SCO vs Linux.. Continued

An anonymous reader writes "ComputerWorld has an interview with Chris Sontag, from SCO. Now the story has a pretty face." The interview has a variety of comments worth noting like how much source code SCO thinks has slipped from unix to linux. This story continues to amaze me.

Comments

Last 2 questions

[Waab]

I find the last two questions and answers to be particularly interesting.

Do you intend at any point to begin offering licenses to Linux users? We would hope as quickly as possible to develop solutions with the industry to allow customers to move forward with whatever platforms they wanted to choose, so long as the appropriate intellectual property foundation is in place.

Why didn't you act earlier? ... The point is we're really only recently seeing significant moves by many players, specifically IBM, to come out and state that they are moving wholesale to Linux.

So they'd be generous enough to sell Linux licenses and they didn't realize there might be a problem until a really big company started backing the competition.

Would it be ok to spell it $CO from now on, especially since they seem to be in bed with M$?

Re:Last 2 questions

[PetiePooo]

I'm not arguing the point that they should have immediately stopped distribution once the alleged violation was discovered. That delay could be a case of the left foot not knowing what the right foot is doing. Or the order to bring a lawsuit being handled before the order to stop their own distribution. You'll find that the courts tend to give a bit of leeway when it comes to timing issues like that. Hence my statement:

Whether they stopped their distribution soon enough after discovery of the violation is a matter for the courts to rule on.

You'll notice that I didn't mention the timing of the withdrawal of their distro in my recap of the fallacy. I think we both agree that,
IF their IP claim is valid, and
IF someone leaked that IP into the kernel source tree, and
IF they were unaware of it,
THEN the source tree contributer was not authorized to license the code, and
THEREFORE the GPL is not valid for that portion.

Elsewhere there are indications that they should have known of the alleged violations for some time. They themselves were contributors..

Re:Last 2 questions

[Wavicle]

Would it be ok to spell it $CO from now on, especially since they seem to be in bed with M$?

In bed with M$?? Let's look at what the article says:

What do you see as a company's options in the face of your warning? I would suspend any new Linux-related activities until this is all sorted out.

This is not in bed. This is in a dirty bathroom stall, in a seedy part of town, with one party on his or her knees.

what?

[pebs]

It is many different sections of code ranging from five to 10 to 15 lines of code in multiple places that are of issue

10 to 15 lines of code? That's such a small amount that similar code could be entirely coincidental.

Re:what?

[RealAlaskan]

The actual quote:

How many lines of code in the Linux kernel are a direct copyright violation? It's very extensive. It is many different sections of code ranging from five to 10 to 15 lines of code in multiple places that are of issue, up to large blocks of code that have been inappropriately copied into Linux in violation of our source-code licensing contract. That's in the kernel itself, so it is significant. It is not a line or two here or there. It was quite a surprise for us.

So, there are also some big blocks. What do you want to bet that those big blocks are the things which have been copied from BSD? What do you want to bet that they match up to SCO's stuff because the unix code that SCO bought the rights to sublicense has in it the BSD code which AT&T illegally copied?

In other words, anything in Linux which ``belongs to SCO'' has probably actually been copied, perfectly legitimately, from BSD. And of course, anything BSD is safe from SCO, whether SCO has the copyrights or not. At worst, Linux will have to incorporate the BSD advertising. More likely, the advertising clause was removed before the copying was done.

I suspect that IBM knows this. It would help explain their lack of panic.

Re:And the drama continues

[haraldm]

Well if the stock drops low enough every Linuxer is going to buy a LOT, performing a hostile takeover, ultimately :-))

Re:And the drama continues

[StealthBadger]

I don't think SCO cares if anyone believes them or not. The only thing that matters is whether or not a judge finds their legal argument sound enough to award them a judgement. Anything else they say is just smoke and mirrors, a nod in the general direction of Publik Opeenion.

The only way the price will go down and stay down is if someone leaks the so-called evidence (or some juicy internal memos...) and the Emperor is discovered to be wanderin' around buck nekkid, so to speak.

I'm just wondering why it hasn't happened yet.

code revealed in open court?

[jfruhlinger]

A question: If this ever gets to court, will SCO have to reveal its proprietary code in open court in order to prove that Linux has ripped it off? If so, won't that just disseminate their code further ?

jf

Re:code revealed in open court?

[sugarbomb]

Unlikely. I've worked on a few IP litigations, and a great deal of effort is put in to keep secret documents out of the public view. When the documents are discussed in court, the doors are closed, and anyone who is not allowed to see the info is removed from the courtroom. The exhibits the court uses are never released to the public, and even the transcripts are never released to the public.

If you are patient enough, a good observer can figure out what was discussed in closed door sessions

Sucks

[smittyoneeach]

to see the legal system used as a denial of service attack on the entire economy.
I hope that there is some actual basis for the claim, because otherwise SCO is just an MS meat-puppet.
Fighting to keep an open mind on the subject.

What this means

[Ender Ryan]

What this means, everyone, is that they knew of the "offending code" for a while without taking any action, and continued in their Linux business, knowingly distributing the offending code themselves.

They have no legal legs to stand on. Hopefully the courts will get this one right, and SCO will become irrelevent.

Re:What this means

[tomhudson]

SCO will become irrelevent.</quote>

Too late for that, it's already happened. Even the suits are now getting it, what with SCO's stock down by a more than a third since yesterday's peak.

I agree. Hopefully, the courts will chop off both their legs, and they'll be just another dot-com-bomb bumming around (pun intended). This, of course, requires them to survive long enough to get to court, which is becoming doubtful.

Re:What this means

[jedidiah]

This is a little bit different from the usual case of patent/copyright neglect. IF they new that the Linux kernel was contaminated and continued to distribute it this can only mean one of two things:

1) They willfully distributed the associated property under the GPL.

2) They willfully violated the GPL.

This would also means that SCO failed to take reasonable steps to inform the offending parties and allow them to make amends.

Litigation really should be the last resort.

Re:What this means

[EllisDees]

>It's true they have no legal leg to stand on, but unlike trademarks, patents and copyright do not diminish with disuse.

Doesn't matter. By distributing their own code under the GPL, they have given up the right to restrict the future use of that code. If they are trying to say that they didn't distribute it under the GPL, then every kernel contributor whose code they distributed can sure them into oblivion.

Re:What this means

[Surak]

How could they NOT know?

From the article:
How many lines of code in the Linux kernel are a direct copyright violation? It's very extensive. It is many different sections of code ranging from five to 10 to 15 lines of code in multiple places that are of issue, up to large blocks of code that have been inappropriately copied into Linux in violation of our source-code licensing contract. That's in the kernel itself, so it is significant. It is not a line or two here or there. It was quite a surprise for us.

Extensively? 10-15 lines of code in multiple places and large blocks of code? How could they NOT know? They distributed a custom-patched kernel with (at least AFAIK) Caldera OpenLinux 2.2 and 2.4, and probably all later versions as well, and maybe some earlier versions. I still have COL manuals and CDs laying around somewhere in my boxes of stuff. ;)

No, you're right. They willfully distributed associated property under the GPL, and they have willfully violated the GPL. FSF and/or Linus should sue them for all their worth (which is more than you think, I think their market cap is only $75 mil or so). IBM should countersue, their work is in the Linux kernel too. I wouldn't be surprised if IBM did so, either.

No, really, I think SCO is fscked and has been fscked and this is their last dying gasp. They figure if they're gonna go out, they might as well go out in style, eh?

Re:What this means

[e2d2]

Wouldn't this mean that other's distributing Linux are also in violation?

Snip from the GPL (emphasis added):
7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.

The wording is typical legalese, vague and foreign to my human mind. But since there is a dispute over the ownership on the source and potential "trade secrets" would linux distributors be in violation ONLY if they have conditions imposed on them or would the mere allegation of such conditions warrant the ending of distributing Linux under the GPL?

All I know is this is a serious issue for Linux and just the accusations of source theft will cause havoc. IMO that is why SCO is remaining so vague about the accusations and not quick to point out the lines of supposed theft. They really have no reason not to release the specific code to the public because it supposedly already has been released in Linux. So what's the problem? If the suspected lines of code are available to the public then why must the recipients of the evidence sign NDAs?

Bullshit and lawyers, the American way.

Re:What this means

[kien]

No, really, I think SCO is fscked and has been fscked and this is their last dying gasp. They figure if they're gonna go out, they might as well go out in style, eh?

Well, I guess they have a different sense of style than I do...but your point is made. :)

Here's a question (and mod me Redundant if someone's already pointed this out...I'm in a hurry): Why can't they just point out the infringing source code in the kernel without revealing their own proprietary code that is being infringed upon? Are they afraid that we might reverse-engineer the genie that is already out of the bottle or is this all just more obfuscation about what is really a legal non-issue?

Also, anyone that has ever worked for a pretty large corporation is aware of the Executive Summary; that being, get the gist of your meaning across very quickly because your average executive has a shorter attention span than a four-year-old with Attention Deficit Disorder. It is interesting to me that the lead answer in this interview is: Think about if I was the CIO of a company and I'm going to be running my business on an operating system that has an intellectual property foundation that, by almost everyone's admission, is built on quicksand. There is no mechanism in Linux to ensure [the legality of] that intellectual property of the source code being contributed by various people.

At this point, I was hearing the Monty Python song with the

s/spam/FUD

modifier applied.

--K.

Re:What this means

[ktakki]

...what with SCO's stock down by a more than a third since yesterday's peak.

But at $6/share that's still three times what it was trading for in mid-March of this year and ten times its July '02 price.

k.

Re:What this means

[praedor]

Could this then be a means of ATTEMPTING to screw the GPL? Think about it. M$ gives them money, they knowingly distributed their code under the umbrella of GPL, now they are going after linux - perhaps to try to test, and defeat, GPL.

I'm not saying they will invalidate GPL, but this is the first court case that will likely directly involve the GPL...

Re:What this means

[Poeir]

Just 10 to 15? Let's see...

void quickSort(int array[],int start,int arraySize) //1
{
int hold=0,i=start,j=end; //2
while(i<j) //3
{
for(i=start+1;(i<end) && (array[i]<=array[start]);i++) //4
for(j=end-1;(j>start) && (array[j]>=array[start]);j--) //5
if(i<j) //6
{
hold=array[i]; //7
array[i]=array[j]; //8
array[j]=hold; //9
}
}
hold=array[i]; //10
array[i]=array[start]; //11
array[start]=hold; //12
if(start!=end) //13
{
quickSort(array,start,i); //14
quickSort(array,i+1,end); //15
}
} //Unverified

Okay, give me my billion dollars now. Note that I have no idea if Quicksort makes any appearance in the kernel--I doubt it does--but there is any number of standard-issue algorithms that are 10 lines or more. Also, I know this implementation is kind of sloppy and may be incorrect, I just wanted to see if a perfectly ordinary algorithm would be "10 to 15 lines." All the variables also seem obvious as well.

Re:What this means

[schon]

Just what is SCO alleging anyways?

Is it patent, copyright, or trade secret infringement?

That depends; what day is it?

Re:What this means

[Black Copter Control]

But at $6/share that's still three times what it was trading for in mid-March of this year and ten times its July '02 price.

Well, if their claims to having a solid gripe against Linux don't pan out, and/or people in the Linux community start suing them for Libel and slander, I can see their stock ending up under the $1.00 mark pretty quick.

From what I can see, to prove their case against IBM, they'd not only have to prove that Linux has SYSV code. They'd also have to prove that it's IBM that released the code, and not some third party. If IBM is responsible for a third-party releasing their code, then SCO is on a very slippery slope.

By induction:

• If IBM is responsible for third-party release of source code then any company with UNIX source access is responsible for (accidently) re-releasing the source code when they distribute Linux source code
• If any company is responsible for third-party release, then SCO (as a distributer of Linux and the UNIX license holder) is similarly responsible.
• If SCO is responsible for releasing UNIX source code under Linux, then the company with the right to re-license that code is responsible for (re-)releasing it under the GPL.
• If that's the case, then -- under the GPL -- IBM (and the rest of the universe) has the right to do what they've been doing.

QED.

Re:What this means

[meatpopcicle]

what is this?

SCO: we know "Linux" is guilty of code theft, but since the Linux source code is available we will let them figure out where the offending code is even thought they "technically" dont have access to the code that they are in violation of copying?

How is Linux supposed to prove their innocence with this kind of statement?

This whole thing reeks of a scare tactic by a company that is losing profit.

5-10-15 lines of code in blocks are in violation? Well tell me that a 1000 programmers all programming code couldn't come up with some code that looked like proprietary code!

How do you know you are in violation of copyright when it comes to code writing when you dont have access to it. Should you be liable?

If I wrote some code that was GPL and it looked like copyrighted code would I be in violation even though I technically never saw the other code?

How do you protect yourself from that kind of lawsuit. This whole thing of protecting code like this is ludicrous!!!

What a load of feces....

[NerveGas]

• Think about if I was the CIO of a company and I'm going to be running my business on an operating system that has an intellectual property foundation that, by almost everyone's admission, is built on quicksand
  
  What a load of crap. He's essentially saying that closed-source code is somehow more guaranteed to be more legitimate. I'd say that the reverse is true: There's a lot more incentive to do things legally when the entire world gets to see your source code than when virtually nobody does.
  
  
• I would suspend any new Linux-related activities until this is all sorted out.
  
  How unfortunate that he left out the "... and buy SCO instead."
  
  
• We certainly have a point of contention regarding their interpretation of that contract.
  
  Earlier he said that companies should abandon linux-related projects until SCO's suit is worked out. Now, he's implying that despite the fact that SCO is lying out of their teeth, and that Novell is calling them on it, noooooobody should abandon SCO-related projects. Sounds a bit hypocritical.
  

Here's what it really boils down to: SCO is the armpit of the Unix world, and the headquarters are in a city that's virtually the center of the MLM universe, and also known for immensely brain-dead business executives. SCO sucks, and is going downhill. Why? Because of Linux. Amazingly, Linux is also eating into Microsoft's server market share. Now the two team up, and decide to try and get rid of Linux. It's really not hard to see what's going on.

steve

Re:What a load of feces....

[AlecC]

"Think about if I was the CIO of a company and I'm going to be running my business on an operating system that has an intellectual property foundation that, by almost everyone's admission, is built on quicksand"

What a load of crap. He's essentially saying that closed-source code is somehow more guaranteed to be more legitimate. I'd say that the reverse is true: There's a lot more incentive to do things legally when the entire world gets to see your source code than when virtually nobody does.

You're right. With open source, you have the chance to look at the code and see if bits of it at least appear to have been pirated (e.g. changes in programming styl;e, naming conventions etc.). With propriatary, you have no idea until the person claiming theft has forced your supplier (who they claim has ripped them off) to show the souce (of the program you need top keep using) to a court appointed expert - at which point all legal hell may let loose.

And companies have a motive to plagiarise code, because they can sell it. GPL'ers don't have a financial incentive to do so - and a strong personal incentive not to. A Linux contributor who would be shown to have ripped off SCO (or other) code would be a mockery. Geek cred is valued precisely because it is visibl earned - "I wrote that - judge me by it".

He says there is "only" an honor system (and the law) to stop SCO code getting into Linux. But what comparably system is there in commercial enterprises? So GPL has law PLUS honor, commercial has law MINUS financial incentives.

So commercial software is better/safer how?

Re:And the drama continues

[tuffy]

Well if the stock drops low enough every Linuxer is going to buy a LOT, performing a hostile takeover, ultimately :-))

But then the price will start going UP again. And besides, I already have an ample supply of toilet paper...

SCO's own admission that Novell owns UNIX System V

[Bull999999]

From http://perens.com/Articles/SCO/SCO_10-K.html

The Company has an arrangement with Novell, Inc. ("Novell") in which it acts as an administrative agent in the collection of royalties for customers who deploy SVRx technology. Under the agency agreement, the Company collects all customer payments and remits 95 percent of the collected funds to Novell and retains 5 percent as an administrative fee. The Company records the 5 percent administrative fee as revenue in its consolidated statements of operations. The accompanying October 31, 2002 and 2001 consolidated balance sheets reflect the amounts collected related to this agency agreement but not yet remitted to Novell of $1,428,000 and $1,894,000, respectively, as restricted cash and royalty payable to Novell. The October 31, 2001 balances were reclassified from cash and equivalents and other royalties payable to conform to the current year presentation.

This is SCO's admission that Novell owns Unix System V, all revisions - that's what they mean by "SVRx", and SCO pays Novell 95% of the royalties. SCO gets to keep 5% as administrative agent.

Re:could this be....

[pebs]

and look! if you rearrange microsoft a bit you can get sco rimoft

Interesting, and to take it a step further, you rearrange "microsoft" and you get "sco from it"

SCO stock price

[egoff]

SCO dropped 25% yesterday, and another 10% today (so far)

LPA

[ch-chuck]

Installation Complete! To activate you Linux system, call the SCO Licensing clearinghouse at 888-WEG-OTCHA to obtain an activation key. Remember, if you change more than 3 system components, you will have to obtain a new activation key. Have a Nice Day!

Questions reporters aren't asking SCO

[jonabbey]

SCO is refusing to answer some elementary questions that are essential to put their claims into context. Of course, it suits their purpose to cast FUD on the OSS competitor that is destroying the value of their IP, but there's no reason why reporters should let them do it.

To wit:

Does SCO believe that Linux would be substantially less useful if the code claimed to be excerpted from SYSV were excised? Is the value of the allegedly stolen code significant to the overall value of the Linux system, or is it merely valuable to provide standing for SCO to discourage the use of a free competitor to SCO?

Is the claimed SCO code part of one or more optional components of the Linux kernel, or are they in the kernel's core?

Does the claimed SCO code relate in any way to compatibility with SCO disk partions, file systems, or binary compatibility?

How many lines of code are we talking about?

No, really, how many lines of code are we talking about?

Where is the logic in keeping the outside experts under NDA about what code is believed by SCO to have been copied into Linux? If the code is in the Linux kernel, by definition it cannot be an effective trade secret.. does this mean that the real reason for the proposed NDA is to ensure that Linux developers cannot remove the alleged SCO IP from Linux?

Why doesn't SCO wish for Linux developers to fix the problem, given that SCO has claimed that this is a case against IBM for contract violations?

Does SCO believe that their case for damages would be weakened if the alleged code was removed?

Why does SCO believe it is necessary to prevent Linux developers from fixing the problem, given that there are archives of years of development work on the Linux kernel and utilities. Would SCO consider allowing Linux developers to fix the alleged problem if SCO were given a copy of the entire Kernel development records before revealing this information?

DON'T CLICK THE LINK

[rjamestaylor]

It sends a nasty email to SCO with foul language. Before you think "cool!" realize that you may be charged with harrassment via telecommunication lines, which I seem to recall is a federal offense (in the US). But, more crucially, such emails will be touted by SCO as indicative of the "Linux/Open Source" community and will be used to turn the hearts and minds of the average people away from "our side". They're praying for more idiocy as embodied in this link.

Be respectful in the face of their disrespect. Be honorable in the face of their dishonorable acts. Take the high ground and watch them collapse on themselves.

This story continues to amaze me.

[Znonymous Coward]

...This story continues to amaze me...

What continues to amaze me is the following...

Netcraft reports that SCO's own website is running on Linux.

SCO is still apart of UnitedLinux

SCO's own phone number is 1-888-GO-LINUX

They sure have their hands in a lot of Linux for it being so "illegal".

Re:This story continues to amaze me.

[quantaman]

SCO's own phone number is 1-888-GO-LINUX

It still works, they've just changed their minds about where they want linux to go.

Re:And the drama continues

[bbum]

You are looking at the wrong stock. SCO is not SCO/Caldera. SCO -- SCOR group, traded on NYSE (first clue that it wasn't the right symbol) -- isn't doing well, either, but that is coincidental (unless there are a bunch of stupid day traders that can't tell their symbols apart).

Caldera is SCOX (traded on NASDAQ) -- yahoo summary.

While they are down a boatload in the last couple of days, they are still not below the levels they were trading at the beginning of May (or earlier).

As it stands, it looks like this little publicity stunt has driven the stock up more than down. Given current trends, it looks like it may normalize such that the whole thing is a short term wash in terms of stock value-- whether it proves profitable over the long term remains to be seen.

SCO pays Novell Royalties for SVRx

[zoid.com]

This is from their 10-K filing in Jan. Why would they pay if they owned it?

"Restricted Cash and Royalty Payable to Novell, Inc.

The Company has an arrangement with Novell, Inc. ("Novell") in which it acts as an administrative agent in the collection of royalties for customers who deploy SVRx technology. Under the agency agreement, the Company collects all customer payments and remits 95 percent of the collected funds to Novell and retains 5 percent as an administrative fee. The Company records the 5 percent administrative fee as revenue in its consolidated statements of operations. The accompanying October 31, 2002 and 2001 consolidated balance sheets reflect the amounts collected related to this agency agreement but not yet remitted to Novell of $1,428,000 and $1,894,000, respectively, as restricted cash and royalty payable to Novell. The October 31, 2001 balances were reclassified from cash and equivalents and other royalties payable to conform to the current year presentation."

Re:And the drama continues

[Horny Smurf]

Actually, MS does sell a "Unix Services For Windows" package for 2k/xp. It's the formerly interix stuff.

It's a posix layer and standard utilities (most of the BSD based). It does include gcc, though (and the source code).

It's been available (from interix or MS) since NT4, though, and I doubt it contains any SCO code, and (as I said before), the command-line tools are mostly just recompiles, so it seems unlikely that's why they would get a Unix license (OTOH, suing microsoft is popular, and juries do stupid things sometimes).

SCO sues Linux for contractual violation w/ IBM

[usurper_ii]

SCO has now said it isn't an IP issue or a copyright issue, but a contractual issue. Since Linus had no contract with SCO, how could they sue him for an alleged contractual violation that happened between SCO and IBM?

And the same goes for anything IBM may have leaked, and note I'm not saying they did...but if they did break a contract, how can anyone using a Linux product using such code be held liable for a contractual violation done by IBM, again, when SCO has now said it is contractual issue and not an IP issue or a copyright issue.

On one hand I guess we can be glad SCO are such morons, but on the other hand, can you imagine releasing a press release saying the issue was never about IP or copyrights when they are running around screaming about suing everyone because Linux may have some of their IP in it!!!

Go that way really fast, if something gets in your way, turn

Re:And the drama continues

[haystor]

Market Cap $72.6M.

That's 12 million shares at $6 each.

That's 121 thousand people each spending $600 for 100 shares.

Are there 121 thousand people that would consider it humurous enough to buy the company just to fire fire everyone? They probably all have parachutes set up to rob the company should they be fired and then the company would have to die. Think of this like the blender project only a hell of a lot bigger.

Re: $CO

[JahToasted]

No it would not be appropriate to call them "$CO" since they will be bankrupt very soon.

Re:And the drama continues

[spongman]

This chump actually expects us to believe that Microsoft has been working on more UNIX interoperability for a year or more (yea, right)

Microsoft has been shipping Services for Unix since 2000. It recently won the Open Source Product Excellence Award for Best System Integration Software at LinuxWorld.

Note to Editors:

[misfit13b]

"We will actually be providing some of the evidence next month to various industry analysts, respected press people and other industry leaders so that they don't have to take our word for it or wait until we show some of that evidence in court."

Until this happens, there really isn't too much of a story here. Wake me up in a month.

Just for publicity...

[SharpFang]

Admit this: Who of you purchased any SCO product over the last years? How many of you didn't hear about SCO before, or just heard the name without associating it with anything? What share of the computer market does SCO control?

SCO is forgotten.

So, what's the best way to get out of shadow and stand in spotlight? Oh well, miss Lewinsky showed that to all of us.

1. Make a lot of noise around something famous.
2. Gain fame.
3. Sell products, make claims.
4. PROFIT

The best target would be something as big as M$, but SCO had several reasons not to attack it (including M$ lawyers). So, the next target on the OS market seems obvious...

Why else would SCO care for 15 lines of code, whey would it make so dubious claim, than just to gain publicity? "No matter, good or bad, it's important that they talk about you". Old rule of showbusiness, may apply here too...

I guess the end will be quite mundane. Maybe putting a notice in sources "This part created by SCO". Maybe rewriting that parts of kernel. Maybe the charges will be dismissed. Maybe "SCO will bend under customers' pressure and withdraw its claims". What is important, is that people will talk about SCO over next few years, and whoever plans some new investment, will think "...And maybe consider that SCO thing..."?

What about BSD/USL settlement?

[siskbc]

SCO is refusing to answer some elementary questions that are essential to put their claims into context. Of course, it suits their purpose to cast FUD on the OSS competitor that is destroying the value of their IP, but there's no reason why reporters should let them do it.

To that I would add this: Given the USL/Novell settlement in 1994, BSD was given the right to distribute and license their unix product, which, at the time, was pretty much Sys V if I recall correctly. Given that, can you prove that any of the offending code was written by owners of the historical Sys V code after the settlement?

I remember someone saying in an interview a long time ago that the offending code wasn't BSD code, but I'm having a hard time believing it, and I haven't heard anything of the sort for some time.

Anyone know details on the settlement, as to specifically what rights were granted BSD, and when Sys V developed the symmetric multiprocessor capabilities in question? Or has SCO really broadened its scope beyone SMP to general Sys V operation?

Re:This really is starting to smell like a M$ move

[Piquan]

I've said it before, but they told us to expect it. Remember Halloween VII?

If you don't, Halloween VII was a leaked memo from MS dated Sep 2002. It was a survey report, discussing what types of FUD were most effective, and where FUD was backfiring.

"Linux patent violations/risk of being sued" struck a chord with US and Swedish respondents. Seventy-four percent (74%) of Americans and 82% of Swedes stated that the risk of being sued over Linux patent violations made them feel less favorable towards Linux. This was the only message that had a strong impact with any audience.

And later:

Messages that rely on an abstract discussion of intellectual property rights are not effective.

The discussion of IP rights needs to be tied to concrete actions.

A GREAT detailed explanation SCO/Linux issue

[zoid.com]

This page offers the absolute best explanation about what is going on in the SCO vs. Linux issue. A definate must read!

Re:SMELLTHE FUD!

[jedidiah]

You CAN audit Free Software.

You can NOT audit proprietary source code.

We should repeat this frequently and loudly.

Should the IT pundits come to this realization, all of this gibberish about Open Source "accountability" will backfire horribly.

Even if what this SCO monkey claims is true is true, that still leaves the question of ultimate ownership of the offending code. Linux is "out there". The code and development process is publically documented and mirrored.

If SCO claims ownership to 15 lines of the kernel, how do they independently verify that claim? There are simply no 3rd parties with possession of all the relevant evidence.

It's not just the source itself but also the entire change history.

Merely pointing out common code between the two codebases simply isn't enough to base a conclusion on.

it tells the story of their stock price!

[Thud457]

$0

dollars, cents, nothing

The copied code revealed:

[hoggoth]

Sources at SCO have revealed some of the offending 10 or 15 lines of code, in no particular order:

main( int argc, char **argv)
return;
int i;
{
}
#include <stdio.h>
#include <stdlib.h>
char buffer[MAXBUF];
#define true 1
#define false 0

you be the judge

[bgs4]

here are the lines from linux:
}
}
}

and Unix System V:
}
}
}

Re:And the drama continues

[dAzED1]

They have been working on this for at least four years, if not more.

NO

Good god, its so damn easy to get mod5 as "informative." Whatever. Anyway, the compat layer is for things that are almost completely owned by Solaris in the IP world...nfs, automounter, etc. Those are COMPLETELY Sun, and not even SCO is making claim to them. Think M$ gave money to Sun? Hell no - Sun is a linux ally. They're not trying to discredit Linux. Serves M$ no purpose.

Or, so that you can become somewhat informed on your own, go to M$ all on your own. How about checking out the MS "solution" itself, Services For Unix.

Here's what MS says about it: "Services for UNIX provides file sharing, remote access and administration, password synchronization, common directory management, a common set of utilities, and a shell."

File sharing is through a samba-like util and through nfs, password sync is through ldap-like stuff (as is the directory management), and the utilities have various "uses." Not one of these things has anything to do with SCO.

Of course, if SCO would just mention what is being infringed, that might help the clueless be less confused. Many of us know its bunk without their even bothering, though.

Re:And the drama continues

[TKinias]

scripsit spongman:

Microsoft has been shipping Services for Unix since 2000. It recently won the Open Source Product Excellence Award for Best System Integration Software at LinuxWorld.

...and Teddy Roosevelt once won a Nobel Peace Prize.

It's a strange world we live in.

IP compliance of source code

[Idou]

"There is no mechanism in Linux to ensure [the legality of] that intellectual property of the source code being contributed by various people."

I agree, but how is this any different from proprietary software? How CAN MS ensure that its code doesn't contain any SCO code, unless they license to have direct access to that code?

The only difference I can see is that with closed source code, there is NO WAY for ANYONE (even the owner) to make sure there are no IP violations. With open source code, only the owners of the closed source code are able to ensure their IP is protected (burden falls on the owners).

If one is really concerned about IP, one would require all code that has IP protection "Open Source", that way EVERYONE could verify whether or not a specific part of code is a copy of some other code.

However, it is my opinion that, under the current circumstances, making one's own code "Open Source" is the most one can do to ensure that they have performed "due dilligence" in ensuring that their code is free of IP law violation. Closed source seems to be the model lacking in this area, not Open Source.

Re:Question about their threat to sue Linus Torval

[King_TJ]

I think this coincides with the other comment made by SCO's CEO that they'd go after the infringements all the way to the individual users, if necessary. (By the same token, if they can prove Linus Torvalds somehow stole SCO kernel source and rolled parts of it into the Linux kernel, they'd sue him personally for that act.)

The whole thing sounds ridiculous to me, but as so often happens with software, sometimes there aren't a whole lot of different ways to accomplish a task. A piece of hardware only interfaces with its software drivers through specific commands, sent a specific way, for example. I don't doubt SCO could match up small code segments (or even "big blocks of code", depending on their definition of "big") with what's in Linux.... but it might just be because nobody would really write those routines any other way.

Microsoft DOES use Linux

[Door-opening Fascist]

[SCO gave Microsoft a license because] Microsoft is not using Linux.

Excuse me? Take a look at Microsoft's Netcraft page. The top three machines (UT servers) are running Linux, and are sponsored by MSN.

Re:And the drama continues

[Surak]

Yeah, let's /. SCO's stock! ;)

I know you're joking but...

[lpret]

What about it? What about everyone buy some shares and group together as shareholders to stop SCO from doing this? Then, donate your shares to a single Linux organisation, like Mandrake or Red Hat, and do away with all this madness. This would stop all the doubts people have about the OSS community not being united and only help it. Anyways, just a thought to save linux...

Trade Secrets

[clonebarkins]

I'm sure I just "don't get it" when it comes to trade secrets, but some things don't make sense to me. Obviously, IANAL (and I only sometimes play one on /.), but I believe I heard that if something that is considered a "trade secret" is developed independently by somebody else, there's really nothing you can do about it. That's the tradeoff of patenting your findings -- the information is public, but at least you have rights to it. If you keep it as a trade secret, then you have rights only so far as nobody else discovers/invents the same thing you did.

Now, having said that, obviously there is the IBM component. SCO claims that IBM violated trade agreemnts or NDAs or whatever, and that is how "SCO's code" (if indeed the code even belongs to them) was integrated into linux. But here is the kicker: Just because some lines of code are similar (or even the same) in two different pieces of software, it doesn't mean that the code for one was taken from the other! It seems that SCO not only has the burden of proof of identifying what code they allege is similar, but that they also need to prove that it was IBM (or someone who works at IBM) that actually inserted the code into linux (or at least provided it to Linus et al).

Furthermore, SCO would then need to prove that the code implemented in the linux kernel is 1) critical to the application and 2) actually covered by any patents as being both non-obvious and non-prior art. If some of the matching code is nothing more than an abstracted for loop that increments a counter variable and passes the result to a function or sets another variable (such as an array), then I can't image how any rational person could construe that as patent infringement. But then again, I'm not CEO of a failing company (Q2 earnings aside -- we all know posted earnings don't actually mean anything -- *cough*enron*cough*)

Finally, I like the idea of "whole blocks of code." Obviously his intent is to imply that massive portions of System Unix V code have been "violated," but what he didn't consider is that block has a very technical meaning -- a "whole block" could very easily be a one-line if statement. Not that impressive overall.

Re:Trade Secrets

[the eric conspiracy]

Furthermore, SCO would then need to prove that the code implemented in the linux kernel is 1) critical to the application and 2) actually covered by any patents as being both non-obvious and non-prior art.

You are getting a little mixed up here. A trade secret is something that is NOT patented for any of a variety of reasons. For example, a company could decide that a patent if granted would be unenforceable because it covered a process step that another company could practice in secret making it impossible for the patent holder to determine if infringement was occurring.

The kicker in all of this is the contract with IBM. We don't know what the terms are, and it may well be that it included terms protecting things beyond what is considered trade secret.

I have been involved in some of these things, and I will say that SCO's claims that developers who worked on the SCO project with IBM then later moved on to work on Linux will be dangerous in a court of law. It is very common to claim that knowledge obtained in the first case will inevetably leak into the second project, and courts can and do believe it.

In my case I was privy under NDA to a technology that my employer was considering purchasing from another company - after the decision was made not to go ahead with the purchase I was not allowed by my employer to ever do work in the technology area covered by the NDA for fear of exactly this problem. I also had to destroy all documents involved, and wasn't able to even tell my boss (the VP R&D) the technical reasons behind my decision to recommend against the purchase.

Let's guess the copied code

[jschrod]

10-15 lines? I know them!

* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

Hmpff, what did you write? It's code copied from SCO to Linux, not the other way round? Well, who knows... :-) :-)

BSD

[siskbc]

So, there are also some big blocks. What do you want to bet that those big blocks are the things which have been copied from BSD? What do you want to bet that they match up to SCO's stuff because the unix code that SCO bought the rights to sublicense has in it the BSD code which AT&T illegally copied?

I've been trying like crazy to figure out if this is the case, and if so, if SCO is shit out of luck. I remember a long time ago they said it wasn't the case, but their story might be changing.

What I want to know is whether:

1. The code was pre-1994 from BSD, but they somehow don't think the 1994 agreement is transferrable or even valid, or

2. The code in question was written by Novell or (God forbid) SCO after 1994. If written by Novell, did they explicitly give it to the OSS community or was it just that they didn't enforce the violation? If the latter, this could pose problems...or,

3. Did SCO illegally copy code from BSD (or elsewhere) post-1994? I will say, it will be very important to see source tree commit dates even if they do have some interesting code similarities

I wonder who the "independent experts" they show the source code will be? Probably not anyone with Novell, huh?

Re:BSD

[RealAlaskan]

What I want to know is whether:
# 1. The code was pre-1994 from BSD, but they somehow don't think the 1994 agreement is transferrable or even valid, or

If the code comes from BSD, then we're in the clear [1].

# 2. The code in question was written by Novell or (God forbid) SCO after 1994. If written by Novell, did they explicitly give it to the OSS community or was it just that they didn't enforce the violation? If the latter, this could pose problems...or,

In this case, there's a problem indeed. If Novell sold SCO an exclusive ``license to sublicense'', Novell may or may not be able to let us off the hook. If the code comes to Linux via BSD, as I postulated, we and IBM should be in the clear: we couldn't be expected to know that BSD had somehow stolen it from SCO [2]. If the code comes to Linux via IBM, IBM has a problem [3]. Linus and the kernel gang will have to scramble to clean out the offending stuff in either event.

# 3. Did SCO illegally copy code from BSD (or elsewhere) post-1994? I will say, it will be very important to see source tree commit dates even if they do have some interesting code similarities

This seems to be the most likely scenario. In fact, it seems very probable that they realized, some time back, that they could:

1) copy some good parts from Linux into SCO server/Xenix/whatever
2) claim that Linux copied it from them,
3) sue the deepest pockets around for using ``their'' code
4) PROFIT!

The scary thing is that this is a better business plan than a lot of the venture capitalists' favorite companies had two years ago.

[1] Unless BSD somehow got it from SCO or whoever, as in your point 2. It's more likely that SCO got it from BSD, as I propose in my four steps to (fraudulent) profit, above.

[2] I shouldn't have to say that this is ridiculously unlikely. Should I?

[3] This seems ridiculously unlikely, too.

Lawyers and Juries and Source Code

[Dictator For Life]

I have actually been involved in a lawsuit that involved claims over source code (I was one of two principal coders of the app in question).

In that case, we were accused (in part) of stealing code from an application that was used in the same industry but which not only looked drastically different (hence we could not have even "stolen" look-and-feel), and not only lacked substantial functionality in comparison to our app, but was also first released after our app was in production.

What happened?

We "lost", simply because my employer ran out of money to fight what was unquestionably a preposterous and baseless suit.

But let's look at SCO's claims about "copied code" from the viewpoint of lawyers and likely jurors. They are *not* going to understand the intricacies of kernel code. They are not going to get it when anyone says "Well, the code is the same because it does the same thing." I know this is true, of course, and it's perfectly reasonable: but a jury will try to wrap its heads around this problem by comparing it to things that they *do* understand. So they will compare it to copying books, or movies, or poetry, or something.

Now, if you or I saw a paragraph in a John Grisham book that was identical to a paragraph in a Michael Crichton book, what would we conclude? We would conclude that the paragraph was "obviously" copied.

Given the types of juries that lawyers like to find for themselves (namely, "drooling idiots", all too often), what are the odds that a jury in the United States will really care about learning or understanding the intricacies of programming? What are the odds that they will understand that it's entirely possible for source code to look the same in places when it performs the same function - even if it's written by two different people?

Personally, I wonder whether it might not be better for SCO to be crushed long before this ever gets to a trial. Juries in this country simply cannot be trusted.

How small can IP be?

[Sean80]

ranging from five to 10 to 15 lines of code

What fascinates me is how much intellectual property can you fit into 10 or 15 lines of code? There are only so many ways to structure data in the world, so many ways to allocate memory and so forth. How close does your code have to be to some other piece of code for it to infringe on intellectual property?

Sure, if Linux stole entire libraries of code, then that would be an issue. But how can you lay claim to component parts as small as this?

FSF foresight

[pmz]

With all the FUD and name-calling among SCO, IBM, Novell, Microsoft, etc. etc., I am realizing more and more the foresight of the FSF in establishing its requirements for copyright paperwork when submitting code (link to FSF docs).

It is important to realize that even if the Linux kernel itself is somehow victimized, the GNU tools and the GNU/HURD should be untouched. The BSDs, Mac OS X, and Solaris should fair very well, too, if only because their legal problems are already dealt with. However, I really think SCO's claims against Linux are a long shot (of galactic proportion, unless, of course, they planted the code maliciously), so my hope is that SCO is the only true casualty once all this is over.

I think it's a time issue.

[RabidChipmunk]

Yeah, there's still the "time for FUD" issue. However, for SCO's legal case, the "time to document refutations" could be much more important i.e. SCO doesn't want to give the Linux comuntity time to get their ducks in a row.

SCO knows that the authorship of Linux is much messier than it would be at a traditional company. By making a shotgun claim to many parts of the kernel they can win if any one peice hits. If one author of one peice can't be found, they can win the suit against IBM. If they tell us now what peices they are we can start scouring the globe. If they don't tell us until the legal procedings begin, it becomes a race between the legal procedure and the Linux comunity. Like a life or death game of seek and find. Better, if they can get a judge to only let IBM see the code, it becomes a seek and find where only IBM can play and they can't tell us what they are looking for. They couldn't even say "Does anyone have Linus's email address?" [Or more likely, "does anyone know who wrote lines 1047 to 1052 of kernelfile.c?"]

If they told us what lines were in question, we could all write memiors about how those lines came to be, with CVS snapshots and mailinglist discussions to back it up. If they don't tell us we can either do nothing and be unprepared, or start documenting everything and not get any real work done.

It looks to me like they are testing if the Linux comunity is able to generate a coherent document trail faster than they can generate code. We have lots of data. Can we seperate the wheat from the chaff on demand?

Re:what? That's some pretty expensive code!

[AnglerG]

Let me see, they're suing IBM for 1 billion $.

That comes out to 60-100 million dollars per line of code.

I think we've just found the most expensive ASCII text in the world.

bk annotate?

[Kiaser Zohsay]

With chunks as small as 10-15 lines, it ought to be pretty easy to determine which lines come from which patch, and then from the patch determine the submitter. If the sections of code that SCO is bitching just happen to line up with particular patch submissions, then they might just be able to make a case.

However, more likely, if the code that they are claiming infringes was touched by multiple contributors over a long period of time, and the result of all those edits just happens to kinda sorta resemble a piece of code in SVR5 (aka independent invention), then they are going to have a much harder time making that stick.

Piquan responds to Sontag

[Piquan]

Why should Linux users take your claim seriously? Think about if I was the CIO of a company and I'm going to be running my business on an operating system that has an intellectual property foundation that, by almost everyone's admission, is built on quicksand. There is no mechanism in Linux to ensure [the legality of] that intellectual property of the source code being contributed by various people.

I am aware of no mechanism in SCO Unix to ensure the legality of IP. It is very uncommon for a company to require documentation to verify the legality of their code. In fact, the only company I'm aware of that does this is the Free Software Foundation, who ownes a plurality of the copyrights on the GNU/Linux operating system.

It is a gross exaggeration to say that "almost everyone" feels that Linux's IP foundation is built on quicksand. You are the only one who I have heard state such a belief, despite campaigning by other groups with an interest in discrediting Linux.

The development process has no one that is ensuring that inappropriate code is not getting into Linux.

Please indicate the person or persons at SCO who fills this task.

All that's there is an honor system, and obviously there are a few, at least, that have broken that honor.

In traditional closed-source operating systems, the users must believe the manufacturer's statement that the OS is free of IP entanglements. The open source community, at least, provides IP holders with the means to verify IP issues. Can the same be said for closed-source OSs?

There have been multiple occassions when closed-source software has illegally adopted code from open-source software. And yet, you seem to imply that this is a problem specific to Linux's open-source model.

Your letter to 1,500 end-user companies outlining your claim was vague. What is it that you want from these companies? The one thing that we specifically want from those 1,500 companies that we directly sent those letters to is for them to not take our word on the warning that we sent ... but to seek an opinion of their legal counsel as to the issues that we raised.

Your actions betray your words. You refuse to provide the user base with the information they need to evaluate the issues. Your complaint is vague, and provides no specifics with which the user community could evaluate its authenticity. Does SCO recommend that we stay away from Linux, based on vague claims? Would SCO be willing to pay for the additional costs incurred in a transition to an alternative, if your claim is found to be without merit?

Suppose I made the public claim that SCO had violated my copyrights, but refused to elaborate. Would you then expect all your customers to stop any new SCO-related deployments pending a resolution?

What do you see as a company's options in the face of your warning? I would suspend any new Linux-related activities until this is all sorted out. But first get that opinion of your legal counsel. If they say there is no problem and no issue, then you probably have nothing to worry about. But I doubt there is any attorney worth his salt that is going to say there is no potential of an issue here. There is a big issue.

There is also a potential that you have violated my copyright in creating your own software; it just doesn't seem like a likely thing. Such an action must be judged by an evalution of its merits. Since you do not provide any information by which we could judge the merits of your complaint, there is no reason to act on your accusation.

Should companies remove Linux from their systems? We're not making any specific recommendations at this time. We're still getting our arms around the size of this problem. We're still identifying more and more code from Unix System V that is in Linux, and so we h

liars

[dh003i]

You're claiming that Linux has been polluted with Unix code that you own, but you have not produced any evidence of that. Will you? We will actually be providing some of the evidence next month to various industry analysts, respected press people and other industry leaders so that they don't have to take our word for it or wait until we show some of that evidence in court. We will actually be showing the code, and the basis for why we have made the allegations that we have. We are very confident about our case. Because we are dealing with confidential source code that we have never released without confidentiality agreements, we will have to put in place nondisclosures [agreements] simply to protect the source. But people will be able to give their opinion as to what they think.

It's obvious that this sack of shit is lying. If his claims were true, he'd have no reason not to point out the offending code, since it has already been released to the public for all to see. There is no longer anything that he can protect by keeping the "offending code" secret. This is just a smokescreen for the fact that there really is no evidence because the entire claim is completely fabricated. See the OSI's response to this non-sense.

Also, there are many mechanisms ensuring that FOSS software is properly distributed without violating IP. People are required to sign waivers indicating that the contributed code was not improperly obtained. In many cases, corporations are asked to sign waivers.

Furthermore, since the code is FOSS, any proprietary entities concerned can easily identify any issues and have them resolved. SCO could have done this a long time ago -- it's obvious this is bullshit.

As a general summary, there are more insurances that FOSS isn't misappropriated than there are for proprietary code (which is closed-source, so they can hide misappropriations of IP). Furthermore, it is much more likely that SCO violated that GPL than that any FOSS developers violated SCO's IP.

IBM AIX 5L may be key to this.

[emil]

SCO and IBM worked together for some time on a version of AIX for Itanium. AFAIK, SCO contributed UNIXWare code, and IBM contributed AIX code.

IBM walked away from this agreement.

If IBM contributed anything from this collective codebase (either their own code or SCO's), then SCO's actions become entirely logical.

This may not be about historical UNIX code. This may be about recent development efforts and the sour relationship between IBM and SCO over Itanium UNIX.

Just in From CNET

[Anonymous Coward]

--From the Lone AC, Yippe KY--

SCO Group Chief Executive Darl McBride said a published report that his company may take legal action against Linux founder Linus Torvalds was overstated.
Responding to a portion of a Wednesday story by CBS Marketwatch that has generated intense criticism from the Linux community, McBride told CNET News.com that targeting Torvalds is unlikely.

"Virtually we see no reason why that would ever happen," McBride said. "We're not trying to go down that path."

McBride's comments were meant to address a portion of the story that stated, "McBride added that unless more companies start licensing SCO's property, he may also sue Linus Torvalds, who is credited with inventing the Linux operating system, for patent infringement."

While he would not completely rule out the possibility of suing Torvalds, McBride emphasized with News.com that "I wasn't even talking about patents."

A CBS Marketwatch reporter did not immediately respond to a request to respond to McBride's comments.

Torvalds, meanwhile, said he sees legal action against him as ineffectual but not inconceivable. "I don't see what (SCO) would expect to gain from suing me, but they don't seem to be acting very rationally," he wrote in an e-mail interview.

And while Torvalds said he agrees with some of the criticisms SCO's actions have triggered on Linux-friendly online forums such as Slashdot, he also called for restraint and maturity in dealing with SCO. "I hope this doesn't incite anybody to (launch a denial-of-service attack against) the SCO Web site or something silly like that," he said.

SCO's actions, including legal threats and assertions that Linux programmers couldn't have built high-end features into the operating system on their own, have indeed inflamed the passions of many Linux advocates. SCO's Web site was crushed by such an attack earlier in May; the specific attackers were unknown, but SCO was quick to blame Linux proponents.

SCO sued IBM for more than $1 billion in March, alleging Big Blue illegally incorporated Unix intellectual property that is owned by SCO into Linux. Initially, SCO said it wasn't going after Linux, but it changed its stance when three separate investigations found Unix source code copied into Linux, the company said. SCO has declined to reveal the specific code that was allegedly copied.

The copying of source code could potentially expand SCO's legal actions beyond IBM through copyright infringement claims, but McBride said contracts provide a stronger legal case.

"Our code is showing up inside the Linux kernel. Given the rights we have, where does that take us? The most logical place is the guys we have contracts with," McBride said.

Novell, which owned Unix rights before selling at least some of them to SCO's predecessor in 1995, on Wednesday disputed that SCO has copyrights and patents for Unix. It didn't dispute that SCO holds the contracts under which Unix is licensed to others.

SCO said it will reveal in June the Unix code has been copied into Linux, but only to select people, such as independent analysts who have signed nondisclosure agreements. It won't share that code publicly, saying the Unix code is proprietary.

SCO says it has more than 6,000 Unix licensees, including companies and universities, and that its direct contracts with companies such as Hewlett-Packard, IBM or SGI require that sublicensees protect the Unix code. A sublicensee is a business that has purchased hardware or software with Linux from IBM, for example.

"They sign up for the fact that they may not misappropriate the code," McBride said. Unix is used at the majority of the 1,500 large companies to which SCO sent letters alerting them to legal risks of using Linux, he said.

Contrary to Novell's assertion, SCO contends it does have Unix copyrights and could base legal action on them. "I think it's perfectly clear we have the rights to enforce copyright claims," McBride said. "Clearly copyright is a path you can be taking a hard look at."

Re:Just in From CNET

[NortWind]

SCO said it will reveal in June the Unix code has been copied into Linux, but only to select people, such as independent analysts who have signed nondisclosure agreements. It won't share that code publicly, saying the Unix code is proprietary.

If the code is in Linux, as they claim, then the code is available to the public, and they no reason I can think of to share the infomation about what code they are disputing.

Re:Just in From CNET

[judmarc]

"I'm betting they want the offending code to keep getting used so that they can prosecute the use."

Trouble is, you have a legal obligation to mitigate damages, or in English: You don't get reimbursed in a court of law for harm you could have prevented.

Revealing the code won't do anything to cure any harm to SCO that has already taken place due to the allegedly misappropriation, and SCO can get damages for that if they can prove the improper use of their intellectual property.

So - no damages for future avoidable harm *whether or not SCO reveals the code*. Think about that. It means the reason they're giving for failing to identify the code publicly is a lie. Like CmdrTaco says, the more you hear about this case, the less sense it makes.

Re:Just in From CNET

[budgenator]

We had no way of knowing we were infringing because, after all, the code we were infringing is a secret."

Exactly, any project that has the same inputs, the same outputs and however many million lines of code is going to have code that is exactly the same as an other project that is equivalant, accounting for variable name being different, so when SCO's Chris Sontag said,

It is many different sections of code ranging from five to 10 to 15 lines of code in multiple places that are of issue, up to large blocks of code that have been inappropriately copied ...

to me it seemed to be indicating that the linux people were unable to specificaly avoid using the same code as SCO claims to own, because they were not privy to the code; there are only so many different ways to write that pesky Hello World program. I'd be interested to see what SCO considers large Blocks, my guess is that this is a case of Shankspear sueing several of the infinite number of monkeys with typewriters for copyright infringement.

The other thing I find unusual is SCO has the unix and linux source code, why didn't they scan the code sets, there are programs available that will scan source code. My little brother wrote one that will tell you if a c program and a pascal program were writen by the same person correctly 85% of the time, and it much more accurate if the programs being scanned are written in the same language. Not using something like this on sources worth $billions is pretty irresponsible. Not using something like this before releasing a new version of Linux under the restrictions of the GPL is the stuff shareholder lawsuits are made of.

SCO CEO Publicly retracts threat to sue Linus

[SkArcher]

See here

U.Snooze = U.Loose

[hamhocks]

IANAL, however this sort of thing seems clear:

Why didn't you act earlier?. . . The point is we're really only recently seeing significant moves by many players, specifically IBM, to come out and state that they are moving wholesale to Linux.

With that statement, it seems like SCO provided evidence that it is vulnerable to the "laches defense." According to well established law, you cannot sit back and watch while an infringer enhances and markets your work, then litigate when the infringer starts making big bucks. In effect, SCO let IBM, and many other companies, take the risk and then try to claim the rewards.

Judge Learned Hand wrote, in a 1916 copyright dispute, that:

It must be obvious to every one familiar with equitable principles that it is inequitable for the owner of a copyright, with full notice of an intended infringement, to stand inactive while the proposed infringer spends large sums of money in its exploitation, and to intervene only when his speculation has proved a success. Delay under such circumstances allows the owner to speculate without risk with the other's money; he cannot possibly lose, and he may win.

See the recent (and infinitely puckish) opinion from MGM v. Sony (pdf).

Laches

[yerricde]

unlike trademarks, patents and copyright do not diminish with disuse.

Though trademarks come with the strongest "use it or lose it" responsibilities, patents and copyrights are still subject to some. If a judge finds that a patent holder or copyright owner has harmed an alleged infringer by delaying legal action, the doctrine of laches states that the monopoly holder cannot collect damages for alleged infringements that occurred prior to the alleged infringement.

In addition, copyrights have fair use limitations. If the owner of copyright in a published work refuses to sell copies of the work and refuses to license the work, it could be argued that the copyright owner thereby denies the existence of any "potential market for or value of the copyrighted work" (17 USC 107) that could be harmed by the alleged infringement.

The real question (and thus, not asked)

[SquareOfS]

What do you see as a company's options in the face of your warning?

I would suspend any new Linux-related activities until this is all sorted out. But first get that opinion of your legal counsel. If they say there is no problem and no issue, then you probably have nothing to worry about. But I doubt there is any attorney worth his salt that is going to say there is no potential of an issue here. There is a big issue.

Q: In other words, a company's options reduce to sending a flock of (expensive) lawyers to investigate the legal consequences of a highly complex claim, the factual merits of which you refuse to divulge?

A: Well, when you put it that way . . . still YES! A thousand times YES! Linux development shall come to a screeching halt!

Seriously, when's someone with standing going to countersue?

Open-source software is a common resource; what SCO is doing is analogous to saying "we know that there are poisoned wells. But we're not telling you which ones. Options: (1) drink and maybe die; or (2) pay us to tell you which wells are poisoned.

This is flagrantly abusive, and someone should unleash the flesh-eating lawyers on SCO.

Yet another opinion on the 10-15 lines.

[Dthoma]

A lot of people have posted here making incredibly lame jokes about the possible 10-15 lines of code. However, one point they make is true; we can't tell whether or not the alleged copied code is actually copied or just so obvious that it coincidentally happens to be similar. For certain applications, certain code is going to be identical.

For instance, if you want to use a single string for, say, holding user input, you'll probably use malloc() to declare a char* called 'str' or 'p'. This will probably be about 5 lines of code if you include error detection.

Then there are system calls one uses. If you're outputting a line of text, you'll probably use puts(), or printf(), or fprintf(). If you're getting a list of groups a user is a member of, you'll use getgroups() and/or getgid(). If you want to spawn a subprocess you'll use fork(). If you want to get the name of the current terminal you'll use ttyname(). All this creates code which is likely to look very similar.

Scope of claims

[Monster Munch]

I haven't seen a reference to this article yet?
Are they claiming ownership of C++??

The SCO Group now owns the entire bundle of products that were the property of the AT&T UNIX Systems Laboratory when Novell purchased USL. The SCO Group also owns all the products and property that belonged to SCO when Caldera purchased SCO (including the stuff SCO bought from Novell. It owns all the Caldera products and property. All in all, the SCO Group has a nice collection of products and properties.

For example, a February 1993 press release issued by Novell states: USL develops and markets the UNIX System V operating system, the TUXEDO* Enterprise Transaction Processing System, the C++ Programming Language System and other standards-based system software products to the worldwide computer industry.

They also mention COFF and ELF formats ...

Full article : http://www.mozillaquest.com/Linux03/ScoSource-01_S tory01.html#libraries_included

Both are copied from a common third source.

[refactored]

How many lines of code in the Linux kernel are a direct copyright violation? It's very extensive.

The next question is.....

Can they prove that those lines originate from their proprietary source, and not from some common (shared) free source?

eg. A text book, magazine, a HowTo, a chip manufacturers tutorial sheet, or some other code source.

The onus must be on them to prove that they did in fact create that code and not copy it from some other source.

I wonder if the disputed code is from BSD.

[mellon]

Back in the day, a lot of the Linux networking utilities were based on BSD networking utilities that were released as part of the 4.4BSD release after the USL settlement. I really don't know how many Linux utilities are descended from utilities in the 4.4BSD distribution, but it could be a substantial amount of code.

What led to the settlement between Berkeley and USL (in Berkeley's favor) was that USL had been taking BSD code for years, removing the BSD copyright and license (the first act is forbidden by law, the second by the license), slapping an AT&T proprietary notice on it, and committing it to their repository.

When this was discovered, Berkeley was in the position of being able to say to AT&T "there's no way you can make up for this. You just have to stop selling System V entirely." So they were basically forced to settle.

However, SCO had been receiving SysV tapes from USL for a long time before this settlement occurred. It's quite possible that what they have in their source code repository is a bunch of BSD code with AT&T proprietary notices on it.

Without opening up the legal records from the USL lawsuit and getting testimony from the people who worked on BSD and on System 5 way back when, it would be impossible for them to tell the difference.

To a person who wasn't aware of all this history, they would see a substantial similarity between a lot of "AT&T" code and a lot of Linux code. Not knowing that the "AT&T" code was actually Linux code, they might readily conclude that the code was stolen.

So my point is that it's actually possible that SCO honestly believes they are in the right, because they don't realize that a lot of the code that they think is theirs is actually code came from BSD.

They haven't violated the GPL

[Anonymous Coward]

SCO didn't GPL their code because someone else, _not_ SCO, copied sys V code into the Kernel. That doesn't "auto-magically" make the sys V code GPL code. It is rediculous to think that you can steal someone's code, copy it into a GPLed product, and the owner of the stolen code auto-magically loses all right to that code. Just because SCO distributed linux, and even contributed to linux DOES NOT MEAN THEY knew their code was in linux. THINK about what would be required for them to know sys V code was in linux: Some SCO linux hacker would have to have sys V code memorized (odds are who ever did the linux hacking for SCO had never seen sys V code), and then happen to read the section of Linux containing that code, and through some magic mental diff realize it is the same code. Look, Linux is a huge code base. Sys V is a huge code base. It is very _unlikely_ that a coder would notice stolen code unless they were specifically looking for it and wrote automated tools to do it! Not even Linus knows all the code in the kernel. Christ people, making faulty arguments against the SCO case doesn't help anyone. Too bad nobody will read this post because I don't have a /. account.

This will take decades...

[sterno]

The fact of the matter is that sorting out who owns any particular piece of Unix code could take decades in court. You find two pieces of code that are exactly the same in SCO Unix and Linux. This could be any of the following:

- An amazing coincidence
- SCO licensed software that was copied into Linux improperly
- SCO licensed software that was copied into Linux properly (i.e. by a SCO employee)
- Software from a 3rd party, properly incorporated by both SCO and Linux
- Software from a 3rd party, improperly incorporated by both SCO and Linux
- Software from a 3rd party, that properly/improperly incorporated it into their product which was then incorporated into SCO/Linux

Have you ever taken a look at the Unix family tree? There's no way they are going to be able to sort out the ownership of any individual piece of code. I mean hell, with Novell speaking up, it might be that SCO doesn't own squat (that they are just licensed the copyright that Novell still actually owns). Novell might not even own it either.

It will make some headlines, the world will move on, and in about a decade when this gets settled it will all be irrelevant.

Re:What Befuddles me

[willtsmith]

This is a very good point.

I would think that if someone has a well-defined coding style that the same stuff would crop up in more than one place. The same naming conventions, etc...

The question is ... is the code sections relevant. One can look at music where theft of concept is rampant and effectively expected. Proving infringement is notoriously difficult to do as the simply tweaking the rythm or melody will result in a different work.

It's almost like trying to copyright a design pattern. It's more like a method or a strategy, a way of going about things. In some cases these could be absoluetly trivial problems that have trivial solutions. In such cases, it's no wonder that one would respond (code) in the same fashion.

If their truly is "infringing" code it really has to be LARGE CHUNKS. As we all know, code from two different (but similar) systems is about as interchangeable melted lego bricks. The idea that someone could intentionally chop out small pieces VERBATIM and expect it to work is pretty silly.

Copying large grained, well encapsulated, functional modules of code is the only software plagarism could possibly be effective. 5-10 lines wouldn't work correctly without major modification, it's not worth copying. It would have to be in the range of hundreds and thousands of contiguous lines which match in a functional fashion.

Anything below that threshold is effectively the same as putting a thousand (code) monkeys in a room with PCs and seeing if they randomly repeat AT&Ts greatest sonnets.