SCO versus the Linux world

Most people don’t care about computer operating systems anymore; they’re happy to run Microsoft Windows and pay Bill Gates an occasional tax.  However, for engineers that build 5000-machine server farms or cheap consumer electronics products it is often essential to have an operating system whose source code can be modified and/or that is free.  That’s the role of Unix, whose most popular current variant is known as “GNU/Linux”.


Unix was developed in 1970 at Bell Labs primarily by Dennis Ritchie, Ken Thompson, and Brian Kernighan.  It was substantially improved by University of California Berkeley in the late 1970s.  Richard Stallman and his collaborators in the free software movement, starting in the 1980s, further improved the system and freed Unix from AT&T’s cumbersome licensing restrictions.  Linus Torvalds contributed a free kernel that completed the job started by Stallman.


Through most of its life Unix has represented old ideas, old technology, and an inferior set of features compared to the research and commercial state of the art.  Nonetheless because it was cheap and easy to install on a wide variety of hardware, Unix buried all of its competition except for IBM’s mainframe operating systems and Microsoft Windows.


Under the original 14-year copyright period enacted by the U.S. Congress, SCO’s recent legal attacks against IBM and other Linux users would be impossible.  You couldn’t go to court and say “I want to sit on my butt and collect dividends from this thing that someone else did 32 years ago.”  But copyright today for corporate works has been graciously extended to 100 years, mostly thanks to some Congressmen on the Disney payroll (they didn’t want Mickey Mouse to become a public domain character).  Tim O’Reilly seems to be the only person in the U.S. adhering to the original 14-year term.


Effectively infinite copyright terms are good for Disney’s top managers (and would be good for Disney’s shareholders if the managers didn’t take all of the profits home as salary).  But are they good for American industry?  Microsoft can sit in Redmond making minor improvements to Windows NT/2000/XP/2003, a fairly modern operating system when introduced in the early 1990s but showing its age now, and collect 30% profit on its revenue (one sure sign of its monopoly power; Exxon/Mobil earns about 7% profit by comparison and Toyota earns 5%).    Companies can often make more money by asserting Congressionally-created intellectual property rights in ancient computer programs than they could by building something new and useful.  Being an American corporate manager, swaddled in government-guaranteed rights that never expire, is sort of like growing up in a very rich family.  You could make more money if you tried to work a bit but why strain yourself when you can be quite comfortable without working at all?


(If you want to follow the SCO saga as it unfolds, http://slashdot.org/ is probably the best place for news.)

4 thoughts on “SCO versus the Linux world

  1. Ah—who cares anyway, just have another beer and don’t worry. Bush will take care of all!!!!

  2. The comments about copyright extension at the behest of Disney are probably true, but they’re not relevant for the SCO case, because there really is no legitimate SCO code in Linux to begin with. The claims being made by SCO are so ridiculous that they’ll be thrown out if this ever gets to court. Presumably, SCO just wants to be bought out by IBM on the theory that IBM will decide it’s less costly to fight the case than to pay the blackmail.

    On another subject, it’s sort of quaint seeing people follow the Stallman “GNU/Linux” naming religion. Why don’t you just call it Linux? Calling it “GNU/Linux” is like always referring to Windows as “Microsoft Windows”. Sure, Microsoft (Stallman) want you to do that, and it’s more precise and more descriptive that way (and it gives credit right there in the name itself), but as a name it’s a horrible loser because it’s just too verbose. “GNU/Linux” in prose is about as stilted as an MS press release where it’s always “Microsoft Windows” and “Microsoft Office”, always spelled out in full.

  3. It’s also not clear when the code that SCO claims was copied was originally written. I suspect that it is more recent than 14 years. (Partly because I suspect that it may have been Linux code moved into SCO and not vice versa.)

Comments are closed.