SCO's Smoking-Gun TourBy Larry Dignan | Posted 2003-08-01 Email Print
WEBINAR: Live Date: December 14, 2017 @ 1:00 p.m. ET / 10:00 a.m. PT
Modernizing Authentication — What It Takes to Transform Secure Access REGISTER >
Will a packet of paper with highlighted red and blue codeallegedly lifted from Unix to Linuxgarner licensing fees?NEWS ANALYSIS--Is SCO's smoking gun against IBM a sheet of 8 x 11 paper with pertinent lines of programming highlighted in red and blue?
This code, which was allegedly lifted almost verbatim from Unix to Linux, belongs to a large unnamed hardware vendor that isn't IBM, according to SCO, which was waving it around late in July.
But SCO argues it is evidence that many companies are violating its intellectual property, says Chris Sontag, general manager for SCO's SCOsource unit. SCO acquired not just the source code for Unix System 5 from AT&T years agobut the contracts that pertain to its use by commercial software and hardware makers.
Sontag is making the rounds with press and analysts, arguing IBM is the "ringleader," something akin to an industrywide porting of Unix to Linux without permission.
His presentation boils down to this: IBM "donated" some of the functionality from its own Unix variant, known as AIX, to Linux version 2.4 and beyond. This helped the open-source software grow to handle nonuniform memory access (NUMA), journal file system and other important features for an operating system asked to be a workhorse of enterprise computers. "How did this happen in that short of time?" asks Sontag. "A third or more of the Linux 2.4 kernel is at issue."
According to Sontag, IBM, along with other vendors, gave Linux a helping hand and violated a "derivatives clause" in the contracts for using Unix System 5. The contracts appear to say the source code can be only used for internal purposes and can't be redistributed elsewhere.
For its part, Trink Guarino, IBM director of communications of e-business on-demand, says "SCO hasn't shown us any code specifically that violates our agreement." Sontag says it should be clear what code violates the contract, but acknowledges he hasn't shown IBM a line-by-line comparison. IBM plans to vigorously defend that point and itself.
It's likely that IBM won't see SCO's code until the discovery phase of a trial that isn't scheduled until April 2005. IBM will argue that it added onto the source code, and enhanced functions are gradually turning up in Linux.
So does SCO have a legitimate claim, or is it just a sue-happy company trying to garner licensing fees? IBM will argue the latter, but may change its mind if SCO shows the world offending codea move that would probably result in a lot of workarounds.
There will be more surprises to come. Remember Novell's May 28 challenge whether rights to the Unix source code even transferred to SCO? Turns out a paralegal found an amendment in a dusty file cabinet to prove Novell wrong. On June 6, Novell issued a sheepish release noting that SCO owned the source code.
Bottom line: SCO doesn't even know how much of its Unix intellectual property has allegedly been violated. It's sitting on top of reams of old contracts that could be enforced.
While it sorts itself out, SCO should have a tough time convincing any corporation to pay licensing feesuntil it beats IBM at trial.