When I joined Novell back in 1995, it had just off-loaded Unix. The dream of creating a super-NOS out of NetWare and Unix had turned out, like so many initiatives before it, to have been no more than a nice little side project for honing the coding skills of Novell's then-chief scientist and his cronies; commercial viability was less than an afterthought.

The CEO at the time had more important things on his mind - like getting Novell back to its core objectives. I remember the departing Unix engineers as being somewhat disappointed in the deal but pleased that the network operating system they loved had gone to a company committed to Unix. Most of the lads I briefly knew were off to work for SCO in its various locations around the world. They were looking forward to the challenge ahead - making Unix even better than it already was.

How times change.

SCO has altered its name, ownership and strategy a few times since then but can there be any employees, backers or customers who still genuinely believe that SCO's main commitment is to improving its primary product? Selling some of it would be a nice start.

But try telling that to Darl McBride, SCO's tough-guy CEO, who swaggers around like a red-faced schoolyard bully with Tourette's syndrome. SCO has one objective right now - litigation. And it's proved itself pretty good at it in the past.

Unfortunately, now that SCO has gotten a taste for blood, it wants more and so has decided to go for a big target. And next to Microsoft, the juiciest tuna in the sea is IBM. It's a damn sight easier than setting out a business plan, displaying leadership, recruiting and training dedicated staffers, and generating revenue from an operating system and related products. SCO executives have done their sums and decided that competing in the cutthroat world of software isn't for them; they'd rather cut throats in the soft world of the courtroom. Red Hat, VA Linux and all the other ex-SCO competitors must be heaving sighs of relief at a loss of focus unheard of since the Hubble telescope lost its focus.

SCO isn't the first to take a gamble. Many companies have banked everything on a single big idea (remember WordPerfect?) only to come undone, but to bet all you have on the outcome of a court case (and an intellectual property case, at that) is straying so far into the bounds of uncertainty that I expect that SCO's lawyers believe not just that a big win awaits them but also in goblins, elves and the tooth fairy.

So Microsoft and others have stumped up and paid SCO for its dubious licences? "This proves SCO has something!" cry the wobblers. Maybe, but one can't help but ponder the confidential intricacies of the deal. The monies paid may well be peanuts to Microsoft. (In fact, anything under $2 billion is peanuts to Microsoft, and if the amount paid were even a fifth of that, then McBride, Sontag and company would now have their prize and be on a Caribbean island sipping margaritas and lunching on roast penguin.)

But if SCO is eventually defeated in court, is it really inconceivable that Microsoft may come after some of the moolah it blew on now worthless licensing? Bill Gates may not seem like an avenging angel, but Microsoft possesses attack dogs like no other. If McBride by then considers himself a courthouse veteran, inured to the constant boredom of motions and countermotions, he'll be in for a shock and a "motion" of an altogether more unsavoury kind once he feels the sharp steel of a Microsoft lawsuit.

But let's entertain SCO's quest for a moment and offer the benefit of the doubt - what if SCO's right? What if some stressed-out, overworked IBM employee 20 years ago couldn't bother to write some tricky subroutine and ended up pilfering it wholesale from the shelf marked "Freeze-dried Unix. Special-offer today only"?

It will have no implications for what we know as Linux. Many have already commented that the portions of infringed code would be rewritten, new distributions compiled and DVDs burned within a matter of weeks. But what about the harsh fines, cry the doubters? Sure, SCO may convince a sympathetic court to grant it some caveat allowing monetary recompense for retrospective use.

Big deal. This tack is at best a short-term gain and at worst an inherently dwindling source of revenue. Can you imagine the download links on all the Linux Web sites? "Click here if you wish to continue using our infringed version (SCO payment required) or click here to download our new network operating system that washes whiter-than-white."

This is what in the US is referred to as a no-brainer. Want to know what the sales forecast for a strategy like that looks like? Draw a pair of x-y axes, put a pen in the top left, and draw a thick line down to the bottom-right till you've completed the triangle, then keep drawing till you've messed up your best tablecloth and are down on your knees scribbling on the carpet. If I were a SCO shareholder and this was presented to me at the annual general meeting as the grand business plan, I'd be online within an hour selling every piece of stock I had, just as soon as I'd picked myself up off the floor and wiped away the tears of laughter.

No, the only way SCO can win long term is if open source's underpinnings - the General Public License - is itself blown to bits, and amid the bluster about code obfuscation, who copied what when and copyright infringement, it's not hard to ascertain SCO's real aim: to stop the perpetuation of open-source software and do away with the GPL, its principal method of transmission.

This seems unlikely. McBride claims that the GPL is unconstitutional, and in an open letter in December 2003, he stated that the GPL "prohibits any proprietary use" of Linux code. This misses the point - there is profitable reward in open-source software, and the GPL does allow for revenue generation in the best traditions of capitalism as envisioned by the founding fathers of the US, just not in the conventional way.

Now, there are those who say that charging for software services and maintenance is just licensing in a different guise, but it's still capitalism, and yes, it remains to be seen whether a large firm like Novell can turn such a revenue model into long-term growth. But for the likes of SCO to protest against companies even being allowed to try smacks of a dictatorial style that Saddam himself would have been proud of.

There is a scene in the movie Point Break where Keanu Reeves jumps out of a doomed plane without a parachute. Not too clever, to be sure, but Keanu knew that seconds before, his nemesis, the bad guy played by Patrick Swayze, had himself taken the plunge - only with a parachute. So our hero decides he's going to hurtle on down to Swayze, grab on to him and share a splat-free descent safe in the knowledge that Swayze's chute will take both their weights nicely.

Now replay that scene, but this time take out Swayze and the parachute. That's where SCO now finds itself. McBride and his cohorts are in free fall. Their once normal and near-commercially viable company has become a monstrous, paper-guzzling litigation machine (with a sideline in Unix) weighed down by lawsuit upon lawsuit. And as each case is thrown out of court, SCO is scanning the horizon, desperately seeking out a potential target-cum-saviour who can cushion the fall.

Alas, time is running out, and SCO and its backers must surely soon realise that there is no blond, muscle-bound surfer onto whom they can latch; those wannabe Swayzes that were floating around salivating at the prospect of a share in the spoils have long since woken up and pulled their own cords. The grainy specks that looked so unthreatening two years ago at 30,000 feet are now hastening into view, and guess what. They aren't the giant, dollar-filled air mattresses that McBride et al were hoping for; rather what's looming is the cold, hard concrete of debt and the dusty barrenness of poor sales, disaffected employees and a crippled company culture.

Sure, the legalities will run on forever, but for SCO as a Unix business, time is nearly up.

Paul Coletti is an Edinburgh-based IT consultant and blogger who has worked in IT for 16 years at IBM, Novell and other companies. The opinions expressed in this column are his alone and in no way reflect the views of Novell or any other entity.