I am an BSD Lover,
but am sorry for Linux

Check this:
http://linux.quicksurf.com/index.php?p=1278
http://www.computerworld.com/governmenttopics/government/legalissues/sto...
http://www.eweek.com/article2/0,1759,1752775,00.asp?kc=EWRSS03119TX1K000...

Hmmm... as Rob points out, my choice of language was rather poor. Yes, ELF is not strictly "derived" from COFF in that someone didn't just hack on the COFF code to produce ELF. However, the people responsible for the ELF format didn't start with a blank slate. They looked at COFF (and ECOFF etc. etc. etc.), worked out all the things that were wrong with it, then went and designed ELF. There are many things in common because they started with COFF as a baseline and to that extent, it is a derivative.

As normal groklaw tells the real story.

http://www.groklaw.net/article.php?story=20040722135616439

linux world appears to be a waste of webspace.

Irrespective of all other factors, the predecessor in interest to SCO was party to TISC, and therefore laches applies.

Does SCO have no lawyers on their legal staff, or do they hope for judicial ignorance?

The game of "you dumbass judges don't know shit about coding and operating systems and won't never find out nothing about what high tech does" is over.

Either the ELF claim is true, in which case, laches requires dismissal, or it is not true, in which case, cursory examination will result in a summary dismissal.

Any other possibility requires corruption or stupidity on a monumental level.

I believe that even in good old Bouvier's, which anyone with apt-get can install with no dependency conflicts, even in 2004, culpable liability is still spellt culpable liability. Perhaps someone should send someone a sarge install cd?

Or lawyers should learn to read?

Best,
M

As far a I know, SCO released it's own Linux distribution (Caldera Linux) in the past *UNDER THE GPL LICENCE*...
got the point? They obviously do not :)

Btw if they want some licence fees from anyone, they must first (successfuly) sue every Linux distributor - because they all release it under GPL and anybody can base his rights (according to GPL) on any single release.

If they come to me (not very likely), they will first have to prove invalidity of all Linux distributions licences. Very unlikely.

Fred Down wrote: "I did not state that DLLs and shared libraries are the same. They are similar in function."
I never said that you stated DLLs and shared libraries are the same. They are virtually identical in function, in fact. I was the one who equated the two. Any distinction between the two is, for the purposes of this discussion, hair splitting. The primary difference is that DLLs are specific to Windows, and "shared libraries" as I use the term, is rather generic. You seem to believe that the term "shared library" or "shared runtime library" is somehow specific to the UNIX/Linux world. Fine, whatever. DLL stands for Dynamic Link Library, and as such, means pretty much the same thing as "shared runtime library" or "shared library."

And yes, thank you, I've done Java programming for many years, and know full well that JNI is implemented with DLLs in the Windows world, and with shared libraries elsewhere. That's because these two mechanisms serve the same basic function, and therefore, Sun chose to use the mechanism appropriate to the target platform. Nothing new or exciting here.

I really don't care that your knowledge of ELF comes from practical experience. You said yourself that ELF was developed in part to address shortcomings of COFF, as a replacement for COFF. That would imply that ELF is legitimately a successor technology. I realize that you took exception to someone stating that ELF was derived from COFF; that would be similar to claiming that PNG is derived from GIF, which is untrue (although PNG was developed to replace GIF and improve upon it simultaneously -- I'm one of the PNG spec authors). So I understand your point, but you could have expressed it better -- and you could also acknowledge that, in part, the comment that you were taking exception to wasn't entirely off base in concept. ELF looks like PE, and both ELF and PE/PEF resemble COFF. There are no new ideas here.

"This detracts from my point that SCO by claiming ELF can seriously damage the Open Source."
It took me several read-throughs before I understood your meaning. A couple commas would help your sentence be more comprehensible. It should read "...SCO, by claiming ELF, can seriously damage..."

Your point is well taken. But that damage can only take place if SCO's claim can be substantiated, and that is far from a foregone conclusion.

"I find it strange that you find 'sorta' so upsetting but not comments like [...]"
I find "sorta" upsetting because professional journalists are not supposed to use such language in published news articles. I hold the journalist to a higher standard than the typical geek who posts a comment such as the one you cited. Your objection to the potty humor inherent in that comment is duly noted; however, the person who wrote that comment, while immature, was making a valid point. One easy way to detect bullsh*t and deflate specious arguments is to take someone's argument and substitute some nouns, and see if the same line of reasoning holds. In this case, the choice was crude -- someone chose to use feces in place of source code. The analogy is crude, and doesn't necessarily hold, but it does serve to point up the idiocy of SCO's claims by lampooning them in such an outrageous manner.

Someone at SCO must be sniffing glue.

Microsoft scares me much more than SCO. I think the comments on http://www.cto-corner.com/pages/1/index.htm are a bit closer to the mark than any of us want to believe. MS eventually will patnent breathing, then what do we do?

I'm not as afraid of SCO as I am of Microsoft. I believe IBM, Novell and the community at large can handle SCO, but Microsoft is scary. Read the article at http://www.cto-corner.com/pages/1/index.htm. I think this guy is right on, Microsoft will patent breathing, then what do we do?

ELF was designed to replace COFF and COFFs various extended formats. One of the shortcomings of COFF was that there was no standard shared library mechanism. AIX at least AIX 4.3 still uses COFF (somewhat extended) and its shared libraries are therefore non standard. ELF addressed many other limitations of COFF.

I did not state that DLLs and shared libraries are the same. They are similar in function.

To illustrate what I mean. Java uses DLLs in Windows to implement JNI whereas on Solaris, Linux and various flavours of UNIX shared libraries are used.

My knowledge of ELF comes does not come from 'reading several technical articles’. It comes from converting COFF tools to work with both ELF and COFF formats. My source of information was Mary Lou Nohr Understanding ELF Object Files and Debugging Tools.

This detracts from my point that SCO by claiming ELF can seriously damage the Open Source.

I find it strange that you find 'sorta' so upsetting but not comments like "
All fecal releases I have seen are *substantially similar* to my fecal releases. I claim the shape/form/ip of all fecal releases as my own.
I demand that all SCO persons either pay me to crap , or STOP!"
But then I am not an English major.

Before anyone gets too excited I should declare that I had an aricle published in a magazine owned by IDG. I really have not received any payments from Microsoft.

Fred Down commented on 21 July 2004:

> SCO has a very powerful legal team which may be more relevant than the strength of its technical arguments.

SCO has a legal team with a powerful NAME. The actual quality of the legal work appears to be quite limited, with conflicting assertions in different courts and at different times. The IBM legal team has also engaged in some stiletto humor, clarifying the order of SCO's misnumbered pleading paragraphs, and offering other editorial "help".

It would really be quite funny if the whole thing weren't such an ugly demonstration of the problems with US civil litigation.

This type of anouncements doesn't seam to work anymore.
There are too many people that know the history to contradict them, and contradict them fast with high level of credibility.

A year ago this kind of story would have been cited all around the net and pumped the SCO stock to astronomic levels. Now, not even grooklaw that specializes in reporting from the SCO case mentions this sory and the SCO stock in almost free fall.

If you have a disposition for conspiracy theory, you could almost believe that people from the open source community publishes articles like this where they claim that SCO claims they own technology that is clearly not theirs, and by doing so reducing whatever little credibility SCO may have left.

Fred Down wrote: "The case against SCO is not helped by childish ranting nor by berating journalists who do not write what you want to read."

Sir, it is not that people are berating the "journalist" in this case for writing what they do not want to read. It is that this "journalist" did little or no fact checking, obviously has a poor understanding of the technical matters at hand, and clearly slanted the article in SCO's favor. This is far from un-biased reporting of facts. Furthermore, the "journalist" who wrote this article did not adhere to standards and practices for journalism, such as the use of proper grammar and diction. "Sorta," for example, is considered unacceptable in professional discourse.

As to slanted writing, one need look no further than the mention of the GPL as a "viral" license (something that no self-respecting Linux publication should do, since Linux relies upon the GPL heavily; sadly, since Microsoft now owns the parent company of LinuxWorld, this sort of nonsense is expected). It's also clear from other word and grammar choices that the author clearly accepts SCO's claims at face value, not as the untested assertions they are. This is bias. While some bias is unavoidable in journalism, bias this blatant should be avoided. I don't think it's childish to demand better of a professional journalist.

Regarding Fred Down's comments, I have two rebutting observations:

(1) The comparison between ELF and Microsoft's DLL's is in fact not valid at all, and if you had a grasp of basic deductive logic you'd realize this. Yes, ELF specifies "amongst other things" a standard mechanism for shared libraries. However, ELF's primary function (and this is the key) is to provide a format for binary executables in Linux. DLL's in Windows serve a single, very different function: they solely provide a binary format for shared libraries, and specify a mechanism for loading them at runtime.

In this regard, ELF more correctly maps onto COFF in Windows and a.out in the UNIX/Linux world, not to mention PEF (which I think was mainly a Mac thing). I very much remember the transition from a.out binaries to ELF binaries in Linux, incidentally. ELF may specify a mechanism for loading shared runtime libraries, but its functionality is a superset of the DLL mechanism in Windows, because ELF does so much more.

(2) You claim that ELF is not derived from COFF. Well, what you claim flies in the face of several technical articles that I've read regarding ELF. I'm willing to concede that the ELF implementation does not derive from any COFF-related source code. I'm also willing to concede that the ELF specification may not directly reference the COFF specification (something I can't check at the moment). But it's pretty well established that there are substantial similarities between ELF, COFF, and for that matter, PEF. The core underlying ideas were borrowed freely, as they should be -- why reinvent the wheel?

Mr. Down may earn his living developing for Linux, but he's not the only one. I don't do any kernel hacking these days, but I still write code professionally.

This ELF thing would be very serious, Linux depends heavily on shared libaries, back in the a.out days this was not true. If ELF had to be removed virtually all applications written since the a.out day would break.

More seriously ELF is used by BSD. I beleive it would not have been around when USL and BSD came to their settlement.

If SCO can stop both BSD and Linux then open source will suffer greatly.

My contingency plan should Linux be blocked was to switch to BSD.

SCO has a very powerful legal team which may be more relevant than the strength of its technical arguments.

The case against SCO is not helped by childish ranting nor by berating journalists who do not write what you want to read.

Seems like Linuxworld want technology editors. Time for some real linux advocacy on a psudo-linux site....

http://www.sys-con.com/careers/editor.cfm

Hey, linux moved to ELF from COFF (i think it was COFF), we can move again if need be!

Seems a bit rediculous though as the move to ELF came a long long time ago, surely thats going to make it pretty hard to prove a case based on ELF...

But, of course, my linux history knowledge is a little lacking on the ELF front, all i remember is the first time i compiled something as a shared object... that was a long long while ago

Matt T: Re: [IBM should just buy SCO and dissolve them.]

This was actually SCO's plan from day one... they wanted to be bought, esentially being paid to shut up, and the management gets a nice golden parachute... from which they float down into another company and do the same thing. If you read up on groklaw, you'll see that people such as Darl mcBride, Rob Enderle, etc. see lawsuits, frivolous claims, etc. as simply doing business.

By telling SCO to basically shove it, IBM immediately added legitimacy to the GPL, Linux, etc. You can bet that 90% of the media (read: SCO and Microsoft's mouthpiece) would have spun it as "Linux has IP issues, IBM pays off company to silence credible claims..." you get the idea. The idea of this is to silence this crap the first time around.

Oh yeah, and you don't threaten a company like IBM. They're 1) Huge, 2) Have enormous resources, and 3) possess a rabid army of lawyers. Don't forget...this company went through a large ordeal with the US government years ago.

Just my $0.02

Oh for goodness' sake! There's no doubt that SCO's latest legal move is complete and utter nonsense, and fully deserving of all the invective it receives. The journalist that wrote the article however, is only guilty of not explaining things in a sufficiently technically accurate manner.

It seems these days that anyone who even WRITES about SCO gets the same amount of abuse as the company itself.

Focus people... FOCUS!!!!

everything can be replaced! code in linux can be replaced
by another code! SCO can be replaced by arsehole!

After reading that astonishly badly written, biased and obviously little-researched article, I was about to post regarding the completely unprofessional use of "sorta" and "the stuff". Then I saw that I wasn't the only one. Shame on your linuxworld for publishing this horrendous piece of "reporting"

A quick look at the standard in question at
(http://www.x86.org/ftp/manuals/tools/elf.pdf) reveals that:

1. Both Novell and The Santa Cruz Operation (old SCO) were members of the committee that wrote the standard.

2. "The TIS Committee grants...a non-exclusive, worldwide, royalty-free license to use the information disclosed in this Specification to make your software TIS-compliant..."

total garbage, first time reading an article on this site, and the last.

M$ does own this site, I've even seen banner adds for visual studio .NET! on this site, so there you have it.

Linux world is just a FUD propaganda machine for M$

Matt T commented on 20 July 2004:
> IBM should just buy SCO and dissolve the company.
So that the next nuisance lawsuit can crop up?
No, SCO must be killed through the courts,
that would create a solid barrier against more
anti Linux lawsuits.
In addition, it would kill a lot of FUD out there.

Sco is just trying to "create " IP infringements out of thin air .. now that its clear that nothing is going to harm the GPL movement. It looks like these are the last groans of a dying animal called SCO

It's the time of the year .... everybody has to announce their numbers for the stock market.

SCO is just pumping them a bit for two weeks, then they can go back to their little nest and think of a new strategy to enervate then Linux community.

Give them a week, it'll pass away ....

Could just be a coincidence, but, as of right now, http://www.caldera.com/support/docs/openlinux/1.3/english/intro.html is giving a document not found. Google cache and archive.org have copies, though. :) *Save*

*Snip*
>their "succssors in interest" have been releasing ELF >support under the GPL for years: From >http://www.caldera.com/support/docs/openlinux/1.3/english/i>ntro.html
>"OpenLinux includes support for libc5, ELF, and a.out >binary formats."

Hmmmm... that's interesting. As pointed out in the link below, (which has a link to the published ELF specification dated MAY 1995, as well as a link to the Asset Purchase Agreement which wasn't signed until SEPTEMBER 1995.) OldSCO hadn't even bought ANYTHING from Novell yet, so how could NewSCO claim ELF at all? Evidently the "author" of this "article" did't even bother to do any fact checking, just regurgitated a steaming pile of lies. Nice work.

http://www.groklaw.net/comment.php?mode=display&sid=20040712104852353&ti...

SCO makes me feel ashamed to live in Utah.

Ummm, isn't linuxworld completely ignorable now because it's owned by IDG, a shameless paid Microsoft shill that doesn't really care much about facts or journalism because they are paid hacks? That was my understanding. Why are you all wasting your time here? There are real news sites out there.

I did some of the Elf work in SCO (the old one) Open Server.
I have 2 factual observations.

1 The comparison between ELF and Microsoft DLLs is quite valid. Elf defines amongst other things a standard mechanism for shared libraries. Shared libaries map closely onto DLLs

2 ELF is NOT derived from COFF. They are very different.

I earn my living developing on Linux, and have done for the last 7 years. I have found most of the comments on this discussion to be childish and would appear to be the work of wannabees who most likely have not written a line of professional code.

IBM should just buy SCO and dissolve the company.

Will someone PLEASE put a stop to SCO
these folks spend more time dreaming up ways to
hurt the FREE SOFTWARE FOUNDATION then they do
on creating new products. It is time for SCO to go
I hear there are several ways to build fertilizer
bombs. will somebody please park one of those vans
in the SCO parking lot so that this can all be done!!

Linuxworld,

Um... so this is why SCO has effectively told the Utah court, in filings you've obviously ignored....

1. SCO can't prove copyright infringement in Linux and thus can not refute IBM's 10th Counter Claim (IBM's use of Linux does not infringe... etc..). I'll let you look it up.

2. See 1, therefore they are requesting more discovery.

SCO had no TRUE allegations since day one, they don't know. All they had was a MS bankroll.

just a comment

-

hmmmmmmm.... i dont know if i should cry or laugh. Is SCO out of business or the just dont know what to do with their time? They must be crazy. I still dont understand what was going through their minds. I think they enjoy wasting time and resources. They forget linux plays a major role in todays industry and so its not going to be easy get away with this crab they are trying to put up..............

Have you heard of RESEARCH? Perhaps you should do some before exposing us this type of crap.

This article is as lame and poorly written as SCO's legal documents. If you would bother to read more than SCO's self-serving press releases, you would notice they are trying like hell to avoid any mention of copyrights, and trying to change their stories in the different courts at every turn. Look for their first major butt-kicking tomorrow (7/20/04) when Daimler-Chrysler hands them their head in the hearing.

This analogy between a virus and the GPL is completely false. A virus whether a computer virus or biological does not ask its host for consent when it invades. If a software developer includes a GPL'd work in their product, they are required to use a GPL compatible license. If they not do want to do so, then they can simply not incorporate the GPL'd work.
This issue of consent is the crucial difference.

The GPL is just a license. If I run Windows 2000 Server, Microsoft wants my money. If I run Samba, the samba developers want to be sure that changes to samba that I make AND DISTRIBUTE will be available to improve samba. That's great for me, my colleagues down the street and also for my competitors who may be using samba. It's as though the samba developers aren't content to just give away great software-- they want to give me better software later. Is that what you mean by "viral"?

The TIS Committee grants you a non-exclusive, worldwide, royalty-free license to use the information disclosed in this Specification
to make your software TIS-compliant; no other license, express or implied, is granted or intended hereby.
The TIS Committee makes no warranty for the use of this standard.
THE TIS COMMITTEE SPECIFICALLY DISCLAIMS ALL WARRANTIES, EXPRESS AND IMPLIED, AND ALL LIABILITY, INCLUDING
CONSEQUENTIAL AND OTHER INDIRECT DAMAGES, FOR THE USE OF THESE SPECIFICATION AND THE INFORMATION
CONTAINED IN IT, INCLUDING LIABILITY FOR INFRINGEMENT OF ANY PROPRIETARY RIGHTS. THE TIS COMMITTEE
DOES NOT ASSUME ANY RESPONSIBILITY FOR ANY ERRORS THAT MAY APPEAR IN THE SPECIFICATION, NOR ANY
RESPONSIBILITY TO UPDATE THE INFORMATION CONTAINED IN THEM.
The TIS Committee retains the right to make changes to this specification at any time without notice.
IBM is a registered trademark and OS/2 is a trademark of International Business Machines Corporation.
The Intel logo is a registered trademark and i386 and Intel386 are trademarks of Intel Corporation and may be used only to identify
Intel products.
Microsoft, Microsoft C, MS, MS-DOS, Windows, and XENIX are registered trademarks of Microsoft Corporation.
Phoenix is a registered trademark of Phoenix Technologies, Ltd.
UNIX is a registered trademark in the United States and other countries, licensed exclusively through X/Open Company Limited.
* Other brands and names are the property of their respective owners.

I think that SCO is yet again talking out /dev/null. I guess they can, they own it... or so they claim.

JFS comes from OS/2 not AIX and even if it did, you can't taint IBM's invention of JFS by compiling it for UNIX.

ELF and ABI have already been shown to be untainted IIRC.

"viral" need only be applied to proprietary licences. If you see Microsoft's Shared Source you are tainted and any work you produce from that point forward is a potential theft of Microsoft's IP.

I'm just waiting for the day my eyes, when used in conjunction with my brain are considred a potentialli infringing technology. When coupled with my hands and my mouth they certainly are.

They are right you know. SCO did create it and Linux stole it. That's not all either. Linux stole the idea of executing code on x86 processors from SCO. Oh let's not forget using RAM, SCO had that one first too. As a matter of fact the hole idea of Linux is just a rip-off of SCO concepts and IP.. keyboards, monitors, harddrives, charachter devices... where will it stop?

Hmmm, didnt groklaw do all the reports on the sco vs ibm case ?

also i thought journalism was supposed to be objective, so using "viral license" seems to me a subjective thing.

Is it me or has Microsoft paid another person/company to do their bidding. The article makes any person that doesn't know enough about Linux or Unix believe that Linux is infringing upon SCO IP. Also the author of this article needs to go back to school. The wording of various sentances, the slanted view of the article and the odd words that are used (like "sorta") makes one bad article.

All fecal releases I have seen are *substantially similar* to my fecal releases. I claim the shape/form/ip of all fecal releases as my own.
I demand that all SCO persons either pay me to crap , or STOP!

Maureen O'Gara clearly has a tenuous grasp of computer technology amd zero understanding of copyright law. She also clearly does not check facts or statements for their veracity. In short, her drivel is not worth reading.

SCO is/has been well aware of ELF in Linux: theyand their "succssors in interest" have been releasing ELF support under the GPL for years:
From http://www.caldera.com/support/docs/openlinux/1.3/english/intro.html
"OpenLinux includes support for libc5, ELF, and a.out binary formats."