Skip

MS + SCO + Linux...oh my...
March 4, 2004 4:43 AM   Subscribe

It comes as no surprise that Microsoft seems to be paying SCO to fight the fight. [via #joiito]
posted by yonderboy (23 comments total)

 
If SCO ends up losing the case then I hope there are a large helping of fraud and racketeering charges to go around. SCO has been very reluctant to disclose exactly what has been misappropriated which to me indicates that their case is pretty tenuous. It's a bit like a department store telling the police that a specific person was a shoplifter but not being willing to tell them exactly what was stolen. A claim like that should be met with a great deal of suspicion.

Microsoft's investment in SCO at the time seemed potentially dangerous. In the short term there were probably some companies who could be swayed into deploying on Microsoft products instead of linux. In the longer term Microsoft is jeopardizing their companies reputation. If the suit is actually determined to be fraudulent and it becomes common knowledge that Microsoft helped the suit along then they'll have done damage to their name.

So Microsoft knows that the case should be valid or they were misled (which I would also thing should result in a loss of reputation - a company with as much resources as Microsoft should be difficult to mislead) or a third possibility is that they don't worry about negative outcomes.
posted by substrate at 5:16 AM on March 4, 2004


Microsoft beat Unix last time by pointing out how Unix had many masters, and offered themselfes as the source of a consistant platform.

Given the over 200 seperate forks of GNU/Linux, this time the UNIX market is even more fragmented. And this time around, they have the additional point of how the "Linux community" as a general rule will refer to anything that touches the Linux kernel as "Linux". So suddenly seperate projects like Apache, Sendmail, PostgreSQL, MySQL, PHP are "a Linux Web/Mail server". Such confusion creates support headaches Microsoft can use *AND* the Linux 'community' isn't able to defend itself VS such a campaign.

As for SCO's claim of stolen code? I would bet there *IS* stolen code. However, its not worth what they are claiming. If you look in the 2.0.36 kernel, there is one chunk of code where the 'author' puts in a comment 'taken from FreeBSD', yet it only has a GPL licence, the FreeBSD/BSD licence was removed. The ATA code (per /.) and even Bruce Perens has pointed out code taken from BSD where the BSD copyright was REMOVED. (that makes it 'stolen code' folkes.) So, could some code with SCO rights ended up in the Linux kernel w/o observing copyright? Yup, because some 'authors' of Linux kernel code arn't authors, just thiefs.
posted by rough ashlar at 5:20 AM on March 4, 2004


If you look in the 2.0.36 kernel, there is one chunk of code where the 'author' puts in a comment 'taken from FreeBSD', yet it only has a GPL licence, the FreeBSD/BSD licence was removed.

If I'm not mistaken, isn't the BSD license (without the advertising clause) basically 'take what you want, just say where it came from' ? And as far as the Perens pointing, do you have a link for that? Besides that, I don't doubt for a second that there's borrowed code with dubious origins in Linux. Personally, I'd blame sloppy codewriting before outright theft. Some people feel even more strongly about it.

Of course, all this beating around the bush has given Linux developers ample time to find any stolen SCO code and rewrite it, if they weren't too stoned to remember where they put it.
posted by angry modem at 5:37 AM on March 4, 2004


I'd say linux would be far less likely to contain stolen code than proprietary OSs, since there they don't have to worry about people looking at it.

Executives trading in rumours to pump and dump stocks is nothing new, it's just sad that they are attacking an open source project this time.
posted by Space Coyote at 6:03 AM on March 4, 2004


If I'm not mistaken, isn't the BSD license (without the advertising clause) basically 'take what you want, just say where it came from' ?

All of the documentation and software included in the 4.4BSD and 4.4BSD-Lite
Releases is copyrighted by The Regents of the University of California.

Copyright 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
3. All advertising materials mentioning features or use of this software
must display the following acknowledgement:
This product includes software developed by the University of
California, Berkeley and its contributors.
4. Neither the name of the University nor the names of its contributors
may be used to endorse or promote products derived from this software
without specific prior written permission.

Removing #3 (the 'advertising clause') doesn't change #1,#2 and #4.

And thanks for the Spatula link.
posted by rough ashlar at 6:05 AM on March 4, 2004


I should RTFL :-)
posted by angry modem at 6:49 AM on March 4, 2004


I would disagree about the linux community being more fragmented than the UNIX market of old. It's true that there many more distributions of linux there ever were flavors of UNIX. However, from an application developer's point of view the various linuxes tend to be much more similar than the UNIX's were. The effort in porting an application from SunOs to Sys V could be substantial. Moving something from redhat to mandrake requires only a recompilation and upgrading supporting packages most of the time. Obviously, this isn't true for every application, especially stuff that is kernel dependent.
posted by rdr at 6:52 AM on March 4, 2004


The document below was emailed to me by an anonymous whistleblower inside SCO. He tells me the typos and syntax bobbles were in the original. I cannot certify its authenticity, but I presume that IBM's, Red Hat's, Novell's, AutoZone's, and Daimler-Chryler's lawyers can subpoena the original.

Feh. "I can't tell you where this comes from, and I can't prove it's authentic, and I am posting this on an anti-Microsoft and anti-SCO site - so it's gotta be true!"
posted by FormlessOne at 6:58 AM on March 4, 2004


More of a 'this may very well be true, but let's wait until a subpeona turns it up (or not),' FormlessOne. Of course, by the time any subpeona comes out it may be too late . . .

In any case, the judge told SCO that it absolutely must show the infringing code within the next 45 days. Should be interesting.
posted by Ryvar at 7:11 AM on March 4, 2004


Microsoft beat Unix last time...

lol. did I miss something? MS hasn't won anything! Their code is slower, has hundreds of security flaws and is less feature-rich than *NIX-based equivalents. If you are implying that they have won because they have made more money than the folks who are giving their software away for free, then, well, you need to get with the program.

This case is about dirty pool. MS wants nothing more than to own or own by proxy key IP from the *nixes. They want to break the GPL's validity in court and force a licensing structure on GPL'd OSes. They do this because they need to maintain their monopoly. They charge more and more for software that has diminishing value and they like things the way they are, thanks. Like I said, dirty pool.
posted by n9 at 7:44 AM on March 4, 2004


I would disagree about the linux community being more fragmented than the UNIX market of old.

I'll stake out the other side and point out how each GNU/Linux vendor has to figure out 'how to add value to the "standard"', just like the old 'Sure, we are SYS V UNIX, but we are better' days of the past.

It's true that there many more distributions of linux there ever were flavors of UNIX.

And that is a wedge issue.

However, from an application developer's point of view the various linuxes tend to be much more similar

And rather than focusing on, say, Apache being open source and running on many platforms, the focus by many of the chattering masses is 'Linux'. If the focus was "Open Source", when the boss asks "Isn't Linux in a legal suit?", the Open Source defender doesn't need to try to 'address the legal issues', but instead can say "Yes, and to avoid that we are using FreeBSD - based on code that had settled years ago." Tying Open Source to Linux is why Microsoft sees 'Linux' as a threat.

HARDLY ANY ONE USES LINUX. People care about Mozilla/Apache/Sendmail/GIMP/Open Office/et la. The Linux kernel is important, yet isn't needed as HURD/BSD/Mac OS/ or even Windows can run what people care about - applications!

If people would stop refering to 'anything that runs on a linux kernel as Linux', Microsoft can't fight a battle on the many fronts Open Source offers.
posted by rough ashlar at 7:44 AM on March 4, 2004


(please note that my comments refer to server software. It is generally not the case that MS code is less feature rich than OS code on the desktop.)
posted by n9 at 7:45 AM on March 4, 2004


Given the over 200 seperate forks of GNU/Linux, this time the UNIX market is even more fragmented.

There aren't 200 'forks' of GNU/Linux. There are 200 distributions that offer different packages and they're almost all built from a standard Linux kernel which is very much not forked.

So, could some code with SCO rights ended up in the Linux kernel w/o observing copyright? Yup, because some 'authors' of Linux kernel code arn't authors, just thiefs.

If you follow the case, and have an understanding of what SCO is suing IBM over, it'd be clear that code 'with SCO rights' was in the Linux kernel years before SCO could claim to have any infringment of it's IP, which is what this whole case is about.

The suit alleges several enterprise features of Linux--the NUMA (nonuniform memory access), RCU (read-copy update), SMP (symmetrical multiprocessing), schedulers, JFS (journal file system) and XFS (extended file system) portions--all include copied SCO code. Of course, they're missing the point that NUMA, RCU, SMP, JFS, and XFS, were all invented by other companies (IBM and SGI
mostly) many years ago, and they've now been 'gifted' to the linux kernel by those companies.

Thieves indeed.
posted by djc at 8:54 AM on March 4, 2004


Well, let's see, conspiracy theories abound as to the criminal mastermind behind this company killing exercise in legislative tail chasing.

I seem to remember something like this waaaay back when MS stole the GUI concept from Apple and went on to become the asshat behemoth they are today. Is the world a better place because MS has infiltrated every nook and cranny of anything even resembling a computer? No, its a dangerous place with nasty data deleting viruses, DDoS attacks as political protest and "critical" security updates from MS on a weekly basis. And all of this following their recent source code leak.

Even if SCO does have a claim, the world is actually a better and safer place by having some competition in the marketplace. Maybe it'll get MS to actually produce some decent code the first time out instead of the publish now and fix it later crap they pump out.
posted by fenriq at 9:51 AM on March 4, 2004


Moving something from redhat to mandrake requires only a recompilation and upgrading supporting packages most of the time.

Actually, it doesn't require that, almost all the time. Every Redhat package I've installed on my Mandrake desktop machine in recent years has worked without problem. I've rarely had to do that, since almost everything is available for both. For packages that do require modification, they just need simple changes in the configuration and the installation paths. Binary compatibility between distrubutions is pretty good, and getting better.

It's true that one sometimes has to upgrade other packages (like libc) to get new stuff running, but that's usually not due any incompatibility between distributions. They all converge on the most recent versions of system libraries, most of them at about the same rate for practical purposes.

The problem with "fragmentation" of UNIX was that companies would build all sorts of proprietary functions into their version that weren't supported anywhere else. This problem does not exist to any significant extent in GNU/Linux. All the things added by the major distribuitions are freely available for the other distributions to use. For example, the biggest incompatibility to be found is usually the package format. Since none of them are proprietary, almost all the distros can support all the major package formats.

rough a:the "Linux community" as a general rule will refer to anything that touches the Linux kernel as "Linux". So suddenly seperate projects like Apache, Sendmail, PostgreSQL, MySQL, PHP are "a Linux Web/Mail server".

That's just not true. Nobody in the Linux community with half a clue thinks of Sendmail as a "Linux" project. Just about anyone who would be working with mail servers is aware that its history is longer than that of Linux, and that it runs on all UNIX-like systems. And that Apache runs on Windows, too.

HARDLY ANY ONE USES LINUX.

Actually, many system administrators do choose to run Linux in part for the features that the kernel provides. The proc filesystem, high-performance networking, very fast process creation and context switch times, wide driver support, good memory management and multithreaded performance (in 2.6, anyway), easy extensibility, huge configurability, small memory footprint (when configured for it), decent realtime support (when patched for it), open-source code, ... etc. These things do make a difference, you know.
posted by sfenders at 9:56 AM on March 4, 2004


HARDLY ANY ONE USES LINUX.

'hardly anyone' being defined as companies like google.com, Pixar, Fedex, Merill Lynch, Dreamworks, Oracle, NASA, DaimlerChrysler, Reuters, Sony, Amazon.com, a plethora of government agencies and educational institutions, not to mention just about every major ISP in the U.S. But ya, other than that and the thousands of companies I couldn't think of off the top of my head, hardly anyone.
posted by djc at 10:10 AM on March 4, 2004


But ya, other than that and the thousands of companies I couldn't think of off the top of my head, hardly anyone.

Certainly not Mac Users...

(okay, FreeBSD is *technically* not a Linux, but still...anyone saying "HARDLY ANY ONE" won't know the difference anyways. More console candy, anyone?)
posted by Ogre Lawless at 10:25 AM on March 4, 2004


...the "Linux community" as a general rule will refer to anything that touches the Linux kernel as "Linux". So suddenly separate projects like Apache, Sendmail, PostgreSQL, MySQL, PHP are "a Linux Web/Mail server".

Bullshit. In my experience, Linux advocates (even the kids on Slashdot) are very careful about defining terms, and they would never, never conflate Linux with Apache or Sendmail, etc. They will go out of their way to point out that these are all open-source software projects, and examples of the success of the open-source development model.

HARDLY ANY ONE USES LINUX.

Doesn't Linux have like 40% of the web server market? I know it's rapidly becoming the standard for high-end scientific computing in academia.

Methinks our masonic friend doesn't know what the fuck he's talking about.
posted by mr_roboto at 10:47 AM on March 4, 2004


err.. he meant that hardly anyone uses Linux-based systems because of the Linux kernel, that they care only about the applications that run on top of it, which are not specific to Linux. Which is still wrong, of course, but not as wrong as taking that capitalized bit out of context would suggest.
posted by sfenders at 11:14 AM on March 4, 2004


Methinks that this "rough aslar" person should be a little more careful about the statements he makes, or else he'll look like he's just got his head up Darl McBride's arse...

Then again I'm sure "HARDLY ANYONE" will see it this way anyway.
posted by clevershark at 11:52 AM on March 4, 2004


Moving something from redhat to mandrake requires only a recompilation and upgrading supporting packages most of the time.

It seldom even requires that. Extreme case: I plunked a RedHat binary kernel + modules smack into a the middle of a Debian distribution and it ran like clockwork. (There were good reasons for doing this seemingly silly thing, believe me, or I wouldn't have done it.) Linux binaries are generally distribution-agnostic. The most you have to worry about is if you have outdated shared libraries, and even that isn't as difficult a problem as some that I've faced with Windows DLLs.
posted by George_Spiggott at 2:18 PM on March 4, 2004


More console candy, anyone?

Wuh? I'll have you know that at least two of my consoles run linux.
posted by inpHilltr8r at 4:04 PM on March 4, 2004


Let's assume:

Microsoft perceives Linux as a serious threat to their OS (near) monoploy.

Bill Gates may be a lot of things, but he's not stupid.

Funding the SCO suit buys Microsoft some time, but for what?...

Now here's the fun part. Suppose the top brass at Microsoft review the statistics and anecdotal evidence from Y1950 to Y2004 of high tech companies surviving a next wave / new paradigm challenge, and correctly conclude that their own long-term odds are not good. Suppose they then look at strategies of companies that have survived one or more of these waves (e.g. IBM), and they conclude that co-option is their best chance. Ladies and gentlemen, I give you...

Microsoft Linux

Remember, you (possibly) heard it here first.
posted by ZenMasterThis at 6:32 PM on March 4, 2004


« Older Wait till one breaks a record ...   |   Digital Utopia and its Flaws Newer »


This thread has been archived and is closed to new comments



Post