3000 developer job requires no migration
Fresh CEO carries HP rebound hopes again

Ritchie's rich legacy: Unix both vital, hated

Dennis-Ritchie-CompositeOne week exactly after the death of Steve Jobs, Unix co-creator Dennis Ritchie died Oct. 12 of prostate cancer at age 70. In addition to creating Unix at Bell Labs along with Ken Thompson (above), Ritchie is credited with creating the first C programming language. The most technical of community members think Ritchie deserved the same outpouring of grief and praise Jobs received. Some writers compared Ritchie to Tesla, who invented AC current versus Edison's Direct Current. Edison died rich. Ritchie died alone.

Ritchie's inventions deserve praise, and he got much of it during his lifetime from his technical peers. But the invention of C and Unix had as checkered a past as anything Jobs and Apple sold, and a more caustic effect on better-designed inventions. The HP 3000 community in particular suffered from the snake oil of Unix. A better friend to the 3000 has been Ritchie's C -- and not coincidentally, it's that language that's still making magic that 3000s can use.

You can smell that snake oil burning when you read this week's post from Google's Rob Pike, a colleague of Ritchie's and the first man to report Ritchie's death.

Unix was the great equalizer, the driving force of the Nerd Spring that liberated programming from the grip of hardware manufacturers. The hardware didn't matter any more, since it all ran Unix. And since it didn't matter, hardware fought with other hardware for dominance; the software was a given.

Haters HandbookBell Labs was the industry tower that Pike patrolled in the late 1980s, and that certainly wasn't any province close to an HP lab or a 3000 development cubicle or a company's IT director office. Unix was no more of a given than "spoken language" means "something everyone can understand." Unix remains full of byzantine differences that made the hardware more important than ever, because every vendor sold theirs as the One True Unix. Especially HP, which fed 3000 customers into the Unix ovens while it cooked up a larger installed base for HP-UX. The only thing that was a given was that HP was giving away its customers to Unix, while the world was climbing the shaky rope ladder into Windows. Ritchie was writing an "anti-forward" to a notable book exposing fatal flaws in his creation.

The point here is that technology has always been essential to a product's success and the health of its inventing company. But without insight, passion and a drive to create customer appetites and loyalty, technology like Unix can become the DC of electric service. You won't see much kowtowing to Unix this fall, since it's been eclipsed and outclassed by Linux. The Ubuntu version of Linux will actually save some 3000 customers from the fate of those 1990s ovens. But it took more than 20 years of refinements to make something a given: hardware, now replaced by virtualized servers like the CHARON product for the 3000.

Unix meant HP-UX, Solaris, AIX, Tru64, Ultrix, Xenix, IRIX, A/UX (sold by the non-Jobs Apple) and countless more. Unix was such a rabbit warren by the middle 90s that one of its chief makers, Sun, created Java to bind all software together in "write once, run everywhere." That didn't turn out to be any truer than Unix being a force Ritchie created to spark all digital inventions, including Apple's.

Unix might be the only operating system ever to sprout a Hater's Handbook, one widely cherished by those who worked with the OS and now such a legend you can read it for free, after multiple printings by IDG Press. "The Best of the UNIX-HATERS Online Mailing List Reveals why UNIX Must Die!" To be fair, people might have derided Apple's Newton just as much. But the Newton never killed off worthy achievements in technology. And that is what's being stated here: An invention in technology is more praiseworthy than a successful product and company.

Just try telling that to an HP 3000 expert or owner, now looking for work or hoping for a replacement. Ritchie is deserving of the Turing Award, the US National Medal of Technology, the IEEE Richard W. Hamming Medal, and the Japan Prize. He was also honest enough to know what he'd invented was far from a magic catalyst. "UNIX is very simple, it just needs a genius to understand its simplicity," he said. He started to inch closer to Jobs' hubris while saying that "C is quirky, flawed, and an enormous success." Quirky would be generous for the language both Tymlabs and CCC tried to sell to the HP 3000 market in the 80s and 90s. C needed "lint," another of those four-or-less lettered Unix tools, to find all the bugs in a typical program.

Ritchie deserves a memorial, but working in technology's high towers won't summon the sorrows of flowers laid at doorsteps of Apple Stores. There are fine ones written to assay his genius at the New York Times, the Guardian and many more outlets. On that Hater's Handbook cover, Clifford Stoll said "The next time a UNIX addict tries to intimidate you, reach for this book." Intimidation was rife among the Unix acolytes. The next time an acolyte tries to tell you that Jobs was a mere pitchman while Ritchie was an underpraised genius, consider what each has left behind that is working as their companies built it, and what they were doing in the year that they died. We stand on lots of people's shoulders to make tech magic serve us. Nobody is perfect, but the carping about Ritchie oversight versus Jobs belittles both men. Ask HP 3000 customers. Great tech needs a great advocate to survive.

Comments