also hasn't mentioned that SCO has a track record for deliberate and pernicious lies - for example, SCO claimed they would be auditing AIX users after June 13th - something that never happened, because SCO has no legal right to do this. Rob also didn't mention a very shady deal in which SCO purchased a failing and dying company called Vultus. Vultus is another Canopy company that was not public, which wasn't profitable, and also happened to be located in the same building as SCO was. SCO appears to have bought Vultus to transfer monetary assets out of SCO and into Canopy - the owners of Vultus, and coincidentally, the largest shareholder of SCO.
Also, Rob forgot to mention that SCO is currently under investigation for securities fraud with regard to their IPO.
How's that for a frank assessment?
It's about time someone has given a real and logical overview of the SCO Wars. I have had many people say to me that they do not want anything to do with SCO or it's products because 'what it is doing to the Linux world'? What is SCO doing? They are trying to get revenue from a product that looks like, according to the contract between AT&T and IBM in the 80's, their code.
I'm sick of hearing about 'what SCO is doing', get a life and look at the facts. It is not for us to complain but to wait for the court out come. Thanks Rob for your frank assessment on the SCO Wars and if anyone else has the time, read the contract and see for yourself.
Rob, You miss the point. There are pragmatic open source types like myself who believe that individuals and companies should have the right to protect their intellectual property. Thus, if SCO has a legitimate claim then they have recourse through the court of law. I am not going to pre-judge the ruling or the law.
What is getting people upset is SCO's behavior, not their right to protect their IP. They haven't proved their case, but they are sending invoices and threatening third parties with lawsuits.
Here's the analogy: Publisher A publishes a book by Author B. I (party C) buy the book from a bookstore D. Another author comes along (AA) claiming that he actually wrote the book and B stole his work. Now AA instead of just suing A or going after B, AA starts to claim that I have to pay them even though they haven't yet established their claim. Similarly, they are sending invoices to the bookstores asking them to pay up or risk a lawsuit.
This is not about intellectual property rights, it's about thuggery. SCO is acting like a bunch of thugs.
Mr Rob Enderle,
Your article shows clear bias and is full of unsubstansiated "facts".
From your article:
"However, a few months ago, this changed because I started to get disturbing e-mail from people I had previously respected."
Well why didn't you give an example? Tell us who sent you these disturbing e-mails? If you are telling the truth then you have nothing to fear. Or did you just make that up?
"What seemed to be different was that the open-source community seemed collectively to say, "Theft is okay as long as we benefit.""
This is the worst one of all. Provide some evidence backing up your unfounded assertions. Who exactly has said "theft is ok"? Or did you just make that up?
"when there is a doubt about ownership, you have the right to prove that ownership"
Well that seems fair. How about you rephrase it to "when there is doubt about ownership, we have the right to make unsubstanstiated allegations and make it impossible for the offending person(s) to resolve the issue as quickly as possible"
There are many articles pointing out the fact the SCO has made allegations, and at the same time, made it impossible for Linux developers to remove the offending code. One link: http://www.eweek.com/print_article/0,3668,a=49608,00.asp
Apparently you want to say that SCO might have a case.
Personally I doubt it.
So let the court decide about that.
But then you start expressing your oppinion about the legal system like this:
"The court system is an artificial world"
"So far the ones that come down against SCO appear to be paid by the other side,
I can not be sure about the ones for SCO but clearly some of them are paid by SCO"
(Artificial, paid by.)
Next you have an oppinion about the German court system.
"I am not clear why you would want to put this on trial in Germany anyway, folks I know avoid that court system like the plague"
"Folks you know avoid that court system."
I wonder why.
And still your whole story starts like this:
"I started to get disturbing e-mail from people I had previously respected"
Then you start FUDing individual people like this:
"Only one comment on this, they don't have to prove anything to the Linux
If you are right, they will lose, if you are wrong and haven't prepared yourself
or your company, then good luck in your next job
(assuming the blame falls on you or your department). Why take the risk?
It shouldn't be your problem to begin with.
"Why make this a personal risk or one that puts your department or company at risk
if you don?t have to? This isn?t a great time to be unemployed and that first article
you mention was written largely because of my concern that too many IT folks were putting
their jobs and the jobs of their people at risk needlessly because they weren?t doing
a proper job of justifying their decisions."
Very uggly, very, very indeed.
No Rob, you are not convincing at all, nor is SCO.
However, I would not comment yor story at all if you had not written this:
"the open-source community seemed collectively to say, "Theft is okay as long as we benefit."
That is rubbish. Open-sorce people do not think like that.
As far as your knowledge is concerned, yes, as you wrote:
"I personally had little interest in Linux"
As far as your friends and experts are concerned you said:
"I started to get disturbing e-mail from people I had previously respected"
Perhaps you are just bluffing.
I'm from Germany, so maybe the difference within the legal systems between the US and Germany lead me to wrong conclusions. But I would think that a case must be proven in US courts, too. And I don't mean any ominous arguments which maybe impress financial experts, I mean technical facts. And at the moment there isn't any such fact that supports SCO, at least not publicly available. A court here in Germany sentences SCO to pay a penalty for their claims that Linux infringes SCO's IP without a prove. SCO is not allowed to make such claims here anymore without any prove. Is the legal system in the US really so much different ?
have i missed something here have SCO actually proved some of its alligations ? And i cant realy
understand how SCO can start sending out invoices
for a licens fee before anything is settled in court. If they loose, could'nt that be seen as some sort of fraud then ?
The ideas and concepts you're referring to are called patents. SCO has none of those (at least not relevant ones), IBM has many. So, how exactly is SCO going to use that against IBM? Software licence by AT&T is NOT about patents in any shape or form.
Open source community is doing SCO a big favour. They are trying to identify the (alleged) infringing code so that SCO finally have ammunition to go after the copyright infringers. But no, SCO are not doing any of that. Thousands of CDs with Linux are pressed every day, the software is downloaded freely and copyright holder (SCO) is standing by doing nothing. How will they explain this one to the judge?
They won't. They don't want to see the inside of the courtroom by any stretch of imagination, simply because they have no case. None.
Their evidence has been shown to be publicly available code, either without copyright or under a BSD licence, some even developed independently by Linux developers. The code is not secret by any standard (heck, even I have a copy), so their trade secret claims go out the window. JFS and RCU are independent and patented works of IBM (unless SCO now claims they own OS/2 and IBM patents as well). Similar applies to NUMA and other stuff they mentioned in their lawsuit.
Out of all this, you conclude that SCO has some 55% - 60% chance of winning. I think you must have used the same team of non-existent rocket scientists from MIT, that SCO used to find the infringing code, to calculate this number. No?
I hardly know where to begin criticizing your Linux comments. You seem like a nice guy, but you clearly haven't done your research. One thing is certain, the "financial analysts" you claim have helped you understand the case didn't do their homework either. So let's go back to basics.
As you know, SCO was once Caldera Systems International. Caldera was one of the original Linux distributors, but they lost an enormous amount of market share due to bad PR, a so-so distro, and poor marketing. However, before they stopped distributing Linux, Caldera's programmers wrote a great deal of code for the Linux kernel, as you can see here:
Much more importantly, given SCO/Caldera's contention that IBM stole SMP (the ability of a kernel to run on multiple processors) from SCO, is the fact that Caldera bought an important kernel programmer named Alan Cox a dual pentium motherboard so he could SMP to the Linux kernel. SMP was available in Linux kernel 2.0 - though I don't know how reliable it was -long before the existance of the 2.4 kernel SCO is complaining about.
The issue was discussed on line several times prior to the SCO lawsuit. You'll note that the last one is an e-mail from a Caldera employee:
So how can SCO sue anyone for using SMP in Linux when they financed the effort to create Linux SMP?
This is just the tip of the iceberg. Here's a URL where you can read e-mail from a SCO developer discussing code on a Linux mailing list. Note his e-mail address:
Note also that he takes his promises seriously:
Do you think he's contributing without SCO's permission?
Here's another discussion of a SCO employee contributing to the Linux kernel. The employee's e-mail address is someplace in the middle, where credit is being given for the four contributions he made to the modern kernel:
In other words, as Caldera, SCO was initimately involved with creating the Linux kernal. They bought Linux programmers hardware, allowed their programmers to make contributions, and still brag about their contributions to Linux on their website. How in the world are they going to win a lawsuit claiming that copying the Old-SCO kernel code (which is reportedly vastly inferior to the Linux code) harmed them? How can they claim that they have what the lawyers call "clean hands?"
It gets even worse for SCO. After Caldera bought the SCO code, they announced a plan to unify Linux and SCO Unix. (Stuff like this has been vanishing from SCO's web page very quickly.) Read it and tell me if you still thing SCO has a chance in court!
Also note this ZDnet article. Ex-SCO president Ransom Love talks at length about unifying Unix and Linux:
So now we know that the owner of the SCO code had made a corporate decision to unify the two breeds of UNIX and put the end result under the General Public License. When Caldera got new management in the form of Darl McBride, they changed their name to SCO Group, stopped selling Linux, and announced their new IP licensing program. How in the world can they go into court and claim anything was stolen from them when releasing SCO's code to the public was once corporate policy?
This brings us to the Trillian project. Before buying SCO and changing its name, Caldera took part in this project along with IBM, Redhat, CERN, Intel, Hewlett-Packard, SGI, SuSE, TurboLinux and VA Linux Systems. This project was an effort to make sure Linux could run on Intel's Itanium Processor. Among other things, the Trillian Project made sure that Linux would run SMP on the Itanium chip. SMP, as you'll recall, is one of the things that SCO/Caldera claims was stolen by IBM during the course of Project Monterey. However, if you'll look at:
which is a FAQ put out by the Trillian Project, you'll see that IBM didn't provide the project with SMP. The SMP was provided by Intel and VA Linux, (probably a re-write of the Alan Cox kernel code I discussed above.) IBM provided tools for testing and tuning, but did not provide any "enterprise" enhancements. Notice this last bit! IBM was involved in two UNIX efforts and didn't mix them. With Project Monterey they coded. With Trillian they provided testing tools.
Caldera was a late joiner. They were to provide a distribution, but little or no technical help. Caldera was considered a "premier" Linux company at the time, but the best they could do was join late and create a distribution? Meanwhile, RedHat provided an entire software toolkit!
Now this gets interesting - or does it? IBM was clearly hedging their bets. They were participating with SCO in project Monterey, and with Caldera in developing Linux for Itanium.
In August of 2000, Caldera bought SCO. According to LWN, SCO, in October of 1999, (nearly a year prior to being bought by Caldera,) "...entered into strategic agreement with TurboLinux to develop services for TurboLinux's TurboCluster Server and provide Linux Professional Services for TurboLinux customers. Old SCO also made a sizable investment in TurboLinux, Caldera and LinuxMall." (Old SCO, in this case, refers to SCO before it was bought by Caldera) You'll find that story here:
So what's a judge or jury to do with all this? Look at all the devasting questions a lawyer can ask:
How can old SCO develop services for Linux without knowing what's in the kernel? If there was infringement, why didn't they sue back then?
How does SCO/Caldera get off suing IBM for giving Linux SMP when they knew that the sixty-four bit version of SMP, which is what they're claiming IBM lifted from Project Monterey, was in fact provided by Intel and VA Linux?
How can they make any claim on Linux when Caldera provided Alan Cox the motherboard he used to implement the first version of Linux SMP?
How can they claim they didn't understand the GPL when they contributed code to the GPLed Linux kernel and planned to release SCO code under the GPL?
Perhaps worst of all, Caldera was a major Linux company, taking part in a major undertaking along with major players. If code was being stolen from SCO, wouldn't Caldera have been an accomplice to that illegal act, and doesn't that render their hands unclean?
Lastly, notice what the Doug Michels, president of old SCO, had to say about the Trillian Project in August of 1999:
"IBM has joined so many things over the years, itís not funny. I donít think itís put a significant amount of money into Trillian and itís a small amount of money to be part of a trend. Itís important to look at internal sponsorship in IBM. Monterey is supported by Bob Stevenson, whoís head of IBMís server group and Lou Gerstner [IBMís chief executive] has also bought it off as a strategic initiative."
"IBM is not looking at Trillian as an alternative to Monterey. The real interest in Linux is coming from all the software companies that sell databases and transaction based tools because they are frustrated that Microsoft more or less gives these things away as part of the Back Office bundle. So they say Ďif you give us a free OS, weíll make money from it.'"
"Linux may be there sooner than 10 years, but I donít see any efforts there yet. Trillian is not that type of project and I donít see them adding enterprise class services or performance management capabilities. No one is doing it. SGI has made some noises and people are talking about it, but itís not coming this week, next year or before Merced ships."
SCO's "partner" IBM was working on a major Linux initiative and SCO's management didn't care. Talk about a lack of due diligence! Also note the language Michels uses. Sound familiar? You'll find the story here:
And just so you'll have something to sink your teeth into, here are several URLS discussing the Caldera's participation in the Trillian Project.
http://radio.weblogs.com/0120124/2003/07/08.html (about 1/4 of th way down)
(Note the web-link to the Caldera IA-64 distribution at the last site on the list, which has been taken down on the Caldera side. They've removed an enormous amount of stuff from their website in the last couple months. This is not how honest people behave!)
However, you can use the sign-up link from Caldera/SCO for the SCO Linux Itanium processor NetFarm at:
Ultimately, going to the web page below, we find out that SCO knows about IBM and AIX and Linux compatibility because they worked on it together along with Cygnus Solutions, (Cynus is a RedHat subsidiary which specializes in implementing the gcc compiler on new chipsets) EPC, Geodesic, Merant, Parasoft and Roguewave.
Does SCO's case fall apart here? It sure looks like old SCO deliberately donated code to a Linux effort, expecting it to be part of the Linux kernel or some associated program.
To conclude, Caldera had a couple bad years, Ransom Love was thrown out and Darl McBride became president of Caldera. McBride changed the company's name to SCO and began the process of turning SCO from a company which shipped a product into a company which claims to own IP and charges others to license it. In other word, when McBride came along, the company changed tactics and stopped contributing to the Linux kernel or distributing Linux. Now they need a judge to put the worms back in the can so they can charge for the same IP they once donated to the kernel!!
But let's not forget that they don't own the IP. SCO has admitted that IBM has both copyrights and patents to RCU, NUMA, and JFS. Note that JFS is only one of several journaling file systems available to the Linux kernel. If it disappeared, very few people would even notice. NUMA is hardware. C code for connecting one program to NUMA is likely to resemble C code for connecting any other program to NUMA... I'd go on, but I'm sure you get the point.
Rob, this is what real research looks like.
I've given you link after link of facts showing that SCO helped create SMP, proving that SCO contributed to the Linux kernal, planned under previous management to put their kernel under an "open source" license, did actual work on Linux compatibility, (doubtless requiring a look at the Linux kernel,) didn't do "due diligence," knows someone other than IBM contributed the SMP code, and clearly understood the ramifications of the GPL. Any one of these would be damaging to their lawsuit. Taken as a whole, these issues are a mortal blow.
I don't know what your "financial analysts" looked at, but I doubt they looked very hard.
Did they look at the USL vs. BSD issues:
or the comments real programmers have made on the code SCO revealed:
or Eric Raymond and Rob Landley's work on SCO's lawsuit (encyclopedias in themselves:)
or the Groklaw archives, which cover all these issues and more?
Frankly Rob, your "financial analysts" were clueless incompetents who didn't do their research. (Either that, or they own stock in SCO) How 'bout next time you want to know what's up with a technical lawsuit you consult some lawyers and programmers? If nothing else, you won't sound like you get your news by reading press releases and talking to guys who own stock in the company you're covering.
Also, you might want to do something about the software on your site. It won't keep my line breaks.
SCO hasn't proved that any theft has occurred. Any code that they've shown off has either been release under a license that allows its use (such as the BSD license) or was contributed by the copyright owner (IBM, SGI, etc). Any attempt to try to collect fees w/o proving this, is essentially extortion. If they start mailing their 'invoices' to companies, they may very well find a complaint lodged against them for trying to commit mail fraud.
If you are basing the outcome on the behavior of the participants, how could you side with SCO? Their excuses change almost daily, each claim wilder than the next. Any 'proof' that has been offered has been quickly shown to be false.
Sure, they are betting the company on the lawsuit, because basically w/o the lawsuit the company is dead. Even if they were given away, SCO products offer no real benefit over the competition. As a former user of SCO products, I can safely say that they suck. Also, SCO executives couldn't have pumped their stock price to the current levels and made any money on stock sales - which is the real point of the entire exercise.
A careful analysis will point out that the AT&T license that protects against the kind of theft that may have occurred between SCO's intellectual property and linux has already seen court (at least as it pertains to the license itself), against the University of California, Berkeley, which AT&T lost. I think the point that the author may have missed is that SCO has done more than just made allegations against IBM misappropriating SCO licensed code into linux, they have done things like question the validity of the GNU Public License. The GNU public license seems to disgust SCO enough to publicly denounce it, but not to stop selling GPL'd software. Take a look at the number of GNU licensed products that are released with SCO's newest version(Samba and GCC are 2 big ones). They have a business interest to sell the same software that they condemn - thats plain hypocrisy. I'm not arguing that a DOS attack was the way to remedy this, or helps in the slightest, I'm sure it does the contrary. I will say that I am not personally concerned about SCO's current allegations because I have not seen any proof, and in America, you ARE innocent until proven guilty - so prove me guilty. Anyway - to finish my rant, If someone stole code from SCO (it wasnt me :) then do the right thing and allow the community to respond - you can't re-write Harry Potter, but as far as I know there are multiple Word Processors, web servers and even file servers all written with different code that don't steal each others intellectual property - so please, don't try to tell me that SCO owns the license to all memory management functions and that a unique one could not be written without infringing on SCO's property rights. All board games use a board but are still legal, right?
Boy, that was quite a rant :)
The supposed DDOS attack has taken place only during off hours: i.e. on the weekends and during NON business hours.
I have a better explanation as to what happened: SCO was taking their machines off the network when they were out of the office. SCO was also removing documents from the server - one example is the letter that gave permission to the Linux community to use "ancient Unix".
You might want to check the wayback machine:
To see that for some reason, any of the archives since this case have started have seemed to disappear. The newest entry is Feb 20.
To verify the stangeness of the alleged DDOS attack, you can go check the netcraft logs for yourself. It doesn't look like any DDOS attack I've ever seen and SCO doesn't have very good credibility. You can put 2 and 2 together.
I personally was able to ping the gateways and machines on the same subnet as www.sco.com and www.caldera.com without any problem during the "attack", so the gateways weren't flooded as you would expect them to be. If it was a DDOS attack, it was a very unusual one.
A few flaws:
"SCO and its legal team continue to bet the farm that they will win,"
Malarky. They continue the pump-n-dump stock scheme so the big wigs can get out of a dying company with a golden parachute.
"Were you to read the original AT&T license, which I did, you would find that it anticipated things like this and, if it is enforceable, protects against it."
Ignorance. Not even a patent can protect an idea, only a specific implementation of it. You are just plain wrong.
Here is what you present as "evidence:"
"One of the strongest indications I have that SCO might win is that several of the financial analyst groups who work with me, after reviewing the SCO submissions, have concluded that SCO might actually have the advantage."
So your buddies agree with you? Who gives a crap? Cite your sources and the reasons for *their* opinions, or this is just a handwaving appeal to authority. Or do I have to sign and NDA to find out?
"I clearly have become fascinated with the rhetoric coming out of the open-source community"
Clearly to whom? Why should it be clear to anyone?
"However, a few months ago, this changed because I started to get disturbing e-mail from people I had previously respected. When I tried to point out that SCO might actually be in the right, suddenly the attacks shifted to me in a very personal way in an apparent attempt to shut me up. At the time, I wasn't siding with SCO, I was just pointing out that the company's position might have merit"
So you formed your opinion because your feelings were hurt?
"SCO was simply saying that IBM had taken SCO's intellectual property and was giving it away to the open-source community. Given my experience with cases like this, it wasn't unusual that a big company was doing something like this. I'd seen it before."
Examples of your hand-waving red herring please.
"What seemed to be different was that the open-source community seemed collectively to say, "Theft is okay as long as we benefit.""
Cite examples please, so the morons who hold this ideology can be ostracized from the open source community. Such a parasite would be a pariah.
"The company's Web site was attacked and its executives threatened"
How does this affect the facts of the case? It doesn't. Its an attempt to hide the fact you have no facts.
"And a lot of people who should have known better seemed to think this was okay and that SCO was getting what it deserved for being on the wrong side."
Again, examples please? And who are you accusing of personal attacks, Mr. High Ground? There is simply too much malarky to refute it all. Good day sir.
I don't think the harry potter analogy is quite right. Here's a better analogy:[p]
o JK Rowling (AT&T) writes a book called 'Harry Potter and the Philosophers' stone but is barred from publishing it commercially because she has a monopoly on publishing in some other field.
o JK rowling releases the book to literati for academic study in universities, to help teach courses in writing children's fantasy (that's a metaphor for operating systems by the way). In the course of doing this, many students gain understanding of how to write children's literature. This may not be in agreement with the letter of JK's licence but she knowingly does nothing about it as it is currently of no commercial value to her.
o Eventually JK is allowed to publish 'Harry Potter' Commercially. Some legal hassles arise from the fact that she incorporated other people's writing in the commercial version without bothering to tell anyone.
o Various people, including IBM and Caldera, contribute to a generic childrens fantasy novel called 'Tux and the magic item' under the GPL.
o JK sells the rights to 'Harry Potter' to a third party. Eventually they wind up in the hands of Caldera. Caldera changes its name to SCO.
o Caldera decides to sue IBM, claiming that all childrens fantasy is in fact a trade secret of theirs, despite having published it extensively through various channels. SCO, nee Caldera tells the linux community that 'all your work are belong to us' because 'We wont tell which bits we think are really ours.' Sco issues a press release claiming that all readers of 'Tux and the magic item' now owe them $700, and the 'longer you leave it the more the price goes up. Childrens fantasy books do burn, you know ...'
o Legal scholars and various unqualified pundits spout opinions but cover their backsides by claiming (correctly) that their punditry is not legal advice.
o Various other pundits express their 2c worth by making up increasingly tortured metaphors for the case and posting them in blogs and op-ed pieces. Some of them even try to bring up freedom, mom and apple pie as valid arguments for their point of view.
o Business goes on as usual for all but the most spineless of the user community. A small handful drop their trousers and bend over for SCO.
There are many of us (linuxen) who have never said it is ok to steal. However, if someone says your car is theirs, you wouldn't turn it over without them showing some proof now would you?
This is what SCO has NOT done. The IBM countersuit was launched 28 days after IBM requested details of infringements. SCO failed to reply to IBM, instead courting the media. This does not give IBM the chance to rectify any problems. (like the car guy charging you $70 a day for car rental as he claims you've got his car, but he won't show you the evidence that would enable you to agree/disagree with him).
It turned out none of the code SCO showed in their SCOworld actually belonged too them. Apparently one bit had been
stolen by SCO from BSD and the licence ripped from the top.
Others were derivatives of BSD code and book examples. (as derivative works, they don't own copyright). SCO also refuses to stop selling Linux even though the copyright owners (kernel code contributors) claim SCO are in breach of the licence and therefore must stop using their code. Apparently much Linux code
has also been stolen by SCO for use in Linuxware, released under an in-compatible licence (er that's
In addition to all the above, some of the code SCO claims is stolen SCO employees submitted to the kernel as part of the
job SCO were employing them to do. Apparently this code was only ever used on 0.01% of systems (old alphas or something), and doesn't get used by X86s at all.
If an employee of a company in their official role sold a company car, and all the appropriate paperwork was signed and
so on, could the company later claim the person who bought the car had stolen it? I wouldn't have thought so.
You should be balanced in your view.
I imagined that SCO might have some wierd strategy and now we know. In some alternate reality SCO owns intellectual property that others developed. They even escape ip thief status by claiming anything ever run under any unix is theirs. But you blow it with you statement about ideas behind the implementation. IBM and Sequent had the ideas and owned the IP before it was implemented in Unix or Linux. They can donate or withhold the IP pretty much as they wish. SCO picked the wrong dog for a legal fight. The management at SCO will be lucky to get out of this with the clothes on their back.
So approximately 200 lines were copied from system V to linux and than there are some million lines of code that IBM put into linux although it is perhaps derived work - RCU, NUMA, JFS, SMP etc.
Ok, sco wins in court agains IBM, and parts/all of RCU, NUMA, JFS, SMP is removed from linux.
The 200 lines in linux that was indeed copied from SCO's unix is replaced in 2 days.
My computer probably uses 10 lines of code out of the 200, and it is definietly not using RCU NUMA JFS or SMP (one cpu, ext3 filesystem).
So I don't see the big problem, and also don't see why I should be paying $699 to SCO.
Unless you have a computer with several cpu's and you are using jfs (instead of ext3 or reiserfs or xfs, or anything) I don't see the reason to worry. Maybe I don't have enough empathy, or missing something ?
Aha, seems like IBM & lines copied from
system V to linux are two separate issues.
IBM is sued because of putting unix derived code
into linux, whereas users are sued because of
the copied lines.
IANAL, so do correct me if I am wrong: there was
an article linked from LinuxToday that said
afaict that users of a software are copying it too (from the harddisk/cd to ram) so for some reason that's why users of the software can be sued but readers of a book (which was illegally copied by the distributor)cannot.
But from the original IBM vs. SCO trial we will find out if RCU, NUMA, etc. is ok, and from that
trial and RedHat vs. SCO we will find out
which lines were copied from system V to linux.
And by 2006 Linux will back in business with it's full power, until Darl McBride's than current company starts blackmailing linux users again heh ;|
Your analogy to Harry Potter is flawed. We're certainly not talking about an entire volume created by a single author.
A more correct analogy:
A fifth grade class puts together a collection of poems and prints copies for parents and friends. A publisher calls and complains that the collection includes one or more of their copyrighted works. The school apologizes and offers to remove the items in question. The publisher refuses to identify which poem(s) they own because the school would remove them(?!) Then they state they now own the entire collection and everyone whoís read the book must pay them $699.00 per retina.
Oh, and the publisher turns out be a big, fat liar.
You have made me think, and in all reality when it comes to court and heaven forbid, a jury of peers (half of which are below average intelligence based on simple odds); there is actually no telling who is going to win.
However, on your last point you should consider that even copying someones work of art is legal as long as it is sufficiently changed. No new ideas have really ever emerged only sufficiently changed from a previous idea.
The example that has been "exposed" as "copying" doesn't even perform the same function (with relationship to the kernel - see http://www.catb.org/~esr/writings/smoking-fizzle.html) - I am a little iritated as to the lack of thoroughness of SCO's arguement. Show some real evidence. SCO has ample, correct? How much effort would it take to show one example that is completely and blatantly obvious?
Do they really lack the technical saviness to actually research an example fully? Especially when releasing an example?
My thought processes lead me to believe that what ever they are using to compare the respective sources is returning high numbers of "similarities". "Comparison software", however, is not intelligent enough to determine the legitimacy of copying, but humans are - and I suspect it will take quite a technical review to determine whether all those "similaries" have actually been copied.
Also, determining who copied who remains a difficulty. SCO has the edge in that "Unix" isn't open source. After this is over, though - it may be.
Your last comment is the most interesting, because regardless of the outcome this case will likely change how licenses are done and in breaking this license IBM may be opening up its own intellectual property in an uncontrolled way.
But, in general, youíve got my point. The court system is an artificial world and this works for SCO. They are the small guy against the big company, they look like the victim, down to the point where they are actually being beat up illegally (DNS attacks), Iím quite sure that should the judge even rule slightly in SCOís favor on a minor point heíll get blasted by the OSS crowd (Judges really donít like this) and there is even a chance that the OSS crowd will contact some of the jury resulting in a mistrial and possible early ruling for SCO. I wouldnít want to be either side in this, but I really wouldnít want to be IBM legal.
What judges like even less than being blasted is people who try their case in the press. Do you see IBM saying anything? Do you see tons of press releases and such out of them? No. You see a new multi-million dollar add campaign about linux and it's future as if absolutely nothing is wrong. This is the sign of a supremely confident defendant. They (at the direction of their large, highly competent, skilled, and experienced legal department) are saving their statements for the court case.
The big guy vs. little guy junk may hold in small claims or cases of a company against an individual, but this is one publicly traded corporation against another. No judge is going to care one way or the other about either one. So, while the courts may be an "artificial world", it isn't going to help SCO because it still has rules and those, at least in this case, look like they favor the bigger of the two nameless faceless companies, neither of which will engender any feeling of sympathy in anyone but for a few biased writers.
> Your last comment is the most interesting, because
> regardless of the outcome this case will likely
> change how licenses are done and in breaking this
> license IBM may be opening up its own intellectual
> property in an uncontrolled way.
What license has been broken Rob? Can you quote exactly what license and what part of the license was broken and how?
> But, in general, youíve got my point. The court
> system is an artificial world and this works for
> SCO. They are the small guy against the big
> company, they look like the victim, down to the
> point where they are actually being beat up
> illegally (DNS attacks), Iím quite sure that
> should the judge even rule slightly in SCOís
> favor on a minor point heíll get blasted by the
> OSS crowd (Judges really donít like this) and
> there is even a chance that the OSS crowd will
> contact some of the jury resulting in a mistrial
> and possible early ruling for SCO. I wouldnít
> want to be either side in this, but I really
> wouldnít want to be IBM legal.
Let me get this straight:
You're saying our judicial system is so prejudiced that they will side with the "little guy" no matter how bad his behavior has been? This might happen in a small hick town - it doesn't happen in federal court. IBM will take this to the supreme court if it needs to.
> in breaking this license IBM may be opening up
> its own intellectual property in an
> uncontrolled way.
You're assuming that IBM has violated the AT&T UNIX license. Remember that AT&T and IBM agreed to changes to the standard UNIX license which allowed IBM to retain ownership of modifications it made to UNIX, and to use those modifications, and the ideas behind them, in other, independent products. Even if the IBM-developed code that SCO is complaining that IBM added to the Linux system is considered to be a derived work of UNIX, IBM still had the right to place it in Linux.
All of what you've said here is pure unsupported speculation. You seem to think SCO will win simply because the OSS community can't be trusted to keep their mouths shut.
You completely ignore the fact that the executives of SCO can't seem to keep their mouths shut, and much of what they have said in the press will work against *them* in the court case. SCO should have simply filed their lawsuit and then kept quiet about it, like IBM has, as part of the normal procedure for pursuing a lawsuit. Instead, they've made inflamatory remarks about Linux, the GPL, OSS in general. Not to mention they are demanding licensing fees before they've won their case. All of this behaviour is going to be frowned upon by a judge.
All I see in your comments is extreme bias towards SCO and against the OSS community. How about engaging in more balanced reporting, as any good journalist should?
If you'd actually been paying attention to what the OSS community has been saying in response to SCO's claims, you'd know that all they want is a fair outcome in this case. If SCO's code really did end up in Linux, then the approapiate remedy (after any monetary compensation) is to remove the offending code from Linux. To say otherwise is ludicrous.
As you should know, copyright law protects expressions of ideas, not ideas themselves. But the nature of programming languages, and the act of writing code against published APIs and specifications, means that two programmers can write code that is eerily similiar. This is not plagerism, nor it is evidence of copying: computer code is simply less expressive than the English language.
If a judge ends up ruling that code that looks similiar even though it's independently developed is copyright infringement, then a grave injustice will have been done. However, the outcome of ATT vs. BSDI demonstrates that a judge in the past *understood* the nature of computer code, and ruled that BSD Unix was not an illegal copy of ATT Unix. This kind of case law will make it difficult for SCO to prevail in it's claims that Linux is a derivative of SYS V.
Did SCO pay you for this article? This has to be one of the worst ever articles I have read so far. Why is it that all the mainstream press seems to gloss over the fact that SCO have NOT proven that they own the code in question? The few bits that have gotten out have been de-bunked within hours..
Now if only the mainstream press would show both sides of the story....
First of all, why are financial analysts swaying your opinion on a LEGAL case? There is plenty of existing legal analysis of the case, you are ignoring *all* of it?
Your analysis is completely faulty; you mistake patent for copyright law. You cannot copyright an idea. You must be thinking of patents. But SCO's case is not about patents, except for the ones that IBM owns that SCO is infringing on. Furthermore many of the ideas that came from unix are published in the POSIX standard, as well as countless other books.
We all (yes, open source enthusiasts also) do live in a world where you have the right to prove your ownership and that no one can take that away.
But SCO hasn't proven anything yet. The Open Source community hasn't been saying "it's ok to steal". The open source community has been saying "hurry up and tell us so we can get rid of your code". The example they showed in vegas (from BSD anyway) was already written out because it wasn't good enough!
Also, what are the examples that you have seen before of a big company stealing something and then opensourcing it? There's been examples of infringing code in linux before -- but these were resolved without court fights. Why doesn't SCO mitigate its damages?
What is their compelling proof?
They got fined in Germany for not proving anything!!!!!
Because they are independent. Iíve seen a number of attorneys step into this on both sides. So far the ones that come down against SCO appear to be paid by the other side, I canít be sure about the ones for SCO but clearly some of them are paid by SCO. Attorneys can, and are really good at, arguing both sides. Iím using the financial analysts as a gauge of whether what SCO presents is compelling to a jury. My conclusion is that it is enough. Your point on Germany is interesting but neither of us likely understands all the differences there (and I do know there are huge differences in the German legal system) to effectively factor this in. It doesnít hurt your point but it may not help it much either. In looking at this, Germany just represents more risk than SCO was willing to take. (Iím not clear why you would want to put this on trial in Germany anyway, folks I know avoid that court system like the plague).
<<Because they are independent.>>
You mean like Jonathan Cohen, the 'analyst' that hyped SCO on CNBC, before the last stock surge, because the fund he manages is heavily invested in SCO? (Check out http://www.threenorth.com/sco/cohen.html for the details.) An analyst will hype almost anything if they can make a quick profit off of it. Hardly a ringing endorsement of the merits of this case.
Sorry but in what way can you base a conclusion on what financial analysts say about the case? A Financial analysts opinions are only valid when it comes to the field of finance. They are not lawyers who understand the legal background of the case. They are also not programmers who understand the evidence that SCO is presenting. You seem to ignore the more important aspects of the case. Like for instance the evidence that SCO has already shown. It hurt's their case because the code that they showed belongs to BSD and not SCO. In fact SCO is guilty of license anc copyright violation because the BSD license and copyright was removed. Also you confuse patent, copyright and contract laws. The license you read between IBM and SCO is a contract only between IBM and SCO. If it was stripped by IBM or another submiter and then put into linux then that contract does not exist between SCO and the linux developers who never agreed to the terms of the contract.
The problem the open source community has with SCO is that they appear to be using their claim against IBM to extract money from linux users because they can't show their evidence until their court date. Other than the counter suits, they don't have much downside to losing; their company was already faltering before suing IBM. Betting the farm isn't so impressive when the farm is just about bankrupt. There's little question that the users of Linux won't have to wait long for a free version once SCO's claims are revealed. I know I could personally switch to FreeBSD if that were necessary. SCO's handling of the situation, primarily their attempt to hihack Linux for their profit, will probably generate enough ill will to sink their company even if they could win against IBM. Beyond the SCO vs IBM case, there's also the IBM vs SCO suit, which is probably enough to bury them if they don't get strong revenues from Linux licenses, which I just don't see happening.
They (SCO) actually have an impressive war chest at this point and are profitable, both of which suggest something is working. If I were them would I be going down this path? Not on your life. You point out only some of the risk and right now if they lose, you are probably right, they are done. Even if Iím right and this is 60/40 betting the company on those kinds of odds would seem excessively risky. And if it were Intel rather than IBM on the other side Iíd bet differently because Intel uses litigation like an offensive weapon, IBM has been too risk adverse for too long and I blame their legal department for much of this current problem.
> They (SCO) actually have an impressive war
> chest at this point
$10m in cash, according to their most recent quarterly report, which is up from the $6.6m in their 2002 annual report, but down from the $22.4m from their 2001 annual report. That doesn't strike me as terribly impressive.
> and are profitable
They have had a single profitable quarter, with a $4.5m profit, in their 6 year history.
> both of which suggest something is working.
"Something" is SCOSource, which accounted for over $8m in new revenue in the latest quarter. That suggests that if SCO's attempt to control what IBM can do with IBM's intellectual property fails, that the only thing which has enabled SCO to show a profit will evaporate.
> They (SCO) actually have an impressive war chest
> at this point and are profitable, both of which
> suggest something is working.
SCO eeked out a 3 million dollar profit, not from sales of any of their products, but from the money they got from Sun and Microsoft. Their sales from their traditional Unix business declined from 16 million to 13 million in Q3. They have one single patent.
You consider that a "warchest"?
IBM makes in profit several billion per quarter, and has over 30,000 patents - of which I'm sure every Canopy company violates at least one. In David and Goliath fights, David usually loses.
> If I were them would I be going down this path?
> Not on your life. You point out only some of the
> risk and right now if they lose, you are
> probably right, they are done. Even if Iím right
> and this is 60/40 betting the company on those
> kinds of odds would seem excessively risky.
What most of us are trying to figure out is why do you think you're right? You didn't show any supporting evidence. You ignored that SCO has already lost a similiar case in Germany. You neglected to mention that all the evidence they have shown publicly was completely debunked. You forget to mention that if indeed SCO is telling the truth and there are MILLIONS of lines of infringment that's 1/3rd of Linux, and just NOW SCO notices it? To trust SCO in this is naive to an extreme.
> And if it were Intel rather than IBM on the
> other side Iíd bet differently because Intel
> uses litigation like an offensive weapon, IBM
> has been too risk adverse for too long and I
> blame their legal department for much of this
> current problem.
What if SCO was trying to get Microsoft end users to pay $699 per CPU for an undisclosed infringement? What then? You seem to think Linux is something different - it's not. SCO isn't following traditional IP litigation routes. They have to mitigate or they lose all rights to the IP. It's rule #1 in IP litigation.
The Linux developers want to see this go to court. I'd hate to see SCO improperly profit from their extortion.
You claim to have read the original contract between AT&T and IBM, but you seem to have conveniently left out a portion of the contract that doesn't support your case. I'm referring to the side letter to the original contract which clearly states that IBM owns all derivative works that it developes. Aside from the fact that it is hotly disputed whether any of the disputed technologies constitutes a derivative work of unix, even if they did, IBM retains the ownership of said work and the right to do what they please with it.
That's because the side letter that I read only released IBM from "concepts" not derivative works, even methods remained enforced.
> That's because the side letter that I read only
> released IBM from "concepts" not derivative works,
> even methods remained enforced.
3 posters have proven that the above statement of yours was wrong. Tell us how you came to that conclusion, or was it a pernicious lie?
> That's because the side letter that I read
> only released IBM from "concepts" not
> derivative works, even methods remained
"2. Regarding Section 2.01, we agree that modifications and derivative works prepared by or for you are owned by you. However, ownership of any portion or portions of SOFTWARE PRODUCTS included in any such modification or derivitave work remains with us.
9. [...] Nothing in this agreement shall prevent LICENSEE from developing or marketing products or services employing ideas, concepts, know-how or techniques relating to data processing embodied in SOFTWARE PRODUCTS subject to this Agreement, provided that LICENSEE shall not copy any code from such SOFTWARE PRODUCTS into any such product or in connection with any such service [...]"
So IBM retains ownership of derivative works. If those derivative works did not contain any UNIX code (not developed by IBM), IBM would be free to use them elsewhere. Also, IBM is free to use UNIX techniques (methods) as well.
And I quote from Exhibit C on SCO's site:
"we agree that modifications and derivative works prepared by or for you are owned by you". Gee that sounds like it applies to the derivative works, not "concepts" as you have stated.
The side letter I read is Exhibit C in SCO's suit, available at
This side letter says:
"2. Regarding Section 2.01, we agree that modifications and derivative works prepared by or for you are owned by you. However, ownership of any portion or portions of SOFTWARE PRODUCTS included in any such modification or derivative work remains with us."
It goes on to say:
"9. ... Nothing in this agreement shall prevent LICENSEE from developing or marketing products or services employing ideas, concepts, know-how or techniques relating to data processing embodied in SOFTWARE PRODUCTS subject to this Agreement, provided that LICENSEE shall not copy any code from such SOFTWARE PRODUCTS into any such product or in connection with any such service and employees of LICENSEE shall not refer to the physical documents and materials comprising SOFTWARE PRODUCTS subject to this Agreement when they are developing any such products or service or providing any such service."
I'm interested in your opinion about what these clauses mean for IBM's contribution of JFS and NUMA to Linux kernel 2.4.
My opinion is that Exhibit C makes it clear that IBM's derivative works are IBM's property, and that IBM can distribute JFS and NUMA anything else they develop on IBM's choice of license, so long as IBM does not distribute any original SCO UNIX code as part of IBM's products.
But the IBM/ATT contract cannot be applied to a third party. It only affects IBM.
Even if IBM is held to have improperly disclosed trade secrets, this does not affect Linux, since Linus and most linux users have done nothing improper.
Note that SCO has NOT alleged copyright infringement by IBM in their suit against IBM.
>>The other day I was asked what the odds were that SCO would win against IBM. On the basis of how the two parties were behaving, I offered a range of between 55 and 65 percent. I've spent a lot of my life watching litigation, and I believe that you can better tell the outcome by trying to determine what the parties believe and how believable they are than if you just try to dig through their rhetoric.
Please be more specific about how you think the parties are behaving. On the basis of how the parties are behaving, I come to the exact opposite conclusion. That paragraph is really just about the same as not writing anything.
>>The pleadings generally amount to incredibly detailed "they did this, they did that" kinds of arguments, and it becomes difficult to call a winner until you actually see what shows up in court and how the judge and jury react.
OK. On the other hand, if we won't know the winner until the judge says so, maybe SCO should hold off on the statements like "Linux absolutely has our IP" until then.
>>SCO and its legal team continue to bet the farm that they will win, and IBM appears to be trying to limit its risk. Also, IBM doesn't have complete control of its own side, especially with a massive amount of well-wishers who might actually be making it more difficult for IBM to win. For example, who do you think really benefits from a denial-of-service attack on SCO?
SCO didn't have that big a farm to bet. They claim to have been a few quarters away from shutting their doors earlier this year. About who benefits from the denial of service, do you think SCO will ask the judge to declare they win because they suffered a denial of service attack? I don't think SCO will ever provide proof or evidence that denial of services brought their website down for more than a week every time the business day ended.
>>One of the strongest indications I have that SCO might win is that several of the financial analyst groups who work with me, after reviewing the SCO submissions, have concluded that SCO might actually have the advantage. These firms are relatively unbiased and, generally, if they do have bias, it would typically favor a company like IBM with which they have substantial interest as opposed to a company like SCO with which they don't.
Unnamed financial analyst groups. Why not present the evidence that they used to become convinced? Of course legal analyst groups may be in a better position.
>>I clearly have become fascinated with the rhetoric coming out of the open-source community. It appears to me that a lot of people believe the U.S. legal system is based on what people outside of that system believe -- or that simply because something should be the case is enough to ensure that it will be. Unfortunately, life and litigation generally don't work that way.
Why not provide a quotation? Since you didn't I'm not sure I should believe your assertion about the beliefs of the open-source community.
>>Have you ever watched a divorce either between marriage partners or between business partners? From the outside, it can look like the participants were suddenly possessed by evil spirits or aliens. This is because during the early phases of a relationship, when people are working together, they tend to ignore the other person's shortcomings and talk about how wonderful they are.
>>Once the relationship fails, however, the other person is often presented as if they were Satan spawn with no redeemable qualities at all. So it has seemed with SCO. The company started off as friendly underdog Caldera going after Microsoft. Now the company is nasty little SCO going after Linux . As far as I can tell, it really is the same company.
Well, different management for one. But nastiness or underdog-ness is not evidence. Where is the evidence?
>>When the litigation first started, I personally had little interest in Linux and open source , even though I had covered it on and off for well over a decade. While it was clear the movement had engaged others, it hadn't yet engaged me. However, a few months ago, this changed because I started to get disturbing e-mail from people I had previously respected. When I tried to point out that SCO might actually be in the right, suddenly the attacks shifted to me in a very personal way in an apparent attempt to shut me up.
A little on the paranoid side? People who speak in public have whackos who don't like what they say. That's true of you, Linus Torvalds, Dee Ann Leblanc and others. I wouldn't characterize them as trying to shut you up. Anyone who believes he or she can "shut someone up" with an email is too wacko to be believed. In general, if you are afraid of wackos, get out of the publishing business.
>>At the time, I wasn't siding with SCO, I was just pointing out that the company's position might have merit, and that I knew some of the folks and they weren't as evil as people seemed to think. I'm an analyst; this was analysis.
>>SCO was simply saying that IBM had taken SCO's intellectual property and was giving it away to the open-source community. Given my experience with cases like this, it wasn't unusual that a big company was doing something like this. I'd seen it before.
Really, then you could have provided an example and chose not to? Was your example a closed source company like Microsoft inserting code in secret and hoping it would never be exposed? Is the reason you chose not to include your example that it is impossible to add code to Linux in secret hoping that it will never be exposed, so there is no valid analogy? I'm going to guess yes, that's the reason you didn't give us an example.
>>What seemed to be different was that the open-source community seemed collectively to say, "Theft is okay as long as we benefit." SCO was suddenly painted as evil, referred to in terms that implied it was incompetent, greedy and came from questionable parentage. The company's Web site was attacked and its executives threatened. And a lot of people who should have known better seemed to think this was okay and that SCO was getting what it deserved for being on the wrong side.
You need a quotation to support an assertion that the open source community said "Theft is okay." Linus Torvalds has been quoted saying the opposite. Linus could be considered a member of the open source community.
>>Ever watch people who have been caught stealing something defend themselves? They say things like, "I found it," "Someone else gave it to me," "It wasn't yours to begin with" and, "I have as much right to it as you do." In the open-source rhetoric that followed the SCO lawsuit announcement, I saw a tremendous amount of similarity to this behavior.
Ever watch someone lying to run a scam? They say things like I have absolute proof that I can't show you. Specifically SCO has said many times that the code in question was not BSD code. They show the code then, lo and behold, BSD code. Not "obfuscated" but a clean room implementation. Why hasn't SCO shown more evidence that can't be debunked? Seriously, if they have it, they would be in a much better position if they show it. They don't show it because either they are incompetent or it doesn't exist. OK, maybe both.
>>So, what if the SCO folks are not evil incarnate and their stuff was stolen? Wouldn't that make them the victim, and is it really okay to attack the victim if your side benefits? Where do you draw the line between good and evil?
Whew, trying to pull a fast one there huh? Evil incarnate and stuff stolen are two separate ideas. One can be true and not the other. You show a quotation and I'll address the evil incarnate stuff. I have not seen one person accuse SCO of being evil incarnate. What if their stuff was stolen? If it was, they are the ones with the proof. IBM will see whatever evidence they have in discovery long before it gets to a judge. Who are they hiding their proof from? They claim to have millions of lines of code. They could show all of it without giving IBM more information than it will have before the court date. Why not show 1% of it? Once again SCO would be in a stronger position if it shows the code. The fact that they refuse proves they are incompetent, they don't have the evidence they say they have or both.
>>One belief I found to be particularly interesting was that you could get out from under this problem by simply rewriting the sections of a software product that were in violation. Some people evidently think that if you were caught with a line-by-line copy of someone else's software product, all you would have to do is rewrite the offending lines and you could continue to sell the result.
>>To extend the example to the book-publishing world, some open-source proponents have argued that if you started with Harry Potter and the Sorcerer's Stone, you could end up with Parry Hotter and the Wizard's Rock and be just fine. But this sort of line-by-line replacement won't work because the ideas that surround the product are also protected.
How about rewriting code from scratch that provides the same functionality? That's what the Open Source community has actually proposed. In case you didn't know, SCO owns no patents relevant to the Linux kernel, where they most recently claim the violations are. Producing code independently that provides the same functionality does not violate copyrights. You won't believe me so ask your "financial analyst" friends. Assuming they exist.
>>Were you to read the original AT&T license, which I did, you would find that it anticipated things like this and, if it is enforceable, protects against it.
If you were to read the original AT&T contract, along with its side letters, which I did, you would find that it anticipated things like IBM adding features to its OS that didn't include any original System V code. In fact the original side letters *specifically* allow IBM to do whatever it chooses with such code. Will SCO prove the opposite in court? Good luck. Until it does, there is no SCO IP in the form of derived works in Linux.
>>Let's pretend for a moment that we live in a country where you have the right to protect what is yours, regardless of whether you built it or, like SCO, bought it.
>>Let's also pretend that, when there is a doubt about ownership, you have the right to prove that ownership and that no group of vigilantes or large companies has the right to force you to give up what you can prove is yours, or take away your right to try. Let's pretend that people in general in this mystical land of the free have the right to have opinions different from yours without fear of personal physical or verbal attack.
Do you believe the Open Source community is stopping SCO from proving its ownership of the code? How could it do that? SCO has made a decision on its own to refuse to prove its ownership. None of its excuses are valid. The reason it can't prove it is most likely that its claims are not true.
>>I actually think I live in a place like this, so it would be nice if more open-source software folk joined me here.
Your closing was worthy of violins. I'll close by saying if SCO wants to wait for the court of law to decide if it is right or wrong, that is fine. They should have evidence to show the court. If SCO wants the "court of public opinion" to rule in its favor, they have to show their evidence in public. Sorry. You want the public to believe you? You have to show evidence to the public. From their refusal to make their evidence public, we can only conclude that they don't have any.
It's interesting to read another view of the SCO fuss. But you seem to be ignoring most of the facts, especially:
1. The code that SCO showed as "evidence of copying" turned out to be from BSD. AT&T unsuccessfully sued over this code years ago.
2. SCO distributed Linux itself under a GPL license. This license allows everyone to whom they distributed it to re-distribute it without fee. (read the GPL).
3. The "line-by-line" issue: you make it sound as though the whole of Linux were basically copied from somewhere, with minor changes. You ought to know that this simply is not true. Not even SCO is claiming this.
You are making the classic outsider's mistake of assuming that the IBM case ties in with Linux. Simply, it does not. The SCO IBM case is a simple contract dispute with potentially little or no impact on Linux.
In this dispute SCO is claiming that certain derivative works given by IBM to the Linux Community such as NUMA, JFS, RCU etc. have been disclosed publicly against contractual obligations held by IBM. There is *no* dispute on who owns these works - it is IBM (SCO admit to this). The dispute is whether or not they are *derivative* works of UNIX and thus covered by IBM's contractual obligations of non-disclosure.
If the court finds against IBM then damages will be paid (I imagine very substantial damages) and SCO and its shareholders will be very rich. If SCO loses they are history. However you can guarantee that whoever loses will appeal and this process will literally take years.
But what would the impact on Linux itself be. IANAL, but I imagine the best case would be no impact, and the worst case would be removal of those products from the kernel. In this case cleanroom implementations would be done (which I think if you research the case a little more you will find pretty much all the derivative works were cleanroom implementations at IBM anyway - this will come out in court) and Linux will continue.
The chance that SCO will be able to level licence fees is pretty much none. For people who disagree, think of the MS vs. stacker case. MS lost, had to pay damages and withdraw product, but did stacker get the right to retroactively bill Windows customers? That's right, they did not, and SCO will be in the same boat. Again, worst case net effect to Linux - maybe a couple of months cleanroom time...
Now the second part of their attack, and the one where the FUD is flying so thick you can barely see - the code copying rumours. The first piece of FUD is that this is tied to the IBM case. The simple answer is no, it is not - see above. SCO vs. IBM is simple contract law.
SCO would have us believe entire swathes of the kernel are blatantly copied from SYSV UNIX into the kernel. So far SCO has been unwilling to say *what* exactly is the problem here. They are saying the reason for this is that the Linux team will remove the offending code - well excuse me, is this not the point of relief? The few snippets they have shown have been completely demolished as belonging to SCO by various people. Most have been shown to belong to implementations from BSD or the public domain.
I am not even going to bother to attempt to address their pathetic attempts to attack the GPL via US copyright law.
Now what is getting up the open source advocates' nose is that SCO whats to do exactly what it says it is trying to protect. It wants to steal the IP of thousands of code contributions of thousands of developers to the Linux kernel and utilities. I have a small piece of code within the Linux kernel, and you can bet your buns if SCO tries to collect fees on Linux I *will* sue them for copyright infringement. I have released my code under the GPL, and if they break that license they will hear from my attorney, as they are violating *my* copyright. This is an issue that they seem to neglect - and one that when McBride talks about "protection of IP" makes me want to smack him with a cluestick - Darl you moron, you are *stealing* the Linux developers' IP. Are you going to pay Linus Torvalds, Alan Cox, Donald Becker, Rik Van Riel, Andre.... and the other thousands of contributors for their IP? What's SCO going to do when they get hit with thousands of copyright suits?
I will finish with a metaphor. I buy the assets of Xerox PARC. I then say Apple and Microsoft have stolen my windowing systems IP. I then launch a complex legal suit based on some 25-year-old contracts that have been altered many times, and have in fact been seen in court before. I then send every owner of Windows and MacIntosh an invoice for $699 to license my IP, or I will sue you.
Pop quiz: Your response is to :
1) Pour scorn on me
2) Laugh and throw invoice in bin
3) Laugh and sue me for barratry and mail fraud
4) All of the above...or
5) Write an article on how I might have a case, without researching the facts, and help bolster my case.
Well - most normal people will pick options 1-4, it's funny how most of the tech press seem to be picking 5. This metaphor is *exactly* what SCO are doing.
How well, enough ranting - do us a favour. People listen to guys like you. Research your facts before helping to feed the FUD monster. Oh, and research is not talking to another bunch of Windows analysts, it's actually looking at the history of Unix, SCO, the USL lawsuit and the Linux development process and how it protects against the inclusion of other people's IP, and the rapid removal of same.
And what's a better metric to evaluate a case's merit than the way parties are behaving and their rhetoric? Maybe the conclusion of several other financial analyst groups? Now that's extremely professional...
I thought an "analyst" is supposed to actually have a clue on the matter, not just jump in and project his vision based on the above arguments. What about the technical part, which is the core of the case?
SCO's actions don't make sense, neither at a "common sense" level nor at a technical level. From those perspectives they are outrageously bogus, insulting. But somewhere in between (let's call it "analyst/lawyer level"), things suddenly start to make sense and the case appears to have merit. This never seizes to amaze me!
I can't believe there's not a single mention of the many fallacies in this case. SCO's been distributing GPLed software for years and they still are while at the same time claiming that very license to be invalid! Isn't THAT bothering you, Rob? I guess not, Darl's rhetoric is more important...
First and foremost: IANAL.
> SCO and its legal team continue to bet the farm that they will
> win, and IBM appears to be trying to limit its risk. Also, IBM
> doesn't have complete control of its own side, especially with a
> massive amount of well-wishers who might actually be making
> it more difficult for IBM to win.
This isn't IBM's "own side" launching DoS attacks on SCO. Despite what McBride may say, a lot of the people who are angry at SCO actually care nothing about IBM, are not paid by IBM, and are not being set up as IBM's attack dogs. They're angry at SCO for other reasons, very justifiable reasons.
A few of these angry people are rogue elements. They don't represent the Linux community at large. I, for one, certainly don't advocate DoS'ing SCO and never have.
(Also, there's some current debate as to whether SCO is actually being DoS'd. There are conflicting reports coming out of SCO about their four-day outage--a lot of SCO reps said it was just migration hassles from an upgrade. The continuing sco.com outages that look like DoS attacks are coinciding almost perfectly with SCO USA's off-hours. Makes you wonder, doesn't it?)
> One of the strongest indications I have that SCO might win is
> that several of the financial analyst groups who work with me,
> after reviewing the SCO submissions, have concluded that
> SCO might actually have the advantage. These firms are
> relatively unbiased.
These firms are also relatively uninformed in this matter. The history of UNIX, and the interpretation of the contracts and complaint, is a tangled mess at best. SCO's managed to come up with some pretty interesting-looking claims, but none I've seen that really hold water.
> The company started off as friendly underdog Caldera going
> after Microsoft. Now the company is nasty little SCO going after
> Linux . As far as I can tell, it really is the same company.
The company didn't establish itself as a "friendly underdog." The DR-DOS suit may have scored a few points with hardcore MS haters, but it was also clear that Caldera didn't take up the torch just to fight the good fight. Caldera took up the DR-DOS torch because they saw an opportunity to buy a dirt-cheap dead-end set of assets and use it to wring money out of someone.
Also, the company is definitely not the same. The management in particular is different. Ransom Love, while having a tendency to shoot his mouth off, wasn't quite as lawsuit-happy as Darl McBride.
> SCO was simply saying that IBM had taken SCO's intellectual
> property and was giving it away to the open-source
> community. Given my experience with cases like this, it wasn't
> unusual that a big company was doing something like this. I'd
> seen it before.
There's more to it than that. SCO's original complaint was so ill-founded that they had to amend it heavily--and part of the modification was admitting that IBM didn't actually copy code.
SCO is now simply trying to assert some control rights over IBM's intellectual property. NONE of the IP SCO is suing IBM over is actually registered to SCO as a patent or copyright--rather, it's registered as IBM patents on NUMA, RCU, JFS, etc.
SCO's complaint against IBM, as it stands now, relies on a very interesting definition of "derivative works." Essentially SCO now contends that every piece of code that ever touched System V code, even a filesystem driver originally written for a non-UNIX operating system, or a multiprocessor topology design originally written as clean-room code, suddenly comes under the control of SCO after only fleeting contact with the System V code. Even if that IP has no dependence on SVR4.1 code, concepts, or methods, SCO would have us believe it magically becomes irrevocably tied to SVR4.1.
That's what SCO's entire case against IBM is based on. Their "export restrictions" allegations, their "wrongful dissemination" charges, et al, rely on the "derivative works" concept being overly broad--far more broad than what is defined and generally accepted under current case law.
> What seemed to be different was that the open-source
> community seemed collectively to say, "Theft is okay as long
> as we benefit."
Correction on that. At the time the IBM lawsuit was announced, the stance of the Linux community was, "go ahead, it's between you and IBM. If IBM gave us code under false pretenses, though, we'll be quite happy to remove it."
Unfortunately, SCO stonewalled any such cooperation and quickly started making vague threats against Linux users and distributors--threats like "SuSE and RedHat will have their day of reckoning." This isn't how you make friends or encourage good faith. While the rogue elements of the Linux comunity have handled this dispute poorly, the SCO executives haven't handled it any better.
> Ever watch people who have been caught stealing something
> defend themselves? They say things like, "I found it,"
> "Someone else gave it to me," "It wasn't yours to begin with"
Some of these suspicious-sounding arguments can actually hold merit.
For one thing, the "someone else gave it to me" does indeed hold some water. It doesn't allow you to keep stolen goods just because you received them secondhand--but again, Linux developers were perfectly willing to give up any "stolen" IP. SCO has apparently gone out of their way to make that impossible.
The "someone else" argument holds even more water when you factor in that the someone else was, in some cases, Caldera/SCO itself. The examples of "infringement" they put on exhibit at SCOForum (and later gave to a reporter without NDA) were largely proven to be parts of ancient UNIX code that Caldera released under an open-source license. Other portions turned out to be taken from the BSD codebase--something Linux developers have virtually the same rights to as SCO. Yet more portions of the "evidence" were code previously placed in public domain (and in popular programming textbooks) by the original authors such as Kernighan and Ritchie.
Seeing SCO try to backpedal from that was entertaining, to say the least, and it tore a large chunk out of SCO's credibility. First Stowell/Sontag tried to claim that the code they open-sourced was for non-commercial use only. Anyone who read and understood the accompanying license text knew better--there was no mention at all about commercial or non-commercial purposes. Now SCO's official line is that the SCOForum examples weren't actually examples of "infringing" code.
I wonder, then, what were the samples supposed to be? What they turned out to be was a major faux pas on SCO's part.
> To extend the example to the book-publishing world, some
> open-source proponents have argued that if you started with
> Harry Potter and the Sorcerer's Stone, you could end up with
> Parry Hotter and the Wizard's Rock and be just fine. But this
> sort of line-by-line replacement won't work because the ideas
> that surround the product are also protected.
This is a common misconception. I'm sorry if this sounds like a flame, but I just have to call this one out as classic ignorance of IP issues.
The ideas surrounding SVR4.1 are NOT protected--or at least, not for SCO's benefit. SCO only has the SVR4.1 COPYRIGHTS, plus various heavily amended contracts of questionable power. Novell holds the System V patents and is perfectly happy to let the patented ideas be used in Linux. SCO's contracts might give it power over derivative works--but then again, SCO's idea of "derivative works" is overly broad.
Copyrights only protect the expression of an idea, not the idea itself. If you want to protect an idea or concept, you need patents--of which SCO has none. Or rather, I'll amend that--SCO holds a total of one solitary patent on an obscure bit of clustering methodology. Certainly not much help here.
As for Harry Potter analogy, it's a very bad analogy--especially as this is the C language being disputed, not the English language. The C programming language is far more constrained and has far less variation than English and all its dialects, so it's far more prone to "code convergence"--the coincidental duplication of similar or identical code expressions.
> Were you to read the original AT&T license, which I did, you
> would find that it anticipated things like this and, if it is
> enforceable, protects against it.
If you read the extra amendments, you'll also find certain clauses clarifying under what circumstances IBM can divulge concepts and methods. If someone else already divulged certain concepts and methods without IBM's involvement (such as Caldera independently open-sourcing some UNIX codebases), IBM can freely do the same. If someone at IBM happens to retain and divulge "residual" memory of System V concepts and methods---well, that's OK too. Even the fully amended clauses relating to derivative works leaves some leeway for IBM, no matter how broad SCO's definition of derivative works may become.
> Let's also pretend that, when there is a doubt about ownership,
> you have the right to prove that ownership and that no group of
> vigilantes or large companies has the right to force you to give
> up what you can prove is yours, or take away your right to try.
Let's also pretend that you have not only the right to prove ownership, you also have the responsibility to do so. This is a critical first step in mitigation of damages.
As far as IBM is concerned, SCO has supposedly done all it can--or all it's willing to do--to prove "ownership" until it faces IBM in court. Whatever my opinion (or your opinion) may be, that's not our problem at the moment.
As far as the Linux community is concerned, SCO has failed miserably at proving its claims--those claims being "millions of lines" of SVR4.1 code being illegally merged into the Linux codebase. What was originally supposed to be proof (at SCOForum) turned out to be a dog-and-pony show. Now, supposedly, it wasn't supposed to be proof, if you swallow SCO's party line. Yet SCO still contends that it has the right to collect licensing fees and is even threatening to start suing end-users over it. It's been making threats like this almost from the first filing of the IBM complaint.
You may want to look up the term "barratry" as it applies to current case law. That's more-or-less what SCO is trying on Linux users at the moment. It's quite possibly what SCO tried on IBM--which, if true, would be a stupid move on SCO's part. Trying to frighten IBM with lawsuits is like trying to frighten a tiger with a Nerf bat.
Only one comment on this, they don't have to prove anything to the Linux community which likely would not accept any level of reasonable evidence anyway. They have to prove themselves to legal departments and a court. Court is coming at least some legal departments appear to find their arguments compelling.
I mean really, when have you ever seen someone in a case like this just show their evidence and have that be enough for the other side to cave? This is way too complex for that, and the vast majority of people engaged don't have the legal background to make the proper determination anyway.
If you are right, they will lose, if you are wrong and haven't prepared yourself or your company, then good luck in your next job (assuming the blame falls on you or your department). Why take the risk? It shouldn't be your problem to begin with.
The other outcome of the situation is that if SCO wins, they get 3 billion dollars from IBM and the damages have been paid. They are claiming that IBM's alleged transgressions have and will cost them 3 billion dollars (a dubious number considering they have only one measly profitable quarter in their existence). This is not a claim as to what it has cost the company since the 2.4 kernel came out as even IBM hasn't made that much and SCO could never prove they'd earn a fraction of that.
So, in theory, if SCO wins and gets 3 billion, the case is over. The transgression occurred IBM paid for it, SCO made more than they ever could have, and life goes on for everyone. The only way for it to affect the linux community is if SCO is not awarded 3 bln but some reduced amount based on determined losses since the infringement occurred and the right to charge linux users currently.
Since SCO isn't worth anywhere near 3 billion, if it looks like they are going to lose IBM would simply buy them out.
> Only one comment on this, they don't have to prove anything
> to the Linux community which likely would not accept any level
> of reasonable evidence anyway.
You certainly seem to have a bitter, twisted conception of the Linux community. I'm pretty certain it's a misconception, but I don't think I could reasonably convince you of that. We'll just have to agree to disagree on that point.
If SCO wants to have any hope at all of collecting licensing fees for Linux, SCO WILL have to prove its case to the Linux community--not necessarily the IBM case, but the more general allegations of code theft in Linux. SCO has been making blanket accusations against Linux users and developers for ages now and has so far been unwilling or unable to back up their statements.
If SCO provided real, irrefutable evidence of their allegations in an open fashion, Linux developers would accept it and work towards a real solution, preferably by removing the illegal code. So far SCO has failed at that. The SCOForum "evidence" was rejected simply because it was obviously bunk. Even SCO's admitted that now.
>at least some legal departments appear to find their arguments compelling.
Those legal departments must not have done much research into this case then. Many others have and determined that there is no case.
What we really have here is an example of a corporate amnesia. The top executives at SCO are all pretty new at the company and have no real idea what Caldera did as far as merging technology from Linux and SysV - they even published white papers on the topic. Not to mention the volumes of original Unix code that they've open sourced, was widely available w/o NDA in acadamia, or the outcome of the legal battle between AT&T and BSD, which they intend on repeating. They have forgotten (or ignored) all of that and see every similarity as an example of where they were wronged.
Maybe Alzheimer's is a better analogy. They are dying and lashing out because they don't know where they are and where they came from.
Why take the risk with any technology? Who is to say that the latest proprietary software doesn't have holes that will cause your company to lose buckets of money due to downtime and/or corrupted data? Or that the company will be sued because they've stolen code?
"If you are right, they will lose, if you are wrong and haven't prepared yourself or your company, then good luck in your next job (assuming the blame falls on you or your department). Why take the risk? It shouldn't be your problem to begin with."
Because there is no risk to Linux *end users*. The case is a contract dispute between IBM and SCO and *can not* have any bearing on third parties even if SCO wins.
"Only one comment on this, they don't have to prove anything to the Linux community which likely would not accept any level of reasonable evidence anyway."
I find this a rather offensive characterization, Rob, and very unfair to those people actually working on the linux kernel, and to public community leaders like Eric Raymond. I've seen thoughtful commentary (on LWN.net as well as elsewhere) to the effect that SCO's example of the SGI-contributed malloc code may well have been improperly contributed, and that Linux developers need to be vigilant and ready to remediate such issues as they are brought to light.
SCO is claiming an interpretation of the AT&T contracts that would forbid UNIX licensees from taking any independently implemented technology that ever touched a SYSV-derived UNIX from being used without their permission. It is on that basis that they are suing IBM. That's fine, they have a legal theory. IBM has a legal theory in contradiction of SCO's theory, and IBM will defend their theory. If SCO persuades the courts that their intepretation of the AT&T contracts are correct, then fine, those features come out of Linux, no problem. We've got a surfeit of journaled filesystems already, after all, and we can live without RCU and any specific implementation of NUMA.
If that was all that was going on, the linux community would not be kindly disposed towards SCO, but it wouldn't be that big a deal.
SCO is going much farther than that, however.
They have attempted to publicly argue that the GPL is incompatible with US copyright law. They have insinuated that if IBM were serious about fixing the Linux problem, they would somehow 'move away from the GPL'. These are not arguments for remediation of copyright violations, these are arguments for remediation of the fact that Linux exists, no matter who wrote it.
You yourself have argued that open source is vulnerable to this sort of attack because the source code is open for SCO to find problems with. That's fine, we'd love to have the problems identified and repaired.. that's the whole point of open source. You should understand that we'd expect a member of our community like SCO to at least share the ideal of transparency enough to be clear and consistent in what it is arguing, and to provide light as well as heat in its public argumentation.
But when SCO makes a variety of bizarre, ever changing and unsubstantiated public allegations, when SCO pretends to have Linux's best interests at heart while arguing that it should 'move away from the GPL', when SCO attacks Linux users and insults the skills and talents of Linux developers, it's hard not to take it a bit personally.
I have read every single line of SOFT-00015 and amendments as submitted to the court. I have no idea how anyone could reach your conclusions.
Please see an analysis of the contracts at:
Another anti-Linux piece from Rob Enderle. Oh sure, Rob has been writing about Linux "on and off" for years. Always negative. Always critical. Always inflammatory. He never gives constructive criticism, either. It's just plain nastiness.
Rob says "What seemed to be different was that the open-source community seemed collectively to say, 'Theft is okay as long as we benefit.'". Nobody of prominence in the Linux community has said anything of the sort. If Rob took off his anti-Linux sunglasses then he'd realise that SCO is the one trying to collect license fees on the undisputed 3.6 million lines of code to which SCO isn't claiming ownership. That sounds like Rob's "theft" but it is SCO who is the thief.
Rob says "The company started off as friendly underdog Caldera going after Microsoft. Now the company is nasty little SCO going after Linux. As far as I can tell, it really is the same company". Despite a different CEO, a different product line, and a different attitude towards Linux... yeah, exactly the same. Caldera made it very clear that they were going to integrate UNIX and Linux as part of their 2001 business strategy. Now they allege that this has happened without their permission and they want compensation. If this is the "same company" then SCO is schizophrenic.
Rob says "Let's also pretend that, when there is a doubt about ownership, you have the right to prove that ownership and that no group of vigilantes or large companies has the right to force you to give up what you can prove is yours, or take away your right to try". Sure, but SCO refuses to prove that ownership despite repeated requests from Linux companies (Novell, RedHat, IBM) and Linux developers (Linus) and Linux advocates (Perens, Raymond). SCO wants people to believe they have ownership but they don't want to prove it.
I don't know why Rob hates Linux, nor do I care. I just know that his articles are always a good sign that there's plenty of bark and no bite; if Rob thinks SCO has a case, then Linux must be safe.
I have read both this article and your previous article concerning the SCO vs IBM fracas and I would like to make a few comments. Your first article posed a question which was and is legitimate "Suppose SCO is right?" Never mind the actual probabilities of the situation, that possibility does exist. At the time you published that article, there was very little information forthcoming from SCO on which to make much of an assessment of its claims.
I do not know any of your legal qualifications so I cannot comment intelligently on your assessment of SCO's chances of winning their court case, but there are not many of the lawyers that I have read who think their case is very strong. Some of those lawyers are fervent OSS advocates, and others seem to have no axe to grind.
There is one thing that you did not do, and that is to distinguish what is the actual legal case they are trying against IBM and the press campaign they are running. The IBM lawsuit is about breach of contract and divulging of trade secrets, not copyright infringement.
I have read the copies of the AT&T contracts posted on the web and also the other contracts and side letters which put a different light on the derivative works picture. I have read as much as possible concerning the USL vs BSDI case and its outcome which shed considerable doubt on the validity of any copyrights in the 32V code. And the letter which AT&T sent out to licensees clarifying the derivative works clause in their contracts. Do you know what that letter stated without checking on it?
Did you know that SCO does not own the copyrights nor patents to the SMP, NUMA, RCU, nor JFS that is the heart and soul of their breach of contract lawsuit, and their spokesman, Blake Stowell, has publicly admitted this?
Have you checked on the various analyses made on the snippets of code SCO showed in Las Vegas recently? Especially the Berkeley Filter code.
The other snippet of code that was shown was copyrighted by SGI, and its origins go all of the way back to 1973. This was evidently the code seen by Bill Claybrook and Laura Didio. Bill Claybrook is the only analyst that I have heard from who has any knowledge of C programming and UNIX programming.
I do not think that your characterization of the OSS community is correct or fair, no more than I think that the attacks on you for your first article were warranted or fair.
Your statement 'What seemed to be different was that the open-source community seemed collectively to say, "Theft is okay as long as we benefit." does not square too well with the actual facts.
The open source developers collectively do not believe that there is any infringing code in Linux. They, the developers, know what they have wrought and put into the kernel. There are no secrets in the code put into the kernel.
But from the very first, Linus Torvalds and other open source programmers have said that if SCO would show the code, and indeed if it were infringing, it would be removed.
Your Harry Potter analysis is off the mark here, although it can be more aptly applied to SCO itself in this case from the code that was actually shown in Las Vegas. The Berkeley Packet Filter code they showed as theirs did not have the requisite BSD copyright notice in it. It does not belong to SCO. SCO has public domain memory allocation code which has some changes to it, and is claiming a copyright on it, although except for some cosmetics, it is the same as the code first published in 1973.
That is also something SGI did and released it under the GPL. That code had already been removed from the 2.6 kernel before it was known that it was code SCO was claiming copyright to.
Now back to the point. Removing infringing code. The open source community does not want to use code that is not free. They do not want to steal code. If any infringing code comes to light, they will remove it and perform a clean room implementation. This is what Jay Schulist did with the Berkeley Packet Filter algorithm he wrote for the Linux kernel. He used the published specs for the Berkeley Packet Filter to write his own implementation. I think that you are wrong on copyright protecting ideas. It protects the expression of those ideas. The BPF is a case in point.
Linux has been written mostly from scratch using the POSIX specification.
Also, in my reading, it seems to me that SCO has the duty to mitigate damages by allowing unwitting infringers to stop infringing. Please read up on this. The trade secrets defense is not viable. If there were any trade secrets, they are no longer. If SCO were to actually win the case against IBM on the trade secrets part, that still would not affect Linux. Since SCO does not own any of the patents or copyrights on the SMP, RCU, NUMA, nor JFS technology, they cannot cause it to be removed from Linux now that it is no longer a secret technology. This is from several lawyers conversant with the case. Their only dispute there is with IBM and any possible breach of contract there. Which seems to be non-existent since JFS, NUMA, RCU, and SMP as contained in Linux do not contain any SYS V code according to programmers who have access to the SYS V sources.
Have you been following SCO's public claims? Rather than go through all of that again, I will point you to a url where it has been done: http://radio.weblogs.com/0120124/2003/08/19.html.
The open source community has been doing much more than howling, whining, sniveling, etc. They have actually been doing the research on the subject matter and putting their cards on the table and asking SCO for a refutation. All that the different SCO spokespersons have said so far is that they are right without any substantiation whatsoever. When they tried to produce some evidence, it was pretty well blunted by some well known experts.
That is really what we have here. You have talked to some analysts who are from financial groups who have been shown snippets of code under an NDA which restricts what they can say and divulge. Are they programmers? Do they know anything about UNIX programming and history of UNIX code? SCO has publicly shown some code. Can your analyst friends tell you if the code which SCO has shown is the code that they saw? Could they tell you if anything that has been shown and the analyses have changed their minds any? Have you talked to Bill Claybrook?
SCO has the right to protect their intellectual property. They have the right to go to court to prove what rights they may have and to obtain relief for any wrongs that may have been done to them. That remains to be proven in court. My opinion on their chances differs from yours to a considerable degree because of the information I have found in my own readings.
You certainly have the right to your opinion. I object to the one-sided tone and some of the characterizations you have made in your articles. But that is your right. I don't know if you are playing devil's advocate, then I would expect such an article, putting the rosy face on one side and painting the other with a dirty brush.
There are those in the open source community who do the same. Although they are fanatically sincere in their beliefs. And there are others who ask some hard questions, and someone else will come up with the answers.
Unless SCO can come up with a lot stronger evidence than has been shown thus far, their case does not appear to be very good on any front. The contracts are pretty much public. The Linux code is very public. The only thing that is not really easily public is the SYS V source code. But if you have read the OSI position paper on the lawsuit, it just may be moot anyway.
My question, what is the difference between journalism and analysis?
This has been a bit long, so if you are still on board, thanks for hanging in there.
Interesting post and it looks like you did a lot of work. Wish others had done the same. I donít think we disagree on Copyright it does protect the expression of ideas. The real question is ďwas Linux from scratch or did it borrow from UNIXĒ as SCO claims and evidently IBM admits (it was this admission that started this whole mess in the first place, well coupled with a falling out between IBM and SCO). All Iíve ever said was that SCO has a case and that people at risk should turn this over to their legal departments for proper risk mitigation and then move on. Folks who are extremely risk adverse may want to shift to BSD or stay with their proprietary platform until this is settled, and employees should be reminded that being an OSS advocate does not release you form company policy. (Iíll bet some of the comments people are sending around would get them fired given the zero tolerance policies that exist in some firms, and Iíll bet there is an IT manager sweating bullets wondering if one of his (or her) folks is actually doing the denial of service attack).
Your view is the risk is slight, mine is it isnít, in the end we agree there is risk. The people to take it are attorneys, let them do their job and listen to their advice. Why make this a personal risk or one that puts your department or company at risk if you donít have to? This isnít a great time to be unemployed and that first article you mention was written largely because of my concern that too many IT folks were putting their jobs and the jobs of their people at risk needlessly because they werenít doing a proper job of justifying their decisions.
Thanks for posting!
> Interesting post and it looks like you did a lot
> of work. Wish others had done the same. I donít
> think we disagree on Copyright it does protect the
> expression of ideas. The real question is ďwas
> Linux from scratch or did it borrow from UNIXĒ as
> SCO claims and evidently IBM admits (it was this
> admission that started this whole mess in the
> first place, well coupled with a falling out
> between IBM and SCO).
First: you claim IBM admits that Linux borrowed from Unix. Prove this. What exactly did "IBM" say and who from IBM made the statement?
Second: Have you ever heard of POSIX? It's an international standard for an API to implement a Unix-like operating system. If you fully implement POSIX, you get a Unix compatible kernel. POSIX has been an evolving standard for over 30 years.
I have been reading your responses to the different posters and I have noticed that you have not responded directly to any of the points that I or any of the other respondents have raised.
I have no idea how much research you have actually done on this subject or what opinions you have about the points that have been brought up, because you have not addressed them, except with your opinions. I think that is what most of the posters find annoying with your articles.
You did attest as to why you believe the way that you do, based upon discussions with your analyst friends, but there have been many counterpoints raised by others who have given some pretty good reasons for their opinions which you have not addressed.
You do seem to be accepting SCO's word for a lot of things, such as "IBM started this ruckus". However, that point has been addressed by IBM themselves. And a few people have been going back and checking out SCO's public comments and their subsequent actions since at least the first of the year. Some things just do not add up.
Would you care to comment on SCO's announcement that they are planning to send invoices to Linux end users that they have identified through various means? I would really like to hear your opinion on this.
I've noticed the same thing, Bob doesn't seem to be able to defend his position with facts.
Concerning SCO's credibility: I would like to see Bob comment about SCO's claim they were going to audit AIX users as well. I would like to know why Bob thinks that SCO has a chance in court when they've lied about their rights so publicly so often.
I'd bet money that SCO will never ship any licenses, because, quite simply, it's blatant extortion and SCO is aware of this. However, it makes for good PR, even if it's an absolute lie. I'm sure SCO may sell some licenses to "friendly" companies though, like Microsoft.
What about SCO's "revocation" of the Dynix license as well? It's just another blatant lie. I don't see how a company that has such a problem with the truth can be expect to prevail in court.
Or how about this quote from
"Asked why SCO has suddenly started looking at these issues now, after years of declining revenues at his company and the increasing popularity of Linux, McBride said SCO had few options in the late 1990s as Linux began surfacing in the business computing world. "Even if you potentially had a problem [with concerns about Unix code in Linux back then], what are you going to do?" McBride asked. "Sue Linus Torvalds? And get what?""
That statement by McBride alone blows SCO's case out of the water since even if there was infringement, SCO allowed it to continue knowingly and purposely to increase their damages.
> I donít think we disagree on Copyright it does
> protect the expression of ideas.
The expression of ideas, but you claimed that the ideas themselves are also protected. The US Copyright Office disagrees with you.
"How do I protect my idea?
Copyright does not protect ideas, concepts, systems, or methods of doing something. You may express your ideas in writing or drawings and claim copyright in your description, but be aware that copyright will not protect the idea itself as revealed in your written or artistic work."
Rob, I'd really like to hear your response to this from www.copyright.gov since it completely destroys the entire premise of your article.
> How do I protect my idea?
> Copyright does not protect ideas, concepts,
> systems, or methods of doing something. You may
> express your ideas in writing or drawings and
> claim copyright in your description, but be
> aware that copyright will not protect the idea
> itself as revealed in your written or artistic
'The real question is ďwas Linux from scratch or did it borrow from UNIXĒ as SCO claims and evidently IBM admits'
Linux has been in development for over 10 years. It's development has been very public, and very well documented. So the answer to your 'real question' is 'Yes, linux was developed from scratch'. I thought that was pretty obvious. It used ideas and based it's structure on Unix, but if it ever stole anything then it would of been revealed a long time before now.
I would like you to answer the following very simple questions.
1) Why doesn't SCO reveal which code they believe to be stolen to the linux developers, or at least a few of them under a signed NDA?
2)Why don't they want the linux developers to remove the code?
3)Even if some of the code was stolen from them, why does SCO believe they have the right to charge $699 for the rest of the code they did not write?
About question 3. They are not billing people for just the SCO code. They are billing people for linux itself. If they were billing for just the SCO code then I would have the right to make them tell me which parts I am purchasing so that I then have the option of obtaining a competetors version of that product. But as we all know, SCO are not revealing which parts of the linux kernel are theirs.
You have ignored the possiblity that companies that sign SCO's new license agreement will put themselves more at risk! Remember SCO's own statements that contracts are very powerful. Who, then would want to sign a contract with SCO that allows audits, etc.?
Oh, and on the independence of financial analysts, are you including Gartner Group? It seems that there are some financial connections between Gartner Group and SCO!
> Remember SCO's own statements that contracts are
> very powerful.
SCO's exact statement was:
"Copyrights and patents are protection against strangers. Contracts are what you use against parties you have relationships with. From a legal standpoint, contracts end up being far stronger than anything you could do with copyrights."
I still have to laugh that these buffoons are being taken seriously with statements like that.
I think you may have missed some relevant facts.
First - there is likely to be NO Denial of Service attack. SCO's website is strangely up during business hours and down in off hours. What sort of hacker is polite enough to terminate a denial of service attack during non business hours? It's more likely that SCO themselves are taking the website down when they aren't at work. You can consult www.netcraft.com for more information regarding that: http://news.netcraft.com/archives/2003/08/26/is_the_sco_site_down_again.html
Second - SCO has made claims that there are thousands of examples of infringement, yet in Germany, when LinuxTag took them to court, they were unwilling, or more likely unable, to present just one example to prove it. The result was that SCO in Germany was barred from claiming that Linux contained SCO IP illegally copied into it within Germany. In fact, SCO in Germany was just fined 10,000 euros for breaking this agreement.
Third - SCO's examples that SCO showed at SCO Forum have all been completely debunked. One example that SCO was claiming was an example of stolen code was actually Berkeley's code. Another example is code that is in the most famous C book of all time, written by Kernighan and Ritchie, the inventors of Unix and C. I read this book when I was a freshmen in college. It's the bible of C programming written by the gods of C programming and Unix. It's the most available book on the C language in the world. I have a Turkish friend that has the Turkish version.
Fourth - in your divorce analogy, you are incorrect when you state that SCO is the same company that sued Microsoft over DrDOS. The current company is run by completely different people now. There isn't much of a development team left at SCO either. What SCO is today is a shell of its former self with different people running the company.
Fifth - there is no benefit in stealing proprietary code and placing it into Linux, as you can clearly see. Furthermore: it's not true that anybody in the Linux development would condone such an action as theft. If there is infringement, we want to mitigate the problem. This is standard fare in IP litigation, and we know this, since almost all of us actually have jobs and have seen IP litigation. SCO is unwilling to mitigate, which is unprecedented, or to be more plain about it - bizarre.
Sixth - stealing code and placing it into Linux would be a pretty foolish thing to do since all the code is public and can be reviewed. Not only is the code public, but all of it can be traced back to the original author. The punishment for being caught is having your career destroyed and then excommunicated from the Linux development community. Who would be stupid enough to do such a thing? What is the benefit to the developer in doing this? It's a volunteer effort after all. Why would IBM do this when they have AIX?
Seventh - who would have the greatest access to SCO's Unix code in the first place? That would be SCO employees. SCO was modifying Linux themselves. One example that SCO showed at SCO Forum wasn't only traced back to Kernighan and Ritchie, but SCO themselves when they published SVR4 and gave it to the Linux community to use any way they saw fit - back when SCO (then Caldera) was still a Linux company.
In closing, if SCO could prove that IBM in fact did steal code and place it into Linux and jeopardize a 10-year effort representing millions of man hours of work created by volunteers, you can bet that IBM would be an enemy the next day. SCO claims more than a thousand examples of infringements, but is unable to name a single one publicly. This is contrary to all IP infringement cases I've ever seen. Why doesn't SCO take ONE example, and simply prove it's true?
It's all too easy to blame Linux developers since they aren't represented by a company, but it doesn't make any sense. The risk of being caught is very high, and the penalties are equally high and there is no monetary benefit.
If SCO indeed has a case, what they should do is get an injunction against kernel.org which is the entity that distributes Linux - SCO can certainly afford the bond. It's awfully telling that SCO hasn't requested an injunction from having their IP illegally distributed. It's very odd that SCO settled against LinuxTag in Germany. It's strange that their examples of infringement were both debunked, and yes they were debunked, the Linux developer community isn't being coy with words to say that the examples were debunked. SCO has repeatedly changed their story, and all the insiders who are making financial transactions are selling stock.
It doesn't take a Linux developer to know something is up.
I know the eventual outcome of this and found your odds a little surprising. IBM has a good history in the last decade of settling when they find themselves doing something wrong. IBM could have purchased SCO for $5/share back in December if they so desired. This is pocket change for a company that makes billions in *profit* every Q. IBM is quite certain they did nothing wrong, and based on the waffling of SCO over the last 6 months, I'm quite certain that SCO doesn't have any case at all.
I give SCO a 5% chance of prevailing and a 30% chance of being forced to pay restitution to Red Hat for torturous interference.
A couple points:
Regarding SCO being "the same company" as friendly underdog Caldera, that isn't really true; they are under new management, and have shed much of their R&D. The old Caldera embraced and promoted Linux and open source; the new SCO says that the GNU Public License is invalid because (they claim) it clashes with U.S. Copyright Law. That's a pretty dramatic reversal.
The open source community may seem to you to be saying, "theft is ok as long as we benefit" but that seems to me to be a gross misreading of the response to SCO's claims. I have not heard anyone in the open source community defend theft of SCO's IP. The community has called on SCO to provide evidence for this theft, and SCO has refused to do so. SCO's claims also stretch credulity ("millions of lines" of infringing code, representing a substantial fraction of the Linux kernel, which is a collective work of hundreds of developers).
You say that the open source community is responding like someone caught stealing. Claiming innocence is of course a common tactic of criminals... but it is also one chosen by innocent people from time to time.
Regarding the impossibility of rewriting the offending code, perhaps you are not familiar with the BSD system, which could be characterized as a piece-by-piece reimplementation of AT&T Unix. The AT&T/BSD litigation settled the issue of whether this sort of reimplementation was legal. This sort of rewrite would be impossible if SCO held patents that covered more than a particular instantiation of required functionality. SCO did not acquire any Unix patents from Novell when it acquired the System V code base. Their claims against IBM are based on contract violations, not patents.
A couple of things, first they didnít get the patents from Novell because AT&T still owned them (and is apparently negotiating to sell them to SCO). Novell has been nearly clueless in both what they owned and what they sold doing no one any good but SCO at this point.
Caldera was partially, at least in my opinion, conceived to make a run against Microsoft (Ray Norda had a thing about Microsoft). Now they are making a run against IBM and Linux. From my perspective they just changed their targets, and name, after hitting the first one twice. If they were just going after IBMís AIX or still hunting Microsoft I doubt many would be this upset.
You are right innocent people may respond like this too, my experience is that innocent people are more likely to blow this kind of thing off (often to their regret) because they trust too much in the system. Guilty people donít trust the outcome (for good reason) and are more likely to go ballistic screaming to high heaven that they are innocent. Thatís just my perspective.
I am familiar with the BSD system and the rules surrounding it were incredibly well crafted, a number of the BSD folks have indicated that Linux is no where near as well protected and it was this audience that told me they were pilloried when they brought this up.
> You are right innocent people may respond like
> this too, my experience is that innocent people
> are more likely to blow this kind of thing off
> (often to their regret) because they trust too
> much in the system. Guilty people donít trust
> the outcome (for good reason) and are more
> likely to go ballistic screaming to high heaven
> that they are innocent. Thatís just my
So you admit that innocent people have good reason to not remain silent and trust the system, but because we aren't remaining silent and trusting the system, we must be guilty? Nice circular reasoning.
> I am familiar with the BSD system and the rules
> surrounding it were incredibly well crafted, a
> number of the BSD folks have indicated that
> Linux is no where near as well protected and it
> was this audience that told me they were
> pilloried when they brought this up.
The "BSD folks" shouldn't be resting easy, because SCO has been making noises of going after the BSD variants as well, claiming that "there could be issues with the [BSD] settlement agreement". Of course, that settlement was sealed, which makes it difficult for the BSD development groups to adhere to it, but it seems that SCO's preferred strategy is to go after people for violating its "rights" and refusing to tell anyone what needs to be done to stop violating their "rights".
> The "BSD folks" shouldn't be resting easy,
> because SCO has been making noises of going
> after the BSD variants as well, claiming that
> "there could be issues with the [BSD] settlement
> agreement". Of course, that settlement was
> sealed, which makes it difficult for the BSD
> development groups to adhere to it, but it seems
> that SCO's preferred strategy is to go after
> people for violating its "rights" and refusing
> to tell anyone what needs to be done to stop
> violating their "rights".
Actually - the BSD license is on FAR shakier grounds because the BSD license doesn't require re-release of the source code and allows, essentially, relicensing.
If SCO claimed that their code was ilegally copied into BSD Unix, they could claim to actually OWN all the BSD code in that branch after the insertion.
If Linux was under the BSD license and not the GPL, there would be far worse consequences, because SCO could then claim to own Linux outright from 2.2 onward and since the BSD doesn't force source code distribution, SCO wouldn't be under any legal obligation to identify the IP in question outside of the legal requirement for mitigation, which SCO hasn't done anyhow.
"This is because the ideas that surround the product are also protected."
You are wrong. The two ways to protect an idea are with a patent, and trade secrets. SCO is suing IBM for divulging trade secrets, but the problem with protecting an idea by secrecy is that if it leaks out, you have no recourse against anyone except the divulger of the secret. Once it's out of the bag, it's out. Users of your ex-secret owe you nothing.
SCO is mainly talking about copyright - so what? USC17-102b clearly says "(b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work." Also, if it's copyright, it can't be a trade secret.
"Were you to read the original AT&T license, which I did, you would find that it anticipated things like this and, if it is enforceable, protects against it."
Have you read the modifications, the "side letters", where AT&T grants IBM the right to modify and own the modifications? Do you know how copyright law treats "derivative works"?
Here's how I see it.
Say I sell yogurt. Another yogurt fellow comes along and says "Someone stole some cultures from me and put it in your yogurt 10 years ago without your knowledge. I know it, I can prove it, and it's still there. Because your yogurt contained and, in fact, still contains some of my product, you now owe me a percentage of each sale you've made in the last decade, and every sale from now on."
I say, well, prove it's here and it's yours, and I'll remove it and replace it with cultures of my own devising, end of problem.
The owner says "No, I don't think so. Just keep giving me money."
Now, he could have proven, as he says he can, that his cultures were used in the manufacture of my yogurt, but he's choosing not to. Oh, he shows a couple slivers of slides with some other substance mixed in, but they could just as easily be from Dannon or TCBY or any other yogurt made.
Until they are ready to prove something, they've got no right to charge for anything. As far as the law is concerned, until they prove their cultures got into our yogurt, we still own the yogurt. And since they refuse to help us remove their property, they obviously aren't too concerned about ongoing damages.
Well, I too have read the contracts. That is why I sincerely believe that SCO does not have much of a case. There was a side letter signed between AT&T and IBM indicating that IBM owned rights to all code produced by IBM.
Even if that were not the case, as to arguments that simply rewriting won't be sufficient--For the items that have been publically identified by SCO (RCU, SMP, NUMA, et al.), all of those items are covered by patents owned by IBM, SGI or others. SCO has no relevant patents. The code in question is copyright by IBM or others, but for those items SCO has identified, not copyright by SCO.
SCO seems to be relying on some very strong interpretation of the concept of derivative works, which is exactly what seems to be allowed in the IBM/AT&T side letter.
Am I biased? Yes, but I do try to understand why the market seems to be convinced SCO has a case. While it seems to me that Robert Enderle has read the contracts, I don't see any indication that he has read the side letters. I would really like to see an argument made for SCO's case that takes that and the IBM patents/copyrights on the code into question. I definitely realize that I do not know as much about contract law as many--but I also feel I know enough about how lawsuits work to realize that many in the Linux community sound like rosy-eyed idealists when talking about how a judge is going to react. So, why is the side letter ignored in any discussions of the merits of SCO's case?
I sincerely agree that any attacks/threats against SCO, sco.com, or its CEOs are childish, as well as unfortunate for the open source community. Let's just say that there are quite a few youngsters in the group. HOWEVER, this whole fiasco was obviously planned from the beginning. SCO bought the IP for the purpose of suing. They are most decidedly not a victim here. Take a look at this Forbes article, which I found on SCO's web site.
They are actually advertising this link ON THEIR HOMEPAGE! My jaw dropped when I read this, and I felt true pity for souls who aim to get rich in this manner. It's scary. So from a scientific point of view, yeah... they may have a case. But if I were the CEO of the SCO group, let's just say I'd be having trouble sleeping - my conscious is so hard to ignore, the pesky thing...
Honestly, this isn't the path I'd take either. But do think that the system needs to see more of this before it will change. Too many companies are acquiring IP and living off the work of others. It is making it very difficult for new businesses in the US in particular and that will be problematic for the State long term.
A good topic for a future column... Thanks
There is a side letter between IBM and AT&T that to my eye says that IBM owns the right to all code that it adds to Unix.
As to the claim that line-by-line copying won't take care of any problem if in fact there is a direct copying issue, for the material that SCO has identified like RCU, SMP, et al., IBM or SGI or others own the patents on that technology.
So, you have patented material written solely by IBM, a contract side letter that seems to clearly state that derivative works are owned by IBM, and I cannot see why many who look at this issue seem to think SCO has a case.
Am I biased? Yes, but I do try to keep an open mind and do try to understand what is going on. However, even when I read thoughtful articles like this, it seems to me they are basing their conclusions solely on the original or on the SCO/IBM contract language. Why does no one think the side letter relevant?
The claim that ideas are SCO's are completely wrong.
NUMA, JFS are not SCO's. They do not have any patent on it either. Please do not mislead with wrong information. This only begets speculations about the source of bias. I have not read ATT licence, but if ATT licence demands all rights to all ideas that went into AIX be bequeathed to SCO, I am sure it is not enforceable.
You completely ignore the people in legal community who are favoring IBM, instead concentrating on just the non lawyers. 'Rhetoric lacks Reality' applies to your analysis.
Maybe, while Iím clearly not an attorney I did write a number of the contracts that IBM used in the same time frame as the AT&T license and we believed at that time language like this was enforceable, breaking this license will have some interesting implications for IBM and their own contracts and licenses. I believe the license will hold or that IBM will recognize the risk and not work to break any of the major sections. But we will clearly see. One thing is clear if they could get it killed, with the cancellation of AIX, they (IBM) would have done that by now to remove the cloud from that product. I have never seen an IBM product put under a cloud like this and it still looks to me like someone over there really messed up.
> One thing is clear if they could get it killed,
> with the cancellation of AIX, they (IBM) would
> have done that by now to remove the cloud from
> that product.
What cloud? SCO claimed to have canceled the AIX license, and IBM filed a countersuit disputing that. Since SCO never attempted to get any type of injunction to stop the use or distribution of AIX, there's nothing for IBM to "get killed". IBM seems to have made a reasonably proportionate response to SCO's claims, so I think that it demonstrates a bit of bias for you to be claiming that any cloud is over IBM.
> I did write a number of the contracts that IBM
> used in the same time frame as the AT&T license
> and we believed at that time language like this
> was enforceable, breaking this license will have
> some interesting implications for IBM and their
> own contracts and licenses
I would also like evidence to support this claim, Rob. I wouldn't think that IBM would have a non attorney writing contracts. It's not as if IBM's legal department is so tiny it needs help from amateurs.
Out of curiousity - just exactly what work did you do for IBM? You don't seem to be an engineer, you're not an attorney - what then?
"I did write a number of the contracts that IBM used in the same time frame as the AT&T license and we believed at that time language like this was enforceable,"
By even saying there is a "cloud" over IBM is to assume that SCO is right. If you really want to provide fair analysis, why would you take any side's claims as fact? Read the IBM countersuit, it says that they have an irrevocable license.
Should you not give that claim equal weight?
Otherwise why even discuss this?
I wonder how is it that you could come to the conclusion that "someone over there really messed up" if you are equally weighing both company's claims? (and other facts of the case which are detrimental to SCO's case)
How about SCO being negligent in releasing it all under the GPL? How can anyone try to claim damages that stem from their own negligence?
That's preposterous, and would appear to wipe out SCO's case.
But if you are unwilling to consider that (which is not a claim at all, but a fact), what is the point of even discussing it?
Well no offense, Rob, but this is actually about more than just IBM possibly being sneaky. It's about SCO first selling Linux, and then failing at that, telling competitors they'll have to pay them to do it. It's about SCO first saying it wasn't about Linux just IBM, then it was about Linux, then it was about the GPL, and now the whole Open Source community is under attack by SCO. You see it as SCO being attacked by the Linux community?? Exactly who started this ruckus? If the code is illegally copied, let them show it to Linus and his team so they can replace the offending code rather than trying to extort obscene amounts of money out of Linux distributors and users for licenses. You really think SCO are the victims? They're not making any attempt to actually straighten out the situation, only to make money off it and only now that the company is in financial straits. True, some attacks against SCO may be unjustified, but that doesn't make them innocent or victims. If they were sincere about things being fair, they could end this anytime they want rather than playing the litigation game. They'd rather play it dirty, and now others are playing dirty as well. Too bad SCO wasn't prepared to be reasonable about it.
Is it? What I find interesting is this is both not an uncommon position and one that was derived not researched. As far as I can tell IBM started this ruckus. If they had bought the rights to UNIX or maintained the contract with SCO (and not implied that they had no rights) we wouldn't be here today. Right now they (SCO) are positioning for court, in that world you play to win and were I their litigation strategist I wouldn't recommend what you suggest. This is war now and it could have been avoided. (I have to admit some of the stuff they've done will work against them).
IBM started this ruckus IF and only IF they actually did anything wrong. This has yet to be proven in any way, shape, or form. Your bias is showing. If SCO simply filed suit and shut their mouthes, I don't thinnk there'd be any real issue here. It's not SCOs lawsuit I think most people are upset about, it's their actions since filing it. If they had just said "It will all come out in court" whenever they were asked about anything, instead of putting out 100 press releases a week ( and being proud of it to boot) there would only be 1-sided speculation that wouldn't last long without any fuel to fire it.
I am a lawyer and if I was their legal strategist and I thought our case was as solid as you seem to think it is, I'd tell them to file the suit and shut up. Go to court, get your 3bln and then go after others with court proven rights. If your case is weak, you do a lot of handwaving and, in SCOs case, you make a ton of false claims, boost your stock price and hopefully get out without an SEC investigation.
The history of Unix code, the fact that SCO as Caldera was a linux company and had a large number of programmers contributing to Linux, and the fact that works SCO claims IBM derived from unix could just as easily have been derived from linux leaves SCO is a very tenuous position. Proof enough to convince a judge of anything under these confusing circumstances is going to be a very difficult process- at best. While anything can happen in this legal system, I put their chances below 20%.
> As far as I can tell IBM started this ruckus. If
> they had bought the rights to UNIX
IBM did buy irrevocable, fully paid-up, perpetual rights to use and distribute UNIX <http://www.sco.com/scosource/ExhibitD.qxd.pdf>. Why should they have bought more than that? Why buy the cow when you already have the milk?
> or maintained the contract with SCO (and not
> implied that they had no rights) we wouldn't be
> here today.
There's been no evidence presented that IBM did not maintain it's contract with SCO, only SCO's allegations that they did not. IBM didn't say or imply that SCO had no rights, only that SCO didn't have rights to control what IBM does with code developed by IBM (according to <http://www.sco.com/scosource/ExhibitC.qxd.pdf>, SCO doesn't), or the right to revoke an irrevocable license.
What are you talking about? IBM DID buy the rights to Unix, they bought them from AT&T - they just didn't buy exclusive rights to Unix. The contract with SCO hasn't been broken either. I think it's irresponsible for you to even state that SCO is "playing to win" - they're attempting to be bought out. If you remember the DrDOS case that SCO (then Caldera) had against Microsoft, there was no media circus around it. SCO (then Caldera) wasn't releasing PR pieces weekly about the case. It wasn't on their page. Like any intelligent plantiff, they were silent.
The difference is that SCO (then Caldera) had a rock solid case against Microsoft, and knew any thoughtless and wreckless blabbing to the press could only hurt their case. SCO (then Caldera) did court the press on ONE ocassion with regard to the DrDOS case: they proved to the public they were right. The AARD code was published in DrDobbs journal, clearly demonstrating that Microsoft used some fairly underhanded tactics to try to keep DrDOSS out of the marketplace. Want a link?
Why is SCO acting so differently now?
I'll tell you why:
This time, SCO has an extremely weak case, so weak that McBride himself indicated a buyout would end all of this. That's a mark of desperation not confidence, when you KNOW you're going to win, there isn't any possibility of a settlement. They've changed their stories so many times, I have actually lost count. This isn't what you do when you expect to win in court. It's what you do when an old tactic didn't work and you desperately hope that you will be purchased to be shut up.
Don't you realize that SCO was brought to court and then shut down by LinuxTag in Germany? They were just fined 10,000 euros in Germany for essentially breaking a gag order. I ask you, if SCO has such a rock solid case, how did that happen? Do you have any thoughts on that?
SCO is riding a tiger now, and I bet McBride, Sontag, Stowell and crew wish they could get off. If you knew the history of Unix you'd be laughing with us, not at us. I can't even envision what IBM is going to do for trying to extort money out of them so publicly, but I imagine it will be interesting to watch. I have no delusions that IBM is some nice happy teddy bear type company, they're the scariest meanest company that exists on the planet today. If you think of SCO as a street thug, think of IBM as the Mafia godfather and you can understand the situation better. Even if IBM was guilty, they aren't going to lose an IP infringement case. They'd file patent violations on every single Canopy company before they allowed that.
Rob, Im surprised the issue of SCOs own behavior pre-lawsuit hasnt been raised...even putting aside the GPL. For a company to be selling a product it specializes in (whether it be operating systems or shaving cream) to claim they didnt know what was in the product they sold is legally a non-starter.
Caldera sold its own linux product with the supposed violating source code for almost two years after it purchased the rights to system v unix.
If you accept the GPL argument... They sold that product under a license they freely accepted for almost two years while they owned system v. For a company to have that level of expertise(according to their own claims), and not know is gross negligence or it is deliberate. Considering Calderas own attempt to "unify unix and linux", any claim they didnt know what was going on...seems to fail the "giggle test".
If so much code was "stolen" a simple text comparison of the two source codes, would suffice. You wouldnt need expert computer teams. The expert computer teams baloney is designed to "dazzle with bs" these non technically adept analysts.
If Caldera had given away someone elses linux for free then I might agree with you. However they themselves were contributors to the source code and selling the product as well.
In the end SCO as seller of both the violating product and the competing product so they must meet a higher standard of behavior than usual to make a claim of innocence.
They had a reasonable duty to their customers and their partners(such as unitedlinux) to ensure the products they sold didnt violate IP (including their own), they failed in that duty.