Slashdot Mirror

LSB's fee schedule points to the TMLA for the list of countries that have to pay an extra $2,200USD. The TMLA is the Trademark License Agreement which was published in the U.K. by The Open Group in January of 1998. This document, which even has it's own fricken ISBN, is basically a contract that protects the UNIX trademark. Interestingly enough licensing a unix from someone like AT&T doesn't give you the right to use the UNIX trademark.

The Restricted countries, currently only include Taiwan and South Korea.

FSG keeps a list of compliant distributions on its Web site.

All of the certified distros are commercial products. Where are the community distros in all of this?

Could it have something to do with the Fee Schedule? The fees don't seem that steep.

The only thing i complain is, there should be a CTL and Pango enabled binaries available for linux for people viewing indian language sites(UTF-8 encoded).

On the other hand, there is Test Environment Toolkit that noone seems to use. And STAF which requires a huge investment of time just to comprehend.

So, question: what tools do people find useful for build/test automation with C++ ?

http://www.opengroup.org/certification/idx/posix.h tml

6. Mac OS X has full Unicode support in the shell, it's one of the reasons why they switched to bash. Just today I rsynced some font files and noticed that the names of the Chinese fonts displayed correctly as pictograms. So, it's a pretty safe bet that my OS X system is correctly set up for internationalization purposes. Yet OS X does not suffer from broken grep, even with the shell set to en_US locale with full Unicode support.

The message you cite from the Debian list explains that collation rules in en_US are different. The other message you cite is also talking about sort order. Neither deals with regular expressions.

POSIX.1 explicitly states that range operators are only defined for the POSIX locale, which is the C locale. Behaviour of character ranges in other locales is unspecified. As such, I think it is perfectly reasonable to expect grep's regular expressions not to suddenly change their meaning in different English-language locales, particularly when you're talking about a construct as common as [A-Z].

The bash NOTES claims that POSIX.2 requires the broken behavior. That may be true, but every system I use takes the sensible route of not making people's scripts break if the user's locale has different collation, except RedHat.

So the best you can say is that RedHat is Broken As Designed, so as to be slavishly compliant with POSIX.2.

I also note that the bash developers have clearly been struggling with the issue of whether to follow POSIX.2 or take the sensible route, as they went from normal behavior to POSIX.2 to normal and then back to POSIX.2 again in the 2.02..2.05 release stream.

Interestingly, Gentoo's grep man page documents that it has the broken POSIX.2 behavior, but it doesn't actually seem to.

9. If you set up your system to run both editors on the same file at startup, that would be analogous. And yes, it would be stupid. As would having two web servers installed by default and having whichever one happens to be first in the init sequence "win", or having three MTAs. If you don't understand why, let me provide you with a few reasons:

a) More code on the system means more code to keep patched, more security holes.

b) You see the iptables script is in the current run level, so you go edit the iptables config. Whoops, except ipchains runs first. That's the sort of thing that can waste hours of a person's time.

c) ipchains is obsolete. If you're doing a clean install (which I was), why the hell install obsolete software which will never be used at best, and at worst might actually cause me to waste my time or even suffer a security problem?

Under this analogy, wouldn't female dogging about deviations from an English "standard" correspond to female dogging about deviations from specifications such as Single UNIX, LSB, or GNOME HIG?

This topic is probably so old that no one is reading it any more, but I just came across an old note that I left to myself about this document summarizing the differences between POSIX and LSB

a company back in profit, with billions of financial assets:
SUNW Earnings Per Share: -0.119

4Q Earnings after $1.9 billion from MS: $795 million

Regarding 4Q profit:
"Excluding one-time items, the network-computer and software company lost $169 million, or 5 cents a share" - CBS MarketWatch

Java, which is now the most widely requested used development language
My understanding is that Visual Basic is still the most widely used development language. Unless... perhaps you are confusing Java with C#?

Linux is a superb system, and deservedly successful, but its hardly revolutionary.... just a damn good implementation of Unix.
Linux is not an "implementation of UNIX(TM)" according to The Open Group, who owns the UNIX(TM) trademark and certification.

UNIX is the actual operating system (which Linux has made a very powerful and capable clone of). It could be OS X, Solaris, AIX, *BSD or whatever.

I do agree with you though about the GNU software. That is what makes a good Linux/*BSD system.

Within the google search I fonud Developing LSB-certified applications http://www-106.ibm.com/developerworks/linux/librar y/l-lsb.html maybe this is the subject we should be discussing, how many applications ARE following LSB guidelines?

Also, why is the Linux Standard base hosted by the open group? Isn't the Open Group doing it's part already? so the other way arround becomes less relevant. ;)

http://www.opengroup.org/lsb/cert/cert_prodlist.tp l?CALLER=index.tpl

UNIX® describes any operating system sold under a brand licensing agreement with the Open Group. This requires the product to pass a checklist that includes certification to the Single UNIX Specification (free reg. req.) on a given set of supported hardware, based in part on product testing, and payment of brand fees pursuant to the Trademark Licensing Agreement (PDF). Often these brand fees are high enough to shut out publishers of low-volume operating system products.

UNIX® describes any operating system sold under a brand licensing agreement with the Open Group. This requires the product to pass a checklist that includes certification to the Single UNIX Specification (free reg. req.) on a given set of supported hardware, based in part on product testing, and payment of brand fees pursuant to the Trademark Licensing Agreement (PDF). Often these brand fees are high enough to shut out publishers of low-volume operating system products.

UNIX® describes any operating system sold under a brand licensing agreement with the Open Group. This requires the product to pass a checklist that includes certification to the Single UNIX Specification (free reg. req.) on a given set of supported hardware, based in part on product testing, and payment of brand fees pursuant to the Trademark Licensing Agreement (PDF). Often these brand fees are high enough to shut out publishers of low-volume operating system products.

UNIX® describes any operating system sold under a brand licensing agreement with the Open Group. This requires the product to pass a checklist that includes certification to the Single UNIX Specification (free reg. req.) on a given set of supported hardware, based in part on product testing, and payment of brand fees pursuant to the Trademark Licensing Agreement (PDF). Often these brand fees are high enough to shut out publishers of low-volume operating system products.

UNIX® describes any operating system sold under a brand licensing agreement with the Open Group. This requires the product to pass a checklist that includes certification to the Single UNIX Specification (free reg. req.) on a given set of supported hardware, based in part on product testing, and payment of brand fees pursuant to the Trademark Licensing Agreement (PDF). Often these brand fees are high enough to shut out publishers of low-volume operating system products.

I live in the wrong country so I don't think that I can do it. Someone like Red Hat would do nicely (especially if the The Open Group group pitched in).

'Course, I can't seem to find anything that says Solaris is Unix certified, either.

Other than the 14 Solaris entries here? That's not counting the 8 for CDE compliance.

Noticably absent are any Linux / Apple entries. Not that this really means a damn, but it's easy enough to check your story.

It does. Towards the end it reads:

The document is available online , along with a form that can be used to "sign" it.

2> The declaration stinks of pointy haired people sitting in afternoon long meetings. Not sure if I disagree here. Its form reminds me of these stuffy CEPT resolutions I have seen quoted in the old Radio Amateurs Handbook. On the other hand, there are only so many ways such a declaration could be formatted. And isn't the content rather more important, than the form being a literary masterpiece? Besides, the audience may include said PHBs.

3>. IBM now is not the same as IBM was 25 years ago. They probably sees this as relevant to their future business opportunities. They indicate as much, at any rate. And as they are more concerned with selling hardware and services; if they can profit from open standards then they will support that.

Which does raise the question of how much more than an IBM PR thing this really is.

It does. Towards the end it reads:

The document is available online , along with a form that can be used to "sign" it.

2> The declaration stinks of pointy haired people sitting in afternoon long meetings. Not sure if I disagree here. Its form reminds me of these stuffy CEPT resolutions I have seen quoted in the old Radio Amateurs Handbook. On the other hand, there are only so many ways such a declaration could be formatted. And isn't the content rather more important, than the form being a literary masterpiece? Besides, the audience may include said PHBs.

3>. IBM now is not the same as IBM was 25 years ago. They probably sees this as relevant to their future business opportunities. They indicate as much, at any rate. And as they are more concerned with selling hardware and services; if they can profit from open standards then they will support that.

Which does raise the question of how much more than an IBM PR thing this really is.

"Nice to see you snipped the bit about The Open Group suing Apple for infringement of the UNIX trademark."

I thought another poster answered that quite nicely actually. The case is over. It is no longer in Apple's SEC filing which means that either Apple is in some VERY serious trouble with SEC or the case is over.

Come now, that's utter rubbish. There's no mention of it in Apple's May-2004 10-Q, but neither is there any mention of it in Apple Feb-2004 10-Q, nor probably any other. Given the miniscule size of the licensing fees ($US100k/year), this is hardly a surprise.

For a more realistic view, the TOG press articles page contains a link to an osviews.com article from April-2004, in which TOG's Vice President of Marketing explains Apple's infringement.

"The only reason this thread has continued is because you won't admit that a registered trademark is legally valid until overturned in a court of law"

That's because that statement is incorrect. A trademark is invalid upon the reason it is later overturned existing/occuring.

Just as a murderer is a murderer when he kills the victim, NOT when the jury reaches a verdict.

Nonsense. Under the law, a murderer is innocent until proven guilty in a court of law. Under the law, a registered trademark is valid until invalidated in a court of law.

"Interestingly, you haven't provided any links"

I haven't said anything which required a link to back it up? This isn't a court of law, this is a news forum on which I share my opinions with you out of the kindness of my heart. It's not a court of law, and it's not a slashdot STORY posting. Including redundant links doesn't score you any points.

The open group filing suit against Apple is not a conviction and is meaningless.

I've never said the open group didnt have a registered trademark on the word unix. I said that trademark is of questionable validity and that even if it weren't, the open group certification is not what determines whether a system is unix in the common english usage of the word.

In misinformed usage, yes, but the fact that ignorant PC users refer to their computers as 'CPUs' doesn't mean they are in fact CPUs, and incorrect usage of 'UNIX' doesn't mean non-UNIX systems are actually UNIX.

Correct usage refers to Linux as Linux, BSD (including Mac OS X) as BSD and UNIX as UNIX. Their are subtle differences between their APIs, and at the base level, these difference identify each group. I'd love to see all Linux and BSD systems certified as UNIX, but unless and until they are certified, there's no guarantee they conform to any standards, and they simply aren't UNIX.

You sound oh so very high minded about standards and integrity, but you reveal your trollish prejudices.

For the purposes of our discussion, the only Unix standard that matters is POSIX compliance. Hmmmm, can you think of any POSIX compliant OSes that don't have the Unix (TM) on them?

'POSIX' is also a registered trademark, and must also be certified by and licensed from IEEE/TOG. Moreover, the IEEE charge for the POSIX standards, where as TOG freely distributed the Single UNIX Specification. This is one reason the SUS has largely superceded POSIX, especially for open-source developers.

In any case, I doubt Mac OS X is certified as POSIX compliant either. Apple don't seem to be very keen on certifying their systems with TOG.

TOG's problem is that Apple is only one of many who might want to challenge the Unix trademark's viability in court. So the trademark might not fall this year, as you say, but there's always next time.

That isn't true. Apple call their system 'UNIX', but more ethical firms, e.g. Linux firms like Red Hat, SuSE/Novell, etc., don't make that claim, and the commercial UNIXes (e.g. Sun Solaris, IBM AIX, IBM Z/OS, HP Tru64, HP-UX) have been duly certified.

The only system besides OS X I know of that has used 'UNIX' without certification is FreeBSD. However, the FreeBSD website was changed recently, and no longer claims FreeBSD is 'an advanced UNIX operating system', but claims instead it is simply 'an advanced operating system'. Moreover, the FreeBSD organisation have been working with TOG to achieve verifiable compliance with POSIX (with UNIX certification a likely follow-on). Just last month, the FreeBSD organisation received approval from IEEE/TOG to use material from the IEEE/TOG POSIX/Base standards. This is the sort of thing Apple would be doing if they really cared about standards and UNIX.

I'm not sure why you paint me as being cynical just because I have an inkling of how things work.

It's very simple, really. Apple want to claim their OS is UNIX (even using a UNIX logo similar to the TOG UNIX logo), but won't run the tests and pay the small licensing fee (US$100k) to ensure it's compatible with UNIX. They clearly don't give a toss about standards, unlike, for example, IBM, Sun and HP.

The idea that destroying standards for trivial sums of money and minor convenience is just 'the way things work' is an attitude that I consider quite cynical. I still have hope that Apple will come round (by choice or by the force of the law) and submit OS X for certification. If Apple destroy UNIX, they will be despised by many (including me).

Or maybe you do know how things work, but you astroturf for SCO/MS.

This is actually one reason I'm carrying on with this thread. A lot of people who despise SCO (with good reason) seem to think TOG and the UNIX trademark have something to do with SCO.

The reality is the UNIX trademark and The Open Group have absolutely nothing to do with SCO or SCO's legal claims, and never have. SCO claim to own copyrights to a particular set of source code called UnixWare, which they purchased from Novell, who in turn purchased it from AT&T. (SCO also claim their code is in Linux, but comparisons have shown this is extremely unlikely.)

Like all systems that have been certified to comply with the UNIX standard, SCO UnixWare (originally called UNIX System V) can be called 'UNIX', but it has no special position with respect to the UNIX trademark, which is solely owned by The Open Group. Moreover, SCO's licensed use of the 'UNIX' name would be revoked if UnixWare ceased to comply with the standard (the same applies to any UNIX vendor).

Microsoft have nothing to do with The Open Group either, and in fact their UNIX subsystem (Interix) for the NT kernel was certified by Softway Systems in 1988, before it (Interix) was acquired by Microsoft.

"Finding a non-Adobe Acrobat reader that is not itself the result of a violation of an Adobe IP claim [...] is a bit problematic."

No it's not! Adobe has deliberately made PDF a public format; they freely distribute the specs and encourage others to support the format. Finding (e.g.) xpdf is not a bit problematic - there's barely a Linux vendor out there who doesn't ship it. Even The Open Group (the guys who own the UNIX(tm) trademark) have an xpdf page. Getting it to run on your platform might be problematic if you don't run a Unixlike system, but that says more about that platform than about the format.

POSIX is a set of interoperability standards and is protected by US copyright law.

UNIX is a brand name owned by the Open Group, and is licensed for software that passes their certification process.

The entire point of POSIX and UNIX is that there is a standard interface, but differing implementation. Individual developers (including Linus) and the FSF own the copyrights to the code of the kernel and userland. Linus owns the trademark "Linux," but there are not patents on UNIX or POSIX.

If AT&T couldn't play the patent card when they sued Berkeley (BSD Unix), no one will or can now.

Sit back and enjoy your open code...

POSIX is a set of interoperability standards and is protected by US copyright law.

UNIX is a brand name owned by the Open Group, and is licensed for software that passes their certification process.

The entire point of POSIX and UNIX is that there is a standard interface, but differing implementation. Individual developers (including Linus) and the FSF own the copyrights to the code of the kernel and userland. Linus owns the trademark "Linux," but there are not patents on UNIX or POSIX.

If AT&T couldn't play the patent card when they sued Berkeley (BSD Unix), no one will or can now.

Sit back and enjoy your open code...

Not that this is a POSIX call. Here is an example of its use in GnuPG. They use it for security reasons.

Nowdays though, my definitive reference for writing portable unix programs is the merged IEEE POSIX and Open groups's Single Unix Specification. Registration is free.

You can just allocate a data page, decrypt to that, then change the protections on the data page to turn it to code (mprotect(3) in *nix, VirtualProtectEx in Windows). Y'know, just like the hoops a JIT has to jump through. Essentially, you're using the same argument as people who say, "Man, this NT stuff is harder than 95. I'll just grant Full Access to Everyone in the installer."

SCO now claim that Linux is a 'nonliteral implementation' of Unix, and compare their claim to those involving Harry Potter rip-offs and Vanilla Ice versus David Bowie and Queen.

To quote Jon Stewart: "Whaaa?!"

Have they heard of something called POSIX?

I think more accurately it should be called the 'C' desktop.

Yes, because that wouldn't cause any confusion with an existing product either.

who after this will claim they own UNIX?

1. The Open Group still owns UNIX (titled "Backgrounder on the UNIX System and SCO / IBM legal action");
   
2. The SCO Group (TSG) owns UnixWare and its other derivative products, and
   
3. judging (IANAL) from the TSG v Novell court documents, Novell owns the copyrights to Unix System V and never transferred them to (old) SCO.

If you accept the above, and if you consider as well that Novell has told TSG early on that TSG is wrong about the copyrights, and that IBM knew Novell was doing this, you also know why IBM didn't buy out TSG.

That would make the answer to the more appropriate question who will claim to own Unix System V as simple as this: Novell.

Fortunately, JPL has been working on the MSL for quite some time. "Long-term" implies years; MSL - in addition to solar power - will utilize two small nuclear cores, which are currently under design. And "long-term" leads to a need for a WELL-DESIGNED software infrastructure. For this reason, the work on the Mission Data System has been intense. The decision on which of the competing MDS implementations to use is scheduled for 2005. One of the possible implementations is real-time Java.

OS X is not UNIX. It may be UNIX-like, but it is not UNIX. The OP claimed equality that OS X is UNIX, which is not true. Your assertion that Unix is a marketing brand (UNIX) and not a design philosophy (Unix) is, for all practical purposes, crap.

The last time I used OpenServer, it was CDE. However, it's been a while. (2000? 2001? My memory is vague on this.)

I imagine by now they've gone with something else, with corresponding differences in licenses.

As to which X server back then, I don't recall. XF86 now, I'm sure.

Really??? What's your UNIX distribution??? Solaris??? AIX??? HP-UX???

You know I was talking of Linux... and... until SCO is proved wrong, it is UNIX :)

Um, no. It is not. UNIX is a registered trademark of The Open Group. They certify products as being compliant with a known standard. Linux has not been certified, and is not even compliant. Until it is, you are confusing the issue and infringing on a trademark.

Now for the flamebait part of the post: Yes, it's nice that the kids can play on their little OS at home, but please don't confuse it with UNIX.

read much yourself?

I still don't get where the problem is. As I read it, these restrictions apply to redistributions of XFree86 itself, not to 3rd-party programs that link with X libraries. They can't reasonably restrict programs that link with Xlib, Xt, Xaw, etc., because these components can easily be replaced at run-time (with dynamic linking) or compile-time with any other distribution of X11R6 without modification. That's because X11R6 is a standardized protocol, and the Xlib C language binding and Xt toolkit are also open standards.

Since Sun bought "rights" to UNIX from SCO some time ago, I think they can call it UNIX. Otherwise it would be Sunix and the President would need to change his/her last name to Sunis :P

Okay, tell me why SunOS is "real UNIX" but BSD isn't...

On the off chance this is a serious inquiry, SunOS is officially "branded", *BSD and Linux are not. Said brand is required to be a "real UNIX", costs $$$ to obtain. Vaguely recall reading that some Linux distro was going to try for this. Haven't heard of them in quite a while now (iirc it was with kernel 1.2!).

No, because SCOldera doesn't have a clear, uncontested title to "Unix". First of all, the Open Group owns the trademark "UNIX(tm)," secondly, Novell has a competing claim to the code, and finally, large parts of the code are owned by others - most notably the Regents of the University of California.

Caldera/SCO is also the target of more lawsuits than they've filed. They've already lost a suit in Germany, and been additionally fined for violating a court order, and they're under investigation by the Australian government. Doesn't make them look like a very tempting aquisition.

But most of all, yes, MS in particular (who might otherwise be willing to overlook some of these obvious problems) would be in a world of trouble due to their monopoly position. The SEC could and probably would forbid such an aquisition. And remember, US law isn't the only thing they have to worry about. MS also does business in the UK and EU, and would have to deal with major repercussions there as well.

And finally, there are change-of-ownership issues in the contract between Novell and oldSCO. NewSCO seems to think they can sidestep this by claiming that they merged with oldSCO's OS division, and thus the ownership didn't actually didn't change. MS would have a much harder time making such an argument.

There is work to do just that: Java on top of RT/Linux in the Rocky 7 platform (basically the testbed for MER1/2). See here (PDF file.)

Now the question I have is... which filesystem did they use for MER1/2? Is it DOS FAT? If so, I could see how the "too many files" problem could happen quite easily.

Not that DOS FAT wouldn't be OK for spaceflight, it is very simple, reasonably robust and quite mature. Just gotta watch for those FAT limits.

You could already have got that information from the OpenGroup web site. This has always been a useful companion to the Linux manpages when for example one needs to be exactly sure about when a particular error will occur.

and it was but a mere shell of what it used to be.. at least I got to support the FSF by buying a nice-looking "Free Software/Free Society" t-shirt. Other than that, got the google pen and a good look at small displays that would be perfect for replacing my in-dash GPS display screen.. looking for one with multiple inputs. Oh, and I finally got a replacement Linux license plate. The one on my car now has "Compaq" in huge red letters on it, which isn't too cool as an IBM consultant. :) The new one has "OpenGroup" up there now.

Oh, and my vote for the most mis-guided individuals who have no idea how to make the conversion to Linux-for-business: the VOIP people who ran their setup on a chipped X-Box. Are you kidding me, people??? You want a business to buy your product, and you power your display with a video game console? The coolness factor drops way the hell off when you're trying to sell VOIP solutions. Build a damn PC. Jeez.

Anyone remember the IBM party in 2001? They rented out the whole upstairs, had an open bar, great music, a real BattleBots cage, and, well, an ice sculptor. But he cut out a damn fine Tux, too. THAT'S what I think of when I think about the days before the bust.

SCO Files Slander of Title Lawsuit Against Novell.

"The SCO Group, Inc., the owner of the UNIX(R) operating system [...]"

Really? The Open Group should talk to these people. Again.

Whatever. This just makes it more clear, SCOX is a litigation company owned by lawyers, not a technology company. There can be no risk of Novell being competing unfairly with SCOX when Novell clearly isn't a litigation company.

Okay I'll bite where is the proof that SCO's is the original file? Can you actually prove anything? Show me absolute proof which is the original? AT&T vs. BSD was lost because when it came down to specifics, Berkley had proof that AT&T had copied BSD not the other way around. Ever heard of posix? The Open Group Base Specifications IEEE Std 1003.1 ? The so called list of files are all header files from an open standard. Look at this link errno.h

It's but one of the so called offending files.

Now go back to jerking off and leave intelligent discussions to people with brains.

So, do you admit that Linus copied the entire substance of include/errno.h from somewhere else?

I dug out Unix 5th Edition because it's the closest to the original that I could find. Linus says that he copied his errno.h from Minix. Tanenbaum probably copied his from somewhere else. The chain of copying goes back to the original AT&T Unix code.

Compare with the paranoid attitude of gcc: gcc doesn't include copies of other people's system header files from, say, solaris or hp-ux; gcc includes scripts that manipulate the files on the end user's system, if the end user has legal copies of these files.

See this bit from the GNU coding standard:

Referring to Proprietary Programs

RMS has a stick up his a** about a lot of things, but better RMS's stick than Boies' dick.

I haven't read the Posix standard ...

Here is the 2003 version of the standard for errno.h.

The names are standardized in 2003 (and were probably standardized before 1991). But the values are not standardized. If you or I were to write an errno.h from that specifcation, the odds are incredible that we would happen to choose the same 31 values in the same order as the original Unix implementation. Personally, I would follow the order in the specification, which is alphabetical.

Again, from a legal point of view, it's a molehill. The actual value of those identical error numbers numbers is not $3,000,000,000. More like $300 or $3000. Ditto with the ioctl numbers. As you point out, the values of certain signal numbers (like SIGKILL) are well known. And all of these files are completely unrelated to SCO's claims that IBM copied SMP, NUMA, JFS, RCU code into Linux.

But from a PR point of view, it's damaging to us.
This activity looks uncomfortably like SCO's allegations about how Linux is developed. Here's Unix code; here's Linux code; the Linux code is a verbatim copy of the Unix code.

A nasty twist: because this is part of the ABI, it would be difficult and painful to replace include/asm-*/errno.h with a clean room implementation from the Single Unix Specification.

A note here: I'm not pro-SCO, and I'm not trolling. I'm looking to find the strongest SCO points because we're at war with SCO, so it's important to understand the strong points of the enemy's position.

This is why the Federal Information Processing Standards Publication 151-2 is important. The other "open" standards bodies don't hold as much legal weight as the federal standard, because the US Govenment requires all contributing parties to "sign off" on the standard. There are a number of Legislative acts past that mandated such open implementable interoperable standards for Federal Information Processing in Govt tenders ( just don't ask me to name them off the top of my head - It's well over a decade since I worked with Ada ).I'm sure that many Govt organizations would not look too kindly on the self claimed "inheritors of Unix" SCO Group attempt at such a U-turn.

For POSIX compatability all that is needed in reality is a semi successfull attempt at passing compatability and a roadmap. Microsoft got by for years claiming POSIX compatibilty in NT for federal tenders, dispite the fact the the native POSIX layer was not stable or trully functional until late 1998 with the release of Interix. Now The OpenGroup and the Linux Standard Base have a New RoadMap for providing a POSIX shell for Linux LSB.

Very well said, I totally agree. Being that the Open Group owns the UNIX trademark and sets the POSIX standards, I have to wonder if SCO might be stepping on their toes a bit with this. Not good. It'll be interesting to see if the Open Group gets pulled into this. AFAIK, Novell owns the copyrights and maybe a few patents, which it purchased from USL after the BSD case. The trademark (and OSF/1 and POSIX oversight) went to Open Group, and the codebase went to the "old" SCO.

'The SCO Group cannot expect to win any case based upon application interfaces which it's AT&T, USL and Novell predecessors relased in open standards specifically for the purpose of interoperability.

signal.h, errorno.h,and ioctl are all parts of many released standards including The Open Group and IEEE POSIX Base Specifications and the Federal Information Processing Standards Publication 151-2.

Note that The SCO Group does not own the copyrights on any of those standards and it does not own clear title to the copyrights on most of the AT&T Unix base.

From 1989, the then SCO activity pushed for the adoption of the iBCS Intel Binary Compatibility Specifications across *all* i386 Unix vendors

For the benefit of the entire user base, as well as the industry as awhole, SCO encourages all UNIX System vendors for Intel processors to join SCO, USL, Intel, ISC and OSF in supporting the iBCS-2 standard for x86 applications.

Even SCO admits, no requries these definitions to be present in order to be standards compliant.