Martin
Does SCO have a point in their claim or it's juts baloony bs? I can't imagine someone would be so stupid and start accusing others of stealing without any evidence but I may be wrong on this.
SCO had a contract with a group of Unix companies (IBM, HP and others) to port Unix to the Itanic (Itanium) -- then when a Linux port beat them to the door they got dropped and the Unix companies decided to go with Linux. It is remotely possible that some of the code they developed for their unsuccessful attempt may have made it into the ia64 source tree, and might have infected the non-architecture specific code as well. If so, it would be a simple enough process to remove that -- for that matter Linus thinks the ia64 is a piece of junk and would probably be happy to dump the ia64 branch without looking back. (I base this on his criticism of the ia64 hardware architecture on linux-kernel.) Apart from that, from the very beginning Linux has been written with an eye to open standards: POSIX, Unix95, Unix98 etc. propagated by the Open Group which state very specifically what an operating system needs to implement in order to be considered "Unix". Those standards are very specific, so the code created by following those standards by necessity would tend to lead a programmer who had never seen a line of "real Unix" operating system code to create C code that would be remarkably similar to the code in the "real Unix". C is a language with very established and respected idioms so most code that accomplishes the same thing in the most optimum way is going to look more or less the same when done by competent programmers. And to add to the potential similarities, I would say that every Linux kernel programmer has read Kernighan-Ritchie's "The C Programming Language" early in his career, (who are certainly two of the most important contributors to the original Unix code) -- that book teaches the fundamental C idioms in a most convincing manner. To top that off they probably all have studied the BSD sources as well. If it ever comes to court there is probably reason for a bit of worry on our side. (You have to always remember that the judge and jury will know absolutely NOTHING about programming -- you might as well ask your pet dog for a verdict.)