Now how "legal" is this?

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

We can call them SCO or SCOX. Some companies are powerful and threatening enough that they may deserve name-calling. Other companies do things like making slides of evidence that destroys their case and presenting these slides to hundreds of witnesses.

SCO by itself is funnier than anything we can add to their name.

[ Reply to This | # ]

Glenn

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

And is it realistic to expect yourself to be on their first hit list? Let's remember SCO is a pretty small company with very limited resources.

IANAL, so this ain't legal advice!

[ Reply to This | # ]

There are various strategies possible, from presenting the whole scheme as a fraud to your Attorney General, going for the lawsuit against SCO, stretching the case till copyright arguments are settled to silently paying up. I can't promis a thing, but I'm pretty sure that people on the net will be rushing to give you their free advice (without warranty) when you post the details on the correct forum.

IANAL, but I advice against you signing a SCO license contract without a detailed review of all of the contract's clauses. It contains a few that strike me as extremely onerous.

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

Chris Sontag is being clever, like Nixon. "The source code is absolutely owned by SCO." That is likely true. But it is also completely irrelevant. The code was released under the BSD license; therefor while it may be "owned by", it is reusable by anyone, under the terms of the BSD license.

The question of the nessecity of displaying the BSD info as required by the BSD 1.0 license is open.

td

[ Reply to This | # ]

http://www.lemis.com/g rog/SCO/code-comparison.html

See also the comparison from Bruce Perence, with the pictures from SCO Forum here :

http://perens.com/Articles/ SCOCopiedCode.html

Greets.

[ Reply to This | # ]

Similar to the Boston.com article, except references SCO releasing the code themselves under BSD license

[ Reply to This | # ]

Babelfish translation amusing:

Moglen refers to a set of the attorney Mark Heise from the kanzlei, that represents SCO in the procedure with IBM. Mark of Heise had expressed the opinion in the last week that the GPL was altogether invalid , because she permits more than one backup copy of a software. This argumentation calls evenly Moglen now frivol, because hardly an attorney with dignity, who would like to exist with it before court

[ Reply to This | # ]

where rep Perense says:

"SCO don't have enough understanding of the case to be able to lie convincingly. Indeed, they have so little understanding that it's difficult to accuse them of lying."

Any chance that they will roll that out as defense in the end? (Wouldn't work though)

[ Reply to This | # ]

[ Reply to This | # ]

Funny also, that SCO were hiding the "Code" simply by changing the Font to greek ... and this can be reversed just as simple. Great this article from Perens ...

[ Reply to This | # ]

If they do issue a press release, I bet they'll mention the NDA.

[ Reply to This | # ]

What's a business to do? Try your local or federal prosecutor's office.

[ Reply to This | # ]

In brief, it isolates a couple of points which suggest that the current SMP code in SCO's own Unix may in fact have been lifted from Linux.

[ Reply to This | # ]

They can't mention the NDA because the slides that were photographed where not shown under an NDA. They were shown during the keynote to a public audience. Of course, that wouldn't stop them from suing, since 'publicy released' info is still 'absolutely owned by SCO', according to them.

On a side note of the SCO name, I make a prediction. In a couple of years, after this is over, I bet that SCO will be held in the same light within the tech industry as Ishtar is held in the movie industry. It will become a generic, and not too flattering, term. I can hardly wait.

Adam

[ Reply to This | # ]

That's a pretty big *IF*, so big it's not worth speculating on. Of course, if someone's actually seen SCO code with said code... then it's another matter.

[ Reply to This | # ]

Since it's "SCO Group" I suggest "scog" as the IT/IP equivilant of "scud": a trheat that's more annoying than effective.

"Well, he sure scogged us again today (hee,hee)." "Yet another scog attack againt the Linux community..."

[ Reply to This | # ]

Until the next timed press release that may likely bring it up again. :)

[ Reply to This | # ]

Hmmm....Analysis redux...let me get this straight.

Just looking at slide 'A':

0) We suppose both slides show results of comparing the same file taken form SCO kernel and from Linux kernel.

1) Analysis of the 'Linux right side' is easy: It's a definition of function atealloc(). The color coding does not seem to yield useful information. Some guy must like polycolor code. We find:

ulong_t atealloc(struct map *mp,size_t size)

that can be found in http://www.lemis.com/grog/SCO/a te_utils.c which is apparently (the top of the file says so):

Copyright (C) 1992 - 1997, 2000-2002 Silicon Graphics, Inc. and under the GPL

The text in ate_utils.c (not on the slide) uses function kmalloc(), which is Linux-specifc.

2) Analysis of the 'SCOX left side' is easy: It's just commentary. some loser obfuscated the code with a quick poke at Windows. It's evident that it is commentary as the the line of 'stars' goes right to the bottom. Nonobvious to non-coders. Again, the color coding does not seem to yield useful information. Some guy must like polycolor code.

The complete text says:

/* * Allocate 'size' units from the given map. * Return the base of the allocated space. * In a map, the addresses are increasing and the * list is terminated by a 0 size. * Algorithm is first-fit. * * As part of the kernel evolution toward modular naming, * the functions malloc and free are being renamed to rmalloc * and rfree. Compatibility will be maintained by by the * following assembler code: (also see free/rfree below). */

This matches the commentary that can be found *just above*

ulong_t atealloc()

in the source code file:

/* * Allocate 'size' units from the given map. * Return the base of the allocated space. * In a map, the addresses are increasing and the * list is terminated by a 0 size. * Algorithm is first-fit. */

ulong_t atealloc()

3) It is reasonable to assume that the SCOX side, having longer commentary, is actually derived from the Linux (sorry, SGI) side. This should be confirmed though.

Can we assume that the source underneath the commentary on the SCOX side is the same as on the Linux side? Not necessarily. It's just a possibility.

4) Conclusion:

We see that SCOX likes to compare comments We see that they found matching comments in a file A and file B We see that that comment in file B came from GPL-ed code from SGI

We conclude that file A, having amended comment, might be derived from GPL-ed code from SGI.

To do:

-> We need to know more about that 'kernel evolution toward modular naming' operation. Who did it when and why and where? This will shed light on whether that comment was added at SCOX or not. -> We need to know whether 'malloc' and 'free' were ever used in precursors of ate_utils.c, which would explain why malloc and free are mentioned in the commentary, not kmalloc and kfree. This will shed light on whether that comment was added at SCOX or not.

No further conclusions can be drawn, I guess. OTOH, this is all to weird. Does SCOX intend to deliberately mislead by throwing a smokescreen? After all, these are just badly obfuscated comments, begging to be analyzed...

The words "Trust No One" come to mind...The SCOX files ??? LOOOOL!

[ Reply to This | # ]

[ Reply to This | # ]

All I saw were the side by side comparisions of the comments (the one with the Greek encoding). I just read Greg Lehey's writeup. ROTFL There isn't even any code in common!

Even if the code had been copied identically, some courts have allowed revision of as little as 30 percent to "untaint" a work completely. Computer Associates International. Inc. v. Altai. Inc., 775 F.Supp. 544(E.D.N.Y. 1991 ) aff'd F.2d (2nd Cir. 1992). Copyright in software is not what a lot of you seem to think it is. This non-infringing code together with the similar looking comment has already been removed from the Linux kernel anyway....;-)

[ Reply to This | # ]

Also, I bet anything that if SCO starts suing customers, then the customers will get together and file a class action countersuit.

[ Reply to This | # ]

[ Reply to This | # ]

http://news.com.com/2008-1082- 5065859.html

The quote I like about SCO is:

[Cnet Question]: Any thoughts on SCO's claims on Linux?

[Ball's Answer]: I don't know the merits of the lawsuit, but I run their Unix and I'm taking it off that system. I just don't like the way it's being handled. I feel like I'm being threatened again. They never said anything to me, and if I was smart, I probably wouldn't mention it. But I don't like how they're doing it. What they're doing is casting a shadow over the whole Linux community. Look, when you've got Windows 98 not being supported, NT not being supported, OS/2 not being supported--if you're a decision maker in the IT field, you need to be able to look at Linux as something that's going to continue to be supported. It's a major consideration when you're making those decisions.

He has more comments that make the article a good read.

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

Anyone can threaten to sue you at any time. Anyone can threaten to harm you unless you pay them at any time.
First remember this- SCO Group CEO Darl McBride? stated clearly, succinctly and with great conviction,
"Contracts are what you use against people you have a relationship with."
http://biz.yahoo.com/prn ews/030528/law059_1.html
you won't just pay SCO. You'll have to sign a contract with them. Do you really want to have anything to do with that man, that company?

You'll have to spell out for them your company's pattern of Linux usage. Do you really want to divulge all that info to that man, that company?

More reasons not to sign up, even under the threat of a lawsuit are here -
Read this whole page (and please make improvements)
http://twiki.iwethey.org/twiki/bin/view/Main/SCOvsIBMSCOLicensingJustSayNO

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

Pretty amazing for a company that made $80 million last year, as I recall. And pretty amazing that A&TT and Novell sold the rights to Unix for so little when it is really worth so much. Nobody realized the copyrights are worth billions and billions until our Great Leader Darl McBride came along.

Remember when McBride said SCO did not intend to destroy Linux? But if, when you buy a copy of Mandrake for $39, you also have to pay $199 or $699 or $1399 to SCO, then who is going to buy Linux? Everyone would go to BSD. And if that is the case, what happens to SCO's billions in profits? I suppose that McBride, who has the answer for everything, will claim that SCO owns BSD.

[ Reply to This | # ]

[ Reply to This | # ]

"If you only have a few Linux installations, it may be easier and safer to just pay SCO's fee. If you have an extensive installation, there's a lot more money at stake. Meet with legal counsel and your executive immediately to assess your situation. Have a representative contact SCO to clarify its intentions as well as your liability."

This is exactly what SCO wants, IMO. Pay the 'just in case' fee, and SCO can keep running off of small-to-mid size companies for a while. Luckily we have a lawyer on retainer here, but it'd be nice if there was a "set defense" all these small companies could use. Otherwise, a lot of people are going to take advice like Info-Tech gave, and keep this farce going.

Btw, has anyone else noticed the lack of mainstream coverage on the German photos and the open source community's findings? I figured ZDNN, CNET, or Eweek would potentially have that listed on the front page. At least the Screen Savers of TechTV mentioned it last night.

[ Reply to This | # ]

rant start ...

Didn't SCO claim that roughly 1/5 or more of kernel code was copy from SCO's UNIX code? Then let me ask the question: How is it that SCO claims it did not notice the code until recently? As I understand it, they claim there are over 1 million LOC yet they failed to notice 20% of the kernel source were theirs? How the hell did they miss that?

I don't know about you folks, but if 20% of my stuff shows up in your stuff I would have noticed and that would certainly raise my curiousity. Their developers certainly work on kernel code everyday and they want me to believe that they didn't notice it until McBride started to wonder why Linux is getting better so fast and launched an investigation to come up with some bullshit like their code is the Linux kernel?

rant stop ...

[ Reply to This | # ]

"If you only have a few Linux installations, it may be easier and safer to just pay SCO's fee."

Glad to hear you didn't buy that. I have never seen anyone advise publicly that it may be "easier and safer" to just pay those big guys in suits and dark glasses that are hanging around you front door demanding money from you because "something could always happen". People do it of course - it's "easier and safer" - but it's not safer for very long.

[ Reply to This | # ]

[ Reply to This | # ]

http://linuxtoday.com /developer/2003082001326OSCYNT

[ Reply to This | # ]

Aside from the misunderstanding of the GPL (under the GPL, you don't assign your copyright to anyone else, just give rights to others to modify and distribute your work), this is precisely the "pregnant cow defense" that's been discussed and debunked here by legal types, correct?

[ Reply to This | # ]

Cock-eyed ComputerWorld writer shows that a free room and Forum-All-Areas pass can still buy you a lot of press.

http://www.computerworld.com/softwaretopics/os/story/0,10801,84184,00.html

Todd Wiess (todd_weiss@computerworld.com) presents this wonderfully balanced article and demonstrates to the world that when it comes to SCO, we should all just turn that frown upside down®.

[ Reply to This | # ]

In other words, could a reasonable court find that the unintentional, accidental distribution of "their" alleged code by SCO for over two years, continuing even after they were on about infringing code, merits the anullment of the licensing terms under which SCO distributed the code?

Or are they just whining?

[ Reply to This | # ]

Unfortunately, I had to pay SCO $599 to write this comment, $299 to look at my keyboard, $799 to actually type and as I run a dual-monitor system, I have entered into an indentured servitude contract with SCO for the remainder of my life.

[ Reply to This | # ]

[ Reply to This | # ]

Actually I read somewhere everybody who came to the conference had to sign an NDA to see that presentation. I don't know from personal knowledge.

Maybe it's just me, but I think it would be pretty to stupid to mention it (which doesn't mean SCO wouldn't - and I hope they would). If they issued a press release saying something along the lines of "You're not supposed to examine our evidence, it's under NDA.", this could be useful to their opponents later, I think.

Regarding being sued, I still think unless you are in the news for Linux, or a fortune 500 type company, SCO are going to have a hard job finding you. It's also worth remembering they have only $10m or so in cash - so that's not going to make it easy for them to pursue many cases at once. And if they did sue you, I expect IBM/RedHat/FSF to be all over it. And why approach them first, wait to see if they come to you (and I don't mean some form mail sent to everybody) - even if they do - they're going to start by offering you a "license". IANAL, so this isn't legal advice, just my opinion.

Regarding the lines of code, and 20%-40%. I agree with the point, how could they not notice. It's like if Stephen King read somebody's book, and didn't notice every 3rd or 4th chapter was one of his.

Additionally, if they think 20%-40% of Linux 2.4 infringes, and Linux 2.2 doesn't infringe at all - is that even possible (I don't know). My point, at this rate, the lines that they claim, wll be greater than the difference between 2.2 and 2.4

[ Reply to This | # ]

IANAL, but my opinion is yes they are whining. If there is negiglence and the pregnant cow defense flies, AND they get damages - it would seem they would be claiming damages largely or completely caused by their own negligence. That hardly seems just or equitable. I hope any judge or jury would see that.

Regarding mainstream press, the guys who really need to pick this up are Jonathan Berr (sp?) Bloomberg who wrote that piece that feature in Salt Lake Tribune (and elsewhere) about stock selling, and Lee Gomes (WSJ) who analyzed their last teleconference pretty well. I think address for Gomes is lee.gomes@wsj.com (found it in google based on a lucky guess)

[ Reply to This | # ]

The idiots in the article you link to are just too dumb to see that. Heck, the one who has many customers moving from SCO to Linux and Sun should be helping them do so, and picking up more secure business that way. What an idiot.

Vip

[ Reply to This | # ]

Search the amicus brief for the text [fn6] then [fn13] Read section [fn6] and then [fn13] thru [fn14].

The testimony of USL's Mitzi Bond and AT&T's Director of Licensing, Otis Wilson about the ownership of Unix derivatives are very enlightening. These AT&T witnesses testified they didn't own any derivatives unless they incorporated at least some AT&T code.

A lot of the controversey was over differences in Berkeley's rights to derive 32V - which it felt was in entirely in the public domain anyway - and the new licensing terms of System V. Either way AT&T's license would have (at worst) required that Berkeley distribute any of their own proprietary enhancements and modifications (to the extent that they didn't include AT&T code) to anyone without restriction on use, copying, or further distribution. That's hardly the interpretation SCO seems to put on things now is it?

http://cm.be ll-labs.com/cm/cs/who/dmr/bsdi/930107.amicus.txt

[ Reply to This | # ]

http ://www.cbronline.com/latestnews/62cbf9d13b40711e80256d880018c80f Quotes from Heise and Sontag

Anyway they say they're going to start with somebody who uses all of AIX/Dynix/Linux. If that ain't guaranteed to draw an IBM legal response, I don't know what is. And, if you don't have all 3, you *might* be safe for now - not a legal opinion, IANAL - just pragmatic comment.

I forget which article about SCO suing Linux customers, but there was one which said something about talking to their own largest customers about the issue. I'm not sure, but I think it sounded at least possible that they might try suing one of their own customers for buying SCO's own product - gee, I do hope they do.

[ Reply to This | # ]

SCOG vs. IBM as originally stated was a breach of contract suit. RedHat vs. SCOG seems to fall along the lines of intelectual property.

Any thoughts?

[ Reply to This | # ]

was that the extent of the advice? "Pay up"?
Or was there more nuance, like
"If SCO does not know you exist, keep it that way."
"If you get an invoice or a phone call, first get a receptionist to call some law enforcement agencies and register complaints,"
"Then send SCO requests for more documentation, keep delaying SCO until they issue an ultimatum"
"then if it's a pittance pay it, but also consider the risk that your name will get out and others will be calling on you with 'pay us or be sued' scams'

If the advice was a simplistic "pay up", would you be willing to email this newletter writer back, suggesting that his advice is not the most prudent, and adding either my advice or your own?

[ Reply to This | # ]

Excellent! PJ, you need to write about this I think!

No doubt IBM lawyers no about it, but it'd be even better if everybody did, including a decent analysis for us non-legal types.

[ Reply to This | # ]

Attorney Dan Ravicher comments on that case in that article.

I'll have a search function in the new site. That will help.

Gomes has been a true reporter. If I was going to email anyone, I'd email his boss and tell him how great he is, because SCO attacked him, you know. If anyone puts anyone's email address here, please follow this form: soandso at suchandsuch.com, not @suchandsuch.com. I'd hate to repay him with spam from bots crawling the site. Thanks.

[ Reply to This | # ]

This is from Harlan's link

-- START The contracts, letters and communications between the University and AT&T, when read together, prove the University's ownership over its enhancements and modifications of AT&T code and illuminate what limitations actually were imposed on the University's public software distributions. Those limitations are much narrower than what USL now claims.

As early as 1978, the University informed Western Electric (AT&T'ssubsidiary) that it was receiving requests by non-Western Electric licensees for software the University had derived from early versions of AT&T's UNIX. Exhibit A, attached to the Shapreau Decl. Western Electric's response to the University's inquiry regarding its distribution of derivative software to non-licensees set the precedent for the future:

[W]e take no position with regard to your use or distribution of software developed by you which does *not contain* any of our proprietary information such as, without limitation, the computer programs and documentation, or any portion thereof, related to the UNIX Operating System, including methods and concepts utilized therein. (Emphasis added.) Exhibit B, attached to the Shapreau Decl.

Mr. Edwin Baldwin, then Director of Licensing at Western Electric who signed Exhibit B, admitted that any University enhancements which did not "include our computer programs or documentation or the methods and concepts ... would appear not to be governed by the Software License Agreement." Baldwin Depo., at 125:8-128:11. In other words, as long as programs developed by the University from UNIX did notcontain UNIX code, documentation or trade secrets, AT&T had no objection to public dissemination. [fn6] -- END QUOTE

Add my emphasis last sentence (sorry don't know how to do that cool HTML stuff - please nobody explain - I don't want to clutter the discussion)

So my thoughts, are (IANAL so could be misunderstanding!):

1. is the side letter which IBM have saying IBM owns their own modifications enhancements (published in SCO's site, I think it says "derivates developed by you" or similar), like JFS etc., generically apply to all AT&T licensees. It does seem to me, that it looks like it at least applies to University of California too, and perhaps they think/thought it was generic ("set the precedent for the future")

2. To my reading, AT&T seem to have already (to some degree at least) given up on general methods and concepts, and hard restricted their position to don't copy code or documents developed by AT&T

[ Reply to This | # ]

I find it amusing that SCO will show source code from Linux, that is, supposedly, an identical copy of the code they own. They put both examples up, side by side on a slide projector. The amusing part is that the Linux code is displayed with comments and code, yet the SCO code has only displayed the comments and the code was rendered into Greek.

Now, it would seem that if the Linux portion was identical, then they went out of their way to simultaneously obfuscate and display their code to the viewers. It can also be interpreted that the code they obfuscated wasn't identical to Linux and they surely couldn't have their captive audience viewing anomolies of that sort.

'Here's some code from Linux that is IDENTICAL to SCO source code, and over here it the IDENTICAL SCO source code, translated to another language and obfuscated so that you don't see our intellectual property. However that copy on the right isn't our intellectual property, so you can view it. No wait, don't view it either. It's definately ours, but it's strangely not in Greek.'

[ Reply to This | # ]

This is the entirety of the 3rd part of their "action plan" they advise:

"At the end of the day, your organization needs to assess its liability risk. If you only have a few Linux installations, it may be easier and safer to just pay SCO's fee. If you have an extensive installation, there's a lot more money at stake. Meet with legal counsel and your executive immediately to assess your situation. Have a representative contact SCO to clarify its intentions as well as your liability." (Copyright Info-Tech Advisor)

The first 2 steps tell you to conduct a Linux inventory, and review plans to install Linux, and possibly delay any major installations until after the lawsuits with IBM and Redhat have concluded. Pretty much what SCO and MS would like to see. Info-Tech's main theme about this is what I've seen repeated here and elsewhere: "consult a lawyer." The part about paying up just in case was a bit surprising though.

[ Reply to This | # ]

-- START QUOTE

The law is clear that the authority to copy, distribute or display a copyrighted work granted by a license agreement is an absolute defense to a claim of copyright infringement. See, Effects Associates, Inc. v. Cohen, 908 F.2d 555,558 (9th Cir. 1990) (company which created and delivered footage at request of movie maker impliedly granted movie maker non-exclusive license to incorporate footage into movie, and thus movie maker did not infringe on maker's copyright);Apple Computer. Inc. v. Microsoft Corp., 717 F. Supp. 1428, 1432 (N.D. Cal. 1989)(to the extent license agreement authorizes the use of visual displays, it provides a defense to infringement claims based on the use of such visual displays).

-- END QUOTE

[ Reply to This | # ]

--in November 1985, the new software boiler plate which AT&T claims superseded prior licenses acknowledged that AT&T "claims no ownership interest in any portion of such a modification or derivative work" that does not include AT&T code Exhibit G, attached to the Shapreau Decl.];

[ Reply to This | # ]

So, as I read it, the "pregnant cow" case isn't especially germane to the situation thanks to the fact that it covered a contract that had not yet been executed, it covered a situation in which both parties were making a contract under fundamentally mistaken notions, SCO/Caldera had a responsibility and the unlimited ability to determine the facts of the matter yet didn't, and the GPL isn't a contract in the first place, correct?

I guess I need to read more to find out what case law would apply, and under what situations courts have allowed companies, in effect, a "do over", saved them from their own negligence, and hurt lots of other people in the process.

[ Reply to This | # ]

In this situation, in my opinion, you'd have to be a relatively large company for SCO to go after, and I believe they will sue an end user. Why? Despite what McBride says, ALL lawsuits take forever, so if you're SCO, you make your list, check it twice, pick someone and sue. Then you sit back and say, "See, now you've forced us to sue! Better start sending those license checks to SCO!" What's cheaper, buying SCO licenses or defending a lawsuit? I am sure many current Linux users would pay up, and other would-be Linux users will go with another OS.

So, the FUD will continue to support the pump-n-dump, SCOSource licensing revenue will continue to increase, and SCO will drag its feet on every lawsuit in which it is involved. In this case, SCO doesn't have to win, they merely have to present a case that is credible enough to avoid punitive sanctions. By the time the fat lady sings, they will have converted their FUD revenue into a more traditional business plan and will most likely do very much better than they were doing prior to catching lawsuit fever. Did you really expect justice?

Rob

[ Reply to This | # ]

IANAL, but the advice to hide and seek cover and hope SCO misses you makes perfect sense. Seek legal advice the moment you get the first letter from SCO. It will cost you money, but you'll be prevented from making stupid mistakes.

We would like to stay informed on SCO's actions...

[ Reply to This | # ]

IBM's side letter is applicable and is a SCO v IBM Exhibit. The amicus brief simply shows that AT&T had contacted all of it's licensees and offered the same interpretation of the language. It so happens that IBM's agreement is dated February 1, 1985 and the mass mailing was conducted the following August. amicus.txt [fn13] It appears that AT&T similarly tried to restrict other licensees' rights with respect to code they derived from AT&T's UNIX, because AT&T sent out a clarification to all its licensees in August, 1985 which stated: "Section 2.01. The last sentence was added to assure licensees that AT&T will claim ownership in the software that they developed -- only the portion of the software developed by AT&T." IBM has additional rights under Amendment X (the three party Novell/SCO/IBM contract) Para. 6. "the second to the last sentence of paragraph 9 of the February 1, 1985 amendment to SOFT-00015 is modified by deleting the words: and employess 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 services or providing any such service."

[ Reply to This | # ]

IANAL (still!) that AT&T gave up all rights to enhancements or modifications made by all licensees, not just IBM. So it seems clear that SCO should not be claiming some kind of "control rights" on IBM, SGI, etc.

IANAL (still!), 32V point is well-taken. If public domain, then BSD-style license on that code might be legitimate if obtained from that source at that time - and gives Linux rights, but further more, if a different copy is public domain - people can use that freely. My example: I can use public domain works in my own books (say a collection of Shakespeare plays perhaps with commentary of my own added) - evem copyright my book - but I can't stop other people also using the same underlying public domain works in their stuff too.

[ Reply to This | # ]

IANAL (still!) that AT&T gave up all rights to enhancements or modifications made by all licensees, not just IBM. So it seems clear that SCO should not be claiming some kind of "control rights" on IBM, SGI, etc.

IANAL (still!), 32V point is well-taken. If public domain, then BSD-style license on that code might be legitimate if obtained from that source at that time - and gives Linux rights, but further more, if a different copy is public domain - people can use that freely. My example: I can use public domain works in my own books (say a collection of Shakespeare plays perhaps with commentary of my own added) - evem copyright my book - but I can't stop other people also using the same underlying public domain works in their stuff too.

[ Reply to This | # ]

IANAL (still!) that AT&T gave up all rights to enhancements or modifications made by all licensees, not just IBM. So it seems clear that SCO should not be claiming some kind of "control rights" on IBM, SGI, etc.

IANAL (still!), 32V point is well-taken. If public domain, then BSD-style license on that code might be legitimate if obtained from that source at that time - and gives Linux rights, but further more, if a different copy is public domain - people can use that freely. My example: I can use public domain works in my own books (say a collection of Shakespeare plays perhaps with commentary of my own added) - evem copyright my book - but I can't stop other people also using the same underlying public domain works in their stuff too.

[ Reply to This | # ]

[ Reply to This | # ]

If SCO prevails on this theory, I think the TV stations in this land should file suit against all their competitors (cable TV, satellite TV, Internet broadcasters, etc) to eliminate industry competition in their favor. It makes about as much sense.

[ Reply to This | # ]

Absolutely. Isn't this again an appropriate for the well-known citation:

"We have to hang together or we will all hang separately"

The SCOX-Repellent-Howto is yet something that's missing at the Linux Documentation Projet!!!

[ Reply to This | # ]

It seems their counter argument boils down to (with no published evidence or data to support them)

1. Perens' assertions are wrong

2. It's his word against ours

3. SCO knows where the code came from, and it was added in appropriately.

4. SCO owns the code

[ Reply to This | # ]

Go Ernie! Ernie Ball Super Slinkies are the only strings I have ever used on my guitars. Good to see someone who isn't intimidated by SCO.

Now, if you look at the slides, you'll notice that all that is shown are comments. What makes this important is that these comments are NOT code comments! They are function descriptions. Anyone remember all the way back to the good old days of reverse engineering Apple II ROMs or PC BIOS ROMs? They were legal as long as they were "clean." What made reverse engineered ROMs clean? You have two teams: the first team took publicly available function descriptions, then wrote psuedo-code algorithms describing how the actual function code worked. These were then passed on to the second team who took the function descriptions and psuedo-code and made actual code that did the same thing.

Since the slides only display the function descriptions and SOME of the Linux code, this could have simply been a 100% legally reverse engineered function. Without the SCO code, there's no way to tell. The point is that function comments mean ABSOLUTELY NOTHING when it comes to "proving" copied code. If SCO had shown some comments in the middle of a function, that would be different, but they didn't.

From an programming point of view, SCO's so-called proof is a complete wash, regardless of where it came from or what copyright messages had or had not been included.

[ Reply to This | # ]

Yo! Somehow this citation reminds me of the big Luddite unrests at the beginning the industrialization, British Empire, 1840 or so. Jarring changes as people had to move from the countryside to the city and manual labor was being replaced by machines. Don't try to read too much into this association, though, I'm probably just tired. I'll shut up now.

[ Reply to This | # ]

http://www.publicbroadcasting.net/wxxi/news.newsmain?action=article &ARTICLE_ID=536151

[ Reply to This | # ]

[ Reply to This | # ]

50 percent of System V was derived from BSD, which was made available to the public, and even licensed by AT&T. The remaining 50 percent was largely contained in the Unix 32V (for VAX). Caldera is claiming copyright in 32V here based on an AT&T registration filed fourteen years after it started distributing it in an unlimited fashion to the public without any copyright notice:

http://www.tuhs.org/Archi ve/Caldera-license.pdf

32V is the same version that Berkeley based all of their distributions and work on. If it was ever in the public domain, then it still is, and no license is required. Here is what the California Regents said: C. USL FORFEITED ITS COPYRIGHT PROTECTION IN 32V WHEN IT PUBLICLY DISTRIBUTED 32V WITHOUT A COPYRIGHT NOTICE

USL's 32V code was created in 1978. DeFazio Decl., Exh. F. USL forfeited its copyright protection for 32V when its predecessors, Western Electric and AT&T, distributed 32V to the public without a copyright notice (McKusick Decl, at Para. 4 and Joint Decl., at Para. 28.4) and failed to register 32V until 1992, fourteen years after 32V was first published. [fn21]

[fn21] Unlike USL's 32V, post-Berne Convention Implementation Act works, created after March 1, 1989, are not required to have a copyright notice. 17 U.S.C. Section 405.

Under the 1976 Copyright Act, omission of a copyright notice from copies publicly distributed by the copyright owner invalidates the copyright in the work unless (1) notice has been omitted from a small number of copies[fn22], (2)registration is made within five years after publication without notice, or (3) notice was omitted in violation of an express requirement in writing. 17 U.S.C. Section 405(a).Here, USL can find no refuge in any of the three exceptions under 405(a) because thousands of copies of 32V were publicly distributed without copyright notices and registration followed publication by fourteen years. To the University's knowledge, no express requirement by Western Electric that notice be included in 32V was violated. In fact, Western Electric intentionally removed copyright notices from every file contained in 32V prior to publication. McKusick Decl., at Para. 4.

and [fn22]... http://cm.be ll-labs.com/cm/cs/who/dmr/bsdi/930107.amicus.txt

[ Reply to This | # ]

Yes but if the same method and concept is in Version Six and System V, they presumably can't retrospectively take it back. In any case, how could they prove which one of the same method and concept (found in both version 6 and system V) would be the origin, even if they could? I mean, it's the older stuff that is even more widely available, so more likely to be the source of a particular method and concept

And if I'm following all this correctly (which I'm not sure I am!), we're only talking about methods and concepts that originated at AT&T. Methods and concepts in the works by licensees which were added to their implements [JFS], are not part of anything AT&T laid claim to, except for a brief flirtation, which was quickly dropped.

[ Reply to This | # ]

If true, that would now give Linux something like 5 "good" sources: K&R book, public domain old Unix, BSD'd old unix, another BSD'd old unix (by Caldera), and BSD

[ Reply to This | # ]

Yes but if the same method and concept is in Version Six and System V, they presumably can't retrospectively take it back. In any case, how could they prove which one of the same method and concept (found in both version 6 and system V) would be the origin, even if they could? I mean, it's the older stuff that is even more widely available, so more likely to be the source of a particular method and concept

And if I'm following all this correctly (which I'm not sure I am!), we're only talking about methods and concepts that originated at AT&T. Methods and concepts in the works by licensees which were added to their implements [JFS], are not part of anything AT&T laid claim to, except for a brief flirtation, which was quickly dropped.

[ Reply to This | # ]

http://www.technewswor ld.com/perl/story/31386.html

Talking heads here, including Yankee Group.

In response, Chris Sontag, SCO's senior vice president and general manager of SCO Source, told TechNewsWorld that the company has now unveiled the offending code and that it can be remedied.

"The vast majority of the code [in violation] is the derivative work from IBM, so that's a great place to start," Sontag stated. "We're talking about more than one million lines of code that can be remedied."

Linux advocate: More SCO evidence flawed

http://ww w.infoworld.com/article/03/08/20/HNscomoreflaws_1.html

Interestingly this is an example of "obfuscated" code according to SCO, but the developer speaks about how he did it in this article

[ Reply to This | # ]

To the extent that this demo really has anything to do with a "First Fit" (malloc.c) alogorithm, I see little need to find a license anyway. Something like bin sorting just isn't protected by copyright anyway.

[ Reply to This | # ]

[ Reply to This | # ]

http://www.perens.com/SCO/SCO SlideShow.html

Says the original algorithm originates in Knuth. Funny that, I read that speculation before in a lwn.net comment - so maybe somebody knew?

The two examples really do seem to fit what was reported in Byte, Ian Lance Taylor (sp?), and that German guy who didn't sign an NDA [who spoke about the switch statements being similar]

[ Reply to This | # ]

"Yankee Group senior analyst Dana Gardner"

A senior analyst from the Yankee Group, and no DiDio?

[ Reply to This | # ]

"Torvalds: They [SCO] are smoking crack."

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

http://www.technewswor ld.com/perl/story/31386.html

Does anyone know where this "unvieled code" can actually be viewed? Or are they simply claiming ownership of the German publication?

Here is why I think someone needs to propose to publish 32V online as a work "in the public domain", and ask a court for a declaratory judgment regarding the validity of it's copyright (a federal case of actual controversy):

Start extract... Albert said that regardless of how old the code in question is, SCO does hold the copyright that goes back to the origins of Unix. The company has been arguing that the case hinges mainly on whether Linux is a derivative of Unix.

"Unix is an old operating system, so it's been around a while, but that doesn't necessarily mean SCO has less of an issue with it," said Albert. "The argument is not that SCO created these portions; even if they didn't, they still own the copyright on it."

SCO's Sontag rejected claims that the copied code dates back to the origins of Unix, citing SCO's several-year-old System V version 4.1 as the main version of software where illegal copying occurred.

end extract ... System VR4 is not very far removed from it's BSD and 32V predecessors.

[ Reply to This | # ]

" Michael Heise, a partner with Boies, Schiller, & Flexner who's representing SCO, downplayed concerns that the contested code may be covered by an open-source license. In an interview with CNET News.com at the SCO show, Heise said that even if, hypothetically, some older Caldera code were open source, it wouldn't make a difference to the case.

"Let's say you have a hundred files, and you put one of your hundred files under the GPL. That doesn't mean you've lost the rights to your other 99 files," Heise said. "So I don't think it's going to have an impact."

Did I read what I think I just read? Did the legal counsel for SCO just admit that if you place code under the GPL and distribute it you then lose the right to that code except as specified by the GPL? Doesn't this contradict what SCO keeps saying about the GPL, that it isn't really enforceable?

[ Reply to This | # ]

http://www .ecommercetimes.com/images/story-sco-undercuts-1.jpg

for this article:

http://www.ecommercet imes.com/perl/story/31386.html

[ Reply to This | # ]

Finally, we all know the Knuth!

[ Reply to This | # ]

[ Reply to This | # ]

-- the weak point of the GPL will probably prove to be the disclaimer of warranty. A court might hold that unconscionable as an industry practice.

-- this could drive a wedge between Linux distribution commercially (by XXX XXX [tm], say) and non commercial. A court might hold, for example, that XXX XXX or IBM really must warrant their distribution.

-- even if SCO loses, dredging this up could be a problem.

-- we need to mobilize more than just software distributors. Right now, McBride is saying his actions are pro-business. Actually, he is for some businesses (SCO, proprietary model software distribution) against other business segments and models. O'Reilly, McGraw-Hill and friends make more money off of Linux than SCO ever dreamed of.

-- one way to mobilize an industry segment is to scare it with legal risk. If Linux violates copyright, so do the booksellers who distribute it. Part of the reason I pay $40 for a Linux book is the convenience of not downloading Linux (or paying XXX XXX [tm] prices).

-- the business press will wake up if it looks like a software publisher vs. book publisher fight. Especially the part of the press that is owned by book publishers.... Suddenly McBride won't be a business hero. He'll be a threat.

-- how to enlist the booksellers? Sue 'em. Class action. They're exposing us to risk, selling us copyrighted material they have no right to distribute, and disclaiming all warranty too.

Why should we pay $40-100 prices for bad stuff?

[ Reply to This | # ]

[ Reply to This | # ]

I knew what you meant, and enjoyed the Knuth pun very much. I was able to parse your original sentence because the human brain, unlike spectrum analysts looking at code, can decipher meaning from fog. :)

[ Reply to This | # ]

Having discovered that the library at work had a 1st edn copy of K&R I've now looked for the first example in both that and my second edition copy. There is a similar algorithm on P187 of the 2nd edition (also in 1st edn but that's at work so no Page no.) but I can't find the exact code. could people please either give a Page No. or stop saying it is in K&R as as far as I can see only the algorithm is in K&R, not the same code.

A lot of people (not specifically here) seem to be getting confused now about the first example being SGI code contributed by HP. Remember SCO always said that their code viewing circus was not about the IBM code but contributions by other IHVs into which class both SGI and HP fit. I'll make a bit of a guess here and suggest that as most kernel developers didn't have Itanium kit at the time that code was committed the chap from HP who added it was the maintainer of the arch/ia64 area and someone from SGI submitted it to him. That is at least one possibility to address the issue of how this code got in.

Also if anyone hasn't read the eWeek article Brenda linked I'd recommend it - I agree with her that Linus sounds upset.

[ Reply to This | # ]

http://ww w.infoworld.com/article/03/08/20/HNscomoreflaws_1.html

"SCO disputed Perens' claims. "We're the owners of the Unix (AT&T) System V code, and so we would know what it would look like," he said.

SCO strongly claims the intimate knowledge of UNIX System V code. I would love to hear them explain to the judge and the jury how these alleged code got into Linux without being detected by its developers looking at the Linux kernel everyday for, what, approx 2 years?!

Moreover, if they claim to have intimate knowledge of UNIX System V code, then they must also have intimate knowledge of the version of the Linux kernel they distribute since both are their products. It is unconceivable that they did not recognize that the UNIX System V code were copied into Linux. Unless their developers were falling asleep at the wheel while coding.

[ Reply to This | # ]

From Yahoo SCOX forum, http://tinyurl.com/ko2b

Re: Is IBM holding back patents/by: StockSentinel

First of all, all they need is one widely violated patent to stick and they have the wedge they need to split SCO case apart. Second it’s better to go with a few slam-dunk patent violations then mix them up with ones that might be questionable. Third, there is more to those patents IBM chose than people have picked up here, look at IBM’s affirmative defenses. Then think about the 20 year old patent that was tossed in there and how three of the IBM affirmative defenses could come into play if SCO is clearly violating it, and they have no other defense and can’t invalidate the patent. If you figure it out it is hilarious as hell.

I suppose he means the ones about laches, delay, waiver, unclean hands, etc?

[ Reply to This | # ]

I believe the difference between patents and copyrigts are that copyrights give no, or very little, protection to the ideas but last for very long. Patents are wider in scope and protects not only the expression, actual code, but alse methods and priciples, thus stopping reimplementations, but they are limited in time to 15 years.

[ Reply to This | # ]

http://www.converium.com/web/converium/converium.nsf /0/5731FF9F4372B6ED85256B43006EA07D?OpenDocument

Personally, I think SCO is not reading between the lines of IBM's responses, and I think there is a message there that Darl has missed entirely. Also I think it was a big mistake of Darl and Sontag to publicly mock/disparage IBM's filings, as it is goading them not to settle, but get more aggressive.

The hidden message, I think, true to many stories (possibly apocryphal) about IBM's history, is I think: "Do you really want us to search our humongous database of patents for possible infringements by SCOX/Canopy?"

[ Reply to This | # ]

[ Reply to This | # ]

[ Reply to This | # ]

Don't you remember after suing IBM, that SCO made a big huff and puff about how they were going to sue another hardware manufacturer. Is it SGI? Is it Sun? Is it HP? Went the chorus

Then they didn't sue anybody else, just revised their claim against IBM to include Sequent subsiduary.

I just don't think SCO wants to fight 3 major legal battles at once (and that sssumes that nobody else is going to sue them - which I think might be assuming too much). Or maybe I'm underestimating their levels of loopiness.

[ Reply to This | # ]

The oldest is 4,814,746, dated March 21, 1989 and Filed August 11, 1986. So the poster wasn't being precise, just saying that they were really old as patents go.

[ Reply to This | # ]

"Prior to a recent amendment prompted by the Agreement on Trade-Related Aspects of Intellectual Property (TRIPS) accompanying the Uruguay Round GATT, patents were normally issued for a non-renewable period of seventeen years, measured from the date of issuance. See § 154 of Title 35. Under the amended provision (which took effect June 8, 1995) the term will be twenty years measured from the date of application."

So if I understand this patents issued before 95 are for 17 years.

[ Reply to This | # ]

Ullman claimed that the "First Fit" algorithm can differ from an optimal packing by as much at 70%. M. Garey, R. Graham, J. Ullman, "Worst-case analysis of memory allocation algorithms, " Proc. 4th ACM Symposium on Theory of Computing, 1972.

Those who are interested can see if Donald Knuth even felt that "common sense" or heuristic algorithms should be allowed to be patented. heuristic rule Definition: [n] a commonsense rule (or set of rules) intended to increase the probability of solving some problem. Remember that a maitre d' might use a seating chart and a "first fit" algorithm to find a table for a party of six. What's so different about malloc.c? http://www.pluto.linux.it/meeting/meeting1999/atti/no-pa tents/brevetti/docs/knuth_letter_en.html

[ Reply to This | # ]

[ Reply to This | # ]

That was part of the comments on slide 6 of SCO's presentation. If I read that correctly, SCO thinks it owns more than just Linux. What's to keep them from claiming ownership of things like Quake from id? id developed a Solaris version on Solaris machines. I would assume they had a proper license for their Solaris boxes. So by SCO's interpretation, they now own the Quake franchise. I'm sure id will happily turn over all their profits from the game market. Even worse for id, they release their code after a time as open source. How dare they give away SCO's IP!

[ Reply to This | # ]

http://news.com.com/210 0-1016-5066410.html?tag=nl

They quote everything SCO says as matter-of-fact; however, practically everything Perens said is paraphrased and weakened - not giving any strength to Parens arguments.

I thought they were being relatively unbiased till I read that garbage.

[ Reply to This | # ]

Well, the time frame I'm getting from the data gathered is the same one that Project Monterey takes place in and I believe that it is there that SCO believes it has an as-yet-unrevealed smoking gun (or at least a well-hidden dagger laced with poison).

As I recall Sequent, IBM and SCO were partners in this project and it eventually went down the drain, the biggest outcome being that IBM bought up Sequent and SCO sat around twiddling its thumbs.

The timeframe I recall seeing makes this appear to be a revenge scheme to get back at IBM for poking a hole in the Monterey Project. Or rather to use some loose threads in that deal to tap IBM for a few hundred million dollars in a settlement.

Seeing as these partners, and yes SCO and IBM were technology partners at one time lest we forget, it is within reason to believe they would have worked together on the same codebase for Monterey and its possible that code between the three companies were shared since they were merging technologies to create a 64-bit SuperLinux of sorts.

This does not, however, explain to me why SCO would believe that NUMA (as in NUMA-Q or Sequent NUMA), Read-Copy-Update, XFS, JFS etc. would belong to them.

I think the answers lie with Project Monterey and the terms of partnership in that deal. I'm almost certain that Project Monterey is the source of SCO's confidence/arrogance and that it has yet to reveal its final hand.

I hope the community investigates this particular area as I think it will reveal a great deal about what SCO intends to deal in court should that day ever come.

Just my 2 cents.

[ Reply to This | # ]

http://www.silicon.com/ news/500008/1/1005787.html

[ Reply to This | # ]

J.F.,

You got it in one.

This has been their argument, starting around Sept. 2002.

They have grandiose fantasies.

[ Reply to This | # ]

[ Reply to This | # ]

"STOP PRESS. I wrote this yesterday after careful examination of the evidence. I then continued my examinations and discovered that I was wrong. I am now convinced that the code in question was, indeed, derived from UNIX System V.4, and not an earlier version of UNIX, as some other people have claimed. This does not mean that it was stolen. I'll update this page when I have time, hopefully in the next 12 to 18 hours, with detailed reasoning of my conclusions."

Y.

[ Reply to This | # ]

Greg Lehey's page is updated.

Some of his conclusions:

"My analysis differs from that of Bruce Perens and Co. They don't compare the exact code, so they don't notice that the exact version imported clearly came from System V.4, and not from older, freely available versions.

"How important is this code? 17 years ago, BSD had something better. SCO can hardly claim that a couple of error checks and mutual exclusion, both traditional techniques described in many books, are intellectual property worthy of legal protection.

"How can SCO claim to have state-of-the-art SMP technology when they're still using core kernel code dating from the infancy of UNIX?

"There's no doubt whatsoever in my mind that it came from System V.4. Whether it was legal or not is not something that I can assess; given the status of the code, it's very likely that it has been published elsewhere, and that SGI had a right to use it. It's surprising, though, that there's no AT&T copyright notice in there.

"On the other hand, it's difficult to see malice in the incorporated code. Linux already has a malloc() function. Why did it need another one?

"This code was used only in the ia64 port, which was experimental.

"The code was removed from Linux on 4 July 2003, as shown by the commit log. As a result, SCO's claims that this code is in Linux are factually incorrect.

"Even if SCO could prove that this code justified charging license fees, they could only do it from Intel ia64 users, a vanishingly small number of people."

If Greg is right, Marcelo W. Tosatti and SGI may be able to give more details as to the origins and legitimacy of the code.

[ Reply to This | # ]

nexex - about DiDio - I saw that photo - the jaws that bite, the claws that catch - and I shuddered, remembering the jub-jub bird and the frumious bandersnatch, which had lain quiet in my mind up till that point in time. Grabbing my bandana and my harpoon I snuck out of the Baton Rouge flat and crept out - with Bobby McGee of course, my trustworthy companion - to the side of the road and thumbed down a diesel.

No Jub-Jub Bird was going to catch me and mine! I would run as long as I could, and fight, even if the Jub-Jub Bird brought the Frumious Bandersnatch into the hunt ...

At that point in time Hobbes attacked me without provocation and I lost my train of thought, into a deep deep ravine, wherein lurked a Snark of the Boojum variety, which I could escape if I .... ;)

[ Reply to This | # ]

1. Windows Advantage - - Windows 2000 riches await smart migrators Abstract: Windows 2000 riches await smart migrators By Laura DiDio. It" is almost here.. The "It" is Microsoft's Windows 2000, arguably one of the most anticipated productions since David O. Selznick brought "Gone with the Wind" to the big screen.More than six years in the making, Windows 2000 consists of Windows 2000 Pr http://windowsad vantage.com/trenches/11-15-99_riches.asp size 10934 bytes - 11/2/2001 11:09:34 PM GMT   2. Windows Advantage - - Users express cautious optimism about Windows 2000 deployments Abstract: Users express cautious optimism about Windows 2000 deployments By Laura DiDio. All talk of Windows 2000 migration boils down to a single question: "When?. That the question is "when" and not "if" is a testament to Microsoft's continuing clout, marketing muscle and its determination to forge ahead despite http://window sadvantage.com/tech_edge/01-17-00_optimism.asp size 10592 bytes - 11/2/2001 11:08:40 PM GMT   http://windowsadvantage.co m/search/default.asp

[ Reply to This | # ]

I asked a kernel coder your question and a CEO of a Linux company, and here's what they both told me, and this isn't legal advice, because they aren't lawyers. It's a tech perspective. Most companies never use any of the high end stuff at issue here. If you don't install it on your computer, you can't infringe even in SCO's wildest dreams. The 2.2 kernel is more than adequate for most business office functions. If you check, you may find that is what you have running anyway. If so, you have nothing to worry about. Even SCO said it is a solution. Alternatively, you could compile your own kernel, avoiding the famed four, but just running the 2.2 is certainly a lot easier. This is in addition to all the other things you can do, like ask your lawyer.

[ Reply to This | # ]