What Goes Around Comes Around - The Only Clear Winner in This SCO Versus IBM Case is Microsoft

PBS ^ | JUNE 19, 2003 | Robert X. Cringely

[Ernest_at_the_Beach]

Ashton was a macaw that lived in the lunch room at George Tate's software company, Ashton-Tate, home of dBase II, the first successful microcomputer database.  There is a lot about that long-gone company that was unusual.  There was the macaw, of course, which was named for the company, not the other way around.  There was George Tate, himself, who died at his desk when he was only 40, but still managed to get married two weeks later (by proxy -- please explain that one to me).  And later there was Ashton-Tate's copyright infringement lawsuit against Fox Software that pretty much destroyed the company when it became clear that Ashton-Tate didn't really own its database. NASA did, which meant that Fox had as much right to dBase as did Ashton-Tate.  All this came to mind this week while I was thinking (still thinking -- this story seems to never end) about the SCO versus IBM lawsuit over bits of UNIX inside Linux.  There is a lot SCO could learn from the experience of Ashton-Tate.

Those who have stuck with this saga recall that I earlier wondered whether SCO put those bits of UNIX System V into Linux themselves, whether they were scavenged from BSD UNIX into both System V and Linux, or whether the problem lay behind Door Number Three.

Door Number Three it is!  According to some of those who have had a look at the offending code, it DID come from IBM after all.  There are reportedly many lines of identical code, and at least some of the Linux code even carries an IBM copyright notice.  Well, this is a surprise to me and a delight at SCO headquarters in Utah, I'm sure, but I'll bet my house that SCO does not prevail and here's why.

According to Laura Didio of the Yankee Group, "[SCO's] claims are not limited to just one area of the Unix System V kernel. SCO claims there are multiple instances of copyright violations. SCO said these include: NUMA (Non Uniform Memory access) a mechanism for enabling large multiprocessing systems, RCU (Read Copy Update) (and) SMP. All of the aforementioned functions represent high end enterprise performance and scalability functionality portions of the code."

And all those parts appear to have come originally from Sequent Computer Systems, now owned by IBM. RCU was implemented in Sequent's DYNIX/ptx, a legally-licensed derivative of System V, in 1994 for SMPs and in 1996 for NUMAs.  The RCU code inside the Linux version 2.2 kernel even includes the name of Paul McKenney, who was a major contributor to both the DYNIX and Linux versions.  The same guy wrote both pieces of code and probably did do some cutting and pasting between them.  To SCO, this is the smoking gun that makes IBM viable for treble damages because SCO's UNIX licenses cover derivative works.  That means if I have a System V source code license and I change that code, any changes I make live under the original UNIX copyright.

So that makes IBM guilty, right?  Wrong.

If we go back to the Sequent RCU research papers published about this work, we'll see they are very carefully written to present a general way of solving this problem on almost any multi-threaded operating system. It is a general solution. In the key paper, the first mention of some version of UNIX doesn't come until page five under the "implementation" section.  They did this work -- work that was supported by a variety of federal grants and involving more companies than just Sequent -- to develop a concept that they then implemented on UNIX.

Now let's think about the UNIX license and how it concerns intellectual property claims.  I am not a lawyer, but unfortunately, I have been involved in several copyright and trademark cases, and believe I know the law in this area.  SCO looks inside the System V source code and finds this implementation.  They look in the Linux source code and find a similar or identical implementation.  Sure enough, both can be traced to the same programmer at Sequent, which is now owned by IBM.  And SCO, as the UNIX IP enforcer, owns the license for all derivative works -- all derivative UNIX works.  David Boies sees this as his smoking gun and he's going to use it.  But David Boies is not an IP lawyer by trade.  This is key.

The IBM lawyers (who ARE IP lawyers) will strongly argue that none of this matters since we have a case of a single person who did two very similar implementations based on his earlier research.  Both his UNIX and Linux versions (works B and C) were derived from his original research (work A) which was not exclusively limited to UNIX.  His paper shows that was the case and while SCO may see it as the smoking gun, IBM will see it as the proof of innocence.

What SCO owns (forgetting for the moment Novell's contrary ownership claim and perhaps AT&T's) is the copyright on this particular work as applied to UNIX.  But Linux is not UNIX, so applying the same ideas -- even the same code if it comes originally from an upstream source -- is not necessarily copyright infringement. 

Say I write a new high-level programming language, then do nearly identical implementations of that language for UNIX and Linux and the UNIX version is made part of some official UNIX distribution.  Does that mean the Linux version violates the UNIX copyright?  No.  But I wrote both versions and the code is identical.  Surely that is a copyright violation?  No.  This isn't a matter of clean rooms and virgins and reverse engineering, it is a matter of precedence and authorship.  Sequent (now IBM) did not give up all its rights to the code when it was made part of UNIX.  They were very careful to plan it that way.

IBM has the largest legal department of any company in the world.  They are INCREDIBLY sensitive about IP ownership, which produces for them more than $1.5 billion per year in license fees.  They have embraced the GPL very carefully for their Linux work.  The very fact that this code was released under the GPL indicates it was vetted and found acceptable by the IBM legal department.  It's not like sometimes they don't bother to go through this procedure.

The upshot is that I believe David Boies will put on a very good show, but that the case will be thrown out on its merits.

And while this is happening, a whole lot of damage will have been done to vendors and customers alike, with only one party benefiting from the drama -- Microsoft.

SCO is effectively trying to destroy both the UNIX and Linux markets.  This makes no sense, but that is the logical result of their current efforts.  The idea that 1,500 of America's largest companies will be forced to drop Linux and will do so in favor of SCO's UNIXware is ludicrous.  Why would those companies spend big bucks buying licenses from SCO -- a company they are upset with -- when they can comply just as easily, and almost for free, by converting to one of the BSD variants?  Only Microsoft has had success bullying customers into buying its operating systems and SCO is definitely not Microsoft.  This behavior won't sell any software. 

Meanwhile, Oracle is trying to destroy PeopleSoft, one of the most successful application development companies around.  PeopleSoft's Customer Relationship Management (CRM) software is at the heart of many of the biggest Oracle databases.  Oracle, thinking it is reaching for growth in a flat market, actually runs a terrible risk of infuriating its biggest and most important customers.

Microsoft is smart and quick.  They are no doubt angling to take advantage of this new chaos in the software industry.  If history repeats, Microsoft will make very good business decisions.  Everyone else will make very poor, if not stupid business decisions.  The result will be that Windows will be stronger, and Microsoft's own CRM products, acquired when it bought Navision (the Danish CRM company), will gain a foothold in the market against PeopleSoft and Oracle.  A year from now, Microsoft will be a vastly more powerful business even than it is today, which is saying something. 

Where is IBM in all this?  If IBM were smart, they would be beating a path to J.D. Edwards, PeopleSoft, and SAP's doorsteps.  They would be making those companies sweetheart deals to support and resell IBM's Websphere development environment and DB2 database, grabbing some market share from Oracle.  IBM should be helping PeopleSoft hold Oracle at bay, making it worthwhile for customers to move their PeopleSoft and SAP applications from Oracle to DB2.  But this is very unlikely to happen.

Unfortunately, it would take IBM months to recognize such a golden opportunity and more months to approve a plan.  Probably every IBMer who sells or supports products in this "space" (IBMspeak) understands the situation.  But when your leadership is too unaware and too lethargic, well opportunities are missed.

Which brings us back to Ashton the macaw.  When Ashton-Tate sued Fox Software for copyright infringement in 1988, the suit was eventually thrown out because Ashton-Tate was shown to have made false statements in its original copyright application for dBase II.   

The company claimed that it "owned" the source code underlying dBase II -- code drawn from a database called Vulcan that was developed by Wayne Ratliff at NASA's Jet Propulsion Laboratory.  Vulcan was a Z-80 assembler version of JPLDIS, a mainframe database program written at the lab by Jeb Long and others.  Long later joined Ashton-Tate and was responsible for leading development of dBase III and IV.  Where Ashton-Tate apparently made its mistake was in forgetting that buying the marketing rights to Vulcan from Ratliff didn't invalidate the intellectual property rights of Ratliff's employer, JPL.

Think about it.  Ashton-Tate's claim on dBase was, in many ways, similar to SCO's current claim on derivative UNIX works.  They both ignored upstream property rights of others.  What is ironic about this is that Fox Software wasn't the only company sued by Ashton-Tate for this supposed copyright violation.  Fox's co-defendant was SCO. And having been on the other side of such a similar case, they should know better.  

[ShadowAce]

Red Hat offers Advanced Server--but you can't get it for free. SuSE doesn't even offer a downloadable version. Look at the major distributors--they all offer a version of Linux that you must pay for.

Granted--there are many distros that do not have a for-pay version, but the fact that several of the big boys DO, puts the lie to the claim that Linux must be free-as-in-beer.

The whole point of Linux is not its price point--the whole point is that you, the customer, have direct access to the source code. Thus, when something isn't working quite right for your installation, you have the power to fix it--on the spot, rather than placing a call to tech support.

It's not free to run or support--you need the expertise for that--just like you do for Windows.

[Dominic Harr]

Actually I'd prefer a GPL, open source snicker.

:-D

[Bush2000]

SCO's only efforts are to take Linux over lock stock and barrel, period.

So, in addition to being crack smoking open source sycophants, you're also mind-readers? Get real. IBM doesn't own Linux. SCO's suit against IBM, at worst, would prevent Linux geeks from copying technology.

That would only solve the SCO IP issue. That wouldn't solve the issue of anybody else's IP being in Linux, and even worse it would just be a reverse stealing of all the work that the Linux volunteers have put into Linux for years as a gift, not to SCO, but to the world.

Stealing other peoples' IP just so you can put it in Linux is not only illegal but immoral. If this has happened, Torvalds and company are going down big-time.

[Bush2000]

I deploy to both Windows and HP-UX, and I do have one departmental web server and one test app server running Red Hat and Apache.

Yeah, I note you use the word deploy. I'm talking about everyday development. You and I both know that you use Windows, not HP-UX or Linux. Why?!? Why?!? Why?!? Can't Linux measure up?!?

[Bush2000]

The whole point of Linux is not its price point--the whole point is that you, the customer, have direct access to the source code. Thus, when something isn't working quite right for your installation, you have the power to fix it--on the spot, rather than placing a call to tech support. It's not free to run or support--you need the expertise for that--just like you do for Windows.

Harr has been blathering on-and-on-and-on about vendor lock-in. But the fact of the matter is that customers are indeed going to pay some vendor to support their crappy Linux code. So the lock-in still exists. It's simply moved from MS to some other company.

[ShadowAce]

So the lock-in still exists

With Java, I agree. With C, Perl, Python, shell scripting, I don't agree. It all depends on the tools you use. I use Quanta Plus for my development. It allows me to write virtually anything I want in any language, and it will highlight the code according to what I am writing in. It's pretty cool.

[Dominic Harr]

I'm talking about everyday development.

I do dev on NT4 at work (corporate standard, not up to me) and both Windows and Red Hat at home.

I find Linux to be the best server OS going, and altho it is not yet as fully functional as Windows, the only real shortage is quality games.

And as you know, I'm dealing with that in my own way . . . :-D

[Bush2000]

I do dev on NT4 at work (corporate standard, not up to me) and both Windows and Red Hat at home.

So you admit that Windows is a better development environment. Thanks. We can move on now...

I find Linux to be the best server OS going, and altho it is not yet as fully functional as Windows, the only real shortage is quality games.

That makes zero sense. Nobody runs desktop games on a server OS, charlatan.

And as you know, I'm dealing with that in my own way . . . :-D

Rrrrright. You've been talking, talking, talking about it for years. More hot air, as far as I'm concerned.

[Dominic Harr]

You've been talking, talking, talking about it for years.

About a year, yes. That's not too long for one person on a project this size. In fact, by industry standards, it's rather amazing.

And it's just about there -- 95% there.

Early feedback on the alpha was *great*.

[Bush2000]

About a year, yes. That's not too long for one person on a project this size. In fact, by industry standards, it's rather amazing.

Great. You're a legend in your own mind.

And it's just about there -- 95% there.

Based on your half-baked ideas, you don't strike me as a "closer". I won't be holding my breath.

[Nick Danger]

Submitted for your approval: A large American corporation that sends its employees into discussion forums, all around the Internet... spewing insults, cackling like jackals, and behaving like thugs.

What sort of corporation does this? What kind of personality is at its core? And what of the people who join it, like it, and stay?

You're about to be transported into a world of egotistical greed beyond your imagination. A world where truth and lies have no distinction; where anything can and will be said to win the moment; and where insult and bombast are the order of the day.

Welcome to the world of the Internet Thug, and a corporate face that could only have spawned in... the Twilight Zone.

[HiTech RedNeck]

the obviously illogical license

It's very logical. The only prayer SCO has of taking over Linux per se is to win its impossible claim that it owns the IP of all operating systems worldwide. Otherwise the court will force SCO to reveal to Torvalds what needs to be done to purge Linux of SCO content.

[HiTech RedNeck]

"Stealing through barratry is OK for me but not for thee!"

Yeah, that's your hypocritical turkey gobble.

[HiTech RedNeck]

At worst someone else, perhaps in a different country, will pick up the Linux torch. It will be a classic case of corporate slaying of their golden goose.

[Bush2000]

Submitted for your approval: A large American corporation that sends its employees into discussion forums, all around the Internet... spewing insults, cackling like jackals, and behaving like thugs.

What are you saying, Nick? That IBM put you up to this? Tsk, tsk. For shame...

[Bush2000]

Submitted for your approval: A large American corporation that sends its employees into discussion forums, all around the Internet... spewing insults, cackling like jackals, and behaving like thugs.

What are you saying, Nick? That IBM put you up to this? Tsk, tsk. For shame...

[Bush2000]

Submitted for your approval: A large American corporation that sends its employees into discussion forums, all around the Internet... spewing insults, cackling like jackals, and behaving like thugs.

What are you saying, Nick? IBM put you up to this? Tsk, tsk. For shame...

[Bush2000]

At worst someone else, perhaps in a different country, will pick up the Linux torch. It will be a classic case of corporate slaying of their golden goose.

If SCO succeeds in seizing the golden goose, you might as well stick a fork in Linux. Granted, that's a longshot. But it's worthy of speculation. The only reason that Linux has gotten where it has in the past few years isn't because of hackers writing code in their free time; it's because of investment by large corporate sponsors like IBM (and [sic] SCO). It won't matter whether Linux goes offshore. It would be a pariah. And anybody who distributes it openly would get their ass sued.

[Bush2000]

Don't be surprised when Torvalds gets sued. It's just a matter of time.

[InfraRed]

How will they do that, if the code is 'trade secret', and never publicly released?

The code has already been publicly released (by SCO). It's now a matter of determining which lines -if any- need to be replaced. Before the courts can render a verdict, SCO will have to provide evidence. IF the courts rule in SCO's favor, you can bet that the source (a person) of the code will be identified. Code originating from that source would be purged.

the growing loss of income to US software companies who sell products for profit,

There is one particular entity that has caused plenty of US software companies to lose income by bundling "free" application programs with its OS. Over the years, that model has not worked out too well for many companies, and more companies recognize the danger each day. If it's OK to develop and bundle free apps with an OS, it's also OK to develop and bundle a free OS with your hardware or services.

further expansion of insecure code created by foreigners into US markets and government

I am also wary of foreign code, but it's coming at us from all directions (H1B visas, outsourcing, etc). Having US companies inspect and certify distributions/installations is probably the best that can be expected. Unfortunately, there is a great deal of closed source that will never be seen by the eyes of 3rd party analysts.

I also have a problem with Torvalds ego, whereby he refuses to give up personal ownership to IEEE at minimum, but instead insists on complete authoritative rights.

Who knows, maybe some shyster lawyers can back Linus into a corner and force such a decision. It's a real shame that you, personally, don't get to dictate Mr. Torvalds' decisions for him. Maybe if you were to relinquish control of your projects to IEEE, Linus might follow your example. Hey, can you imagine the result if Microsoft gave up ownership of Windows to the IEEE? ;)