Jim Fawcette (head honcho at the eponymous publisher) has a regular column which he typically uses to diss some big tech name or another; recent volleys were aimed at Google, Java, and Microsoft. If you’re going to be a grinch, being even-handed about it is a good thing. The most recent outing, an extended fulmination on Open Source, gets so many things wrong, though, that I have to push back a bit.
First, he makes a blanket assertion that Open-Source software is innovation-free. You hear this a lot, but in general it’s wrong. Granted, a lot of OSS is a rediscovery of something already built: Linux for Unix, MySQL for any number of commercial RDBMSes, and so on. But, uh, you have to work pretty hard to avoid seeing Apache, where a lot of what’s now the conventional wisdom about Web servers was invented, or Perl & Python and friends, which showed that you can do serious work by treating everything as a string and processing everything with hash tables and regular expressions. And it’s idiotic to say Linux is innovation-free. For just one example, look at the futex gizmo in the 2.6 kernel (docs here and here); which looks to me like (finally, after all these years) the right basis for building semaphores and mutexes.
I’d say that OSS is in general neither significantly more nor less innovative than other kinds of software.
Second, he utters the following: “But more importantly, what does it contribute? Why would anyone with excellent computer skills want to work long hours to create code so that millionaire executives at IBM can use it to sell expensive mainframe computers and middleware with six-figure licenses? All for no compensation and little recognition.”
This is spectacularly clueless in two different ways. First of all, IBM is doing very nicely by OSS, but not by selling middleware or mainframes; doesn’t Mr. Fawcette actually look at IBM’s financials? The big winner is Global Services, which can charge the big bucks for deploying systems and not have to compete with software license charges for the buyer’s buck. Basically, IBM learned the lesson that Eric Raymond ably outlined in The Magic Cauldron, that the software biz is (at least in part) naturally about services, not manufacturing.
Secondly, it escapes me how a sentient human being supposedly engaged in the technology sector can use the phrase “No compensation and little recognition” in connection with Open Source. It seems to me that Linus Torvalds gets on more magazine covers than Jim Fawcette, and while few in the OSS community are getting rich, a whole lot of them get paid perfectly decent salaries to do what they love. Hint to Mr. Fawcette: there’s more to life than stock options, better cubicles, and the title on your business card.
Now, I don’t want to come on as an OSS bigot; I firmly believe that the software universe will contain both Open-Source and commercial software for the forseeable future, and the world of commercial software has been pretty good to me.
But unlike some, I don’t see OSS as evil, or as a threat to anything important. This, I’d say, is a case of a publisher who needs an editor.