Slashdot Mirror
Analysis of SCO vs. IBM
icantblvitsnotbutter writes "An excellent -- and clear! -- article over at LinuxWorld.com has a multipoint analysis of SCO's 40-page complaint (this is a brief?!). For all those IANAL's out there, here's something to sink your teeth into. On the balance, the outlook seems positive for IBM. Still, the parallel invocation of a contractural clause potentially nixing AIX lends some credence to claims that this is a just way for SCO to coerce IBM into buying them out..." Some old documents from a similar lawsuit have surfaced, and naturally ESR has his own take on the case.
Looks like it was written by a high school student, but none the less, what a nice brief. Dont those SCO People work hard?
Plus, their case doesn't hold water because, the SCO that we're talking about is not the same SCO as the SCO that provided Unix, after AT&T.
This SCO is caldera. The old SCO is SCO. Do the math, Caldera != SCO. Therefore, I do not see their grounds at all.
If SCO is just about to sell out, and is attempting to inflate it's 'intellectual property equity' in the hope that someone with money may buy them? Or maybe there are already talks underway?
If not, this appears to be a horrid act of desperation.
Averment 41
Shared libraries are by their nature unique creations based on various decisions to write code in certain ways, which are in great part random decisions of the software developers who create the shared library code base.
It's interesting to note that SCO considers the decisions of programmers to be basically random.
Reading the Linux world article SCO claims that Linux advanced features such as failover SMP etc could only come about after many years of development. However I do find it a bit narrow sighted in that they think that IBM is the only one who could procduce this software and port it. There are other blue chips out there who have written failsafe software and ported it to Linux. Personally I think SCO is talking rubbish (well at least on this point at least). I really need to read all 40 pages of their document but don't really have the time. (heh who does :)
Rus
Cheap UK and US VPS
The real reason is that SCO is dying, and wants to be bought out by IBM, thereby knocking up the final share price for their investors.
Got it?
[FUCK BETA]
I know it is the "hip" and "cool" thing to rag on ESR for his views on pretty much anything, however this brief is a very well written document and worth your time to read it. Whatever his precieved faults, he is able to put this issue in clearer prospective for me than the the original posting did.
I just finished reading the brief. I must say that for the first half or so, I was very impressed with it. It was simple, logical and factual. However, by the end, it seemed to devolve into a statement of beliefs and feelings that, to me, did not feel right in a court brief. For example:
SCO's complaint, in all its brazen mendacity, is the last gasp of proprietary Unix. The open-source community and its allies are more than competent to carry forward the Unix tradition. We pray that all assertions of exclusive corporate ownership over this tradition be given a swift and merciful end.
Am I the only one who thought that this was not the forum for such OpenSource flag-waving?
Linux IT Consulting and Domino Development in Michigan
This alone, from SCO's complaint:
Averment 86: It is not possible for Linux to rapidly reach UNIX performance standards for complete enterprise functionality without the misappropriation of UNIX code, methods or concepts to achieve such performance, and coordination by a larger developer, such as IBM.
I hope the rest of their case shows the same degree of arrogance and technical ineptitude. IBM would have little to worry about.
Want to Know How to Cheat the GPL? Read On!
It's well and good to say that ext3 or Reiser beats out JFS, which is probably true, because all installations of linux are going to use those things, including the hacker's box sitting at home. However, not everyone has the time, money, nor interest to develop those areas of Linux that need to be developed in order for it to compete on the level of enterprise class servers. IBM sells those products, and thus they contribute to Linux in ways that a desktop user might not quite see.
Maybe when I'm running a wackload of processors at home on a huge rackspace (when I own my own island too), I can try and contribute to scalability or some such other thing, but until then, IBM fits the bill perfectly.
One thing they missed is the whole BSD "fork" of UNIX. Back in the old days, AT&T gave pretty much every University a site license. One of these of course was UC Berkeley, where they did some some pretty core work, such as this unimportant thing we call TCP/IP. For a long time folks referred to TCP/IP as "Berkely sockets". Hmm, they also gave us C-shell, so maybe they shouldn't get off unpunished. BSD got folded into the "one True UNIX" in SVR4.
As far as "getting up to enterprise grade" speed comparisons goes, they're at some level irrelevant, at least in the way SCO framed it. UNIX vendors had to blaze trails, find their way, mess up, and find the true path again. Linux followed these trails in the form of POSIX. It's much quicker to code to a spec than have to code and lay out the spec, distribute the spec as it changes to your development team, and see how well it integrates after all at the same time. Having the spec also makes all that super coordination "magic" that SCO was talking about seem a little less fairy tale-ish. Lay out a spec, give parts to different people, have them code, and then at the end it all comes together. Who else has been burned by Linux' version of select()? It's coded to the BSD select() man page, unfortunately BSD select() isn't and neither is any other commercial vendor, and Linux select() is not bug compatible with other select()s. There's your miracle for you, the magic of troff. (at least it's not info pages, god no...)
BSDi also had a commercial UNIX, BSD based of course, on x86. I did some work on it. At the time wasn't enterprise ready, at least the version I worked on (the SMP implementation was pretty basic, the kernel was forced to run on a single processor) but in later years it got better and in fact I thimk FreeBSD SMPng is based off ideas from BSDi. My non-lawyer mind wonders how many statements of "fact" in the complaint have to be shown to be false before the case can be dismissed.
On other IANAL notes, I wonder if IBMs law firm is reading these posts and all the other arguments on the other web sites and saying "yeah, thats a good point.. yeah, write that one down."