It's official: I've become an "industry pundit." The January/February edition of The Connection arrived in the mailbox while I was out covering Macworld, and I got myself put into an article about HP's Odyssey Project, in a footnote. It's a tale that like Sister Mary Ignatius, Explains It All For You. (Apologies to Christopher Durang's play, but I'm a lapsed Catholic boy and can feel a lecture in the air.)
At least I didn't get my knuckles rapped with a ruler. Dr. Bill Highleyman, managing editor of The Availability Digest and a past chair of the Tandem User Group, addressed the serious question about the future of Itanium in his rousing conclusion to The Connection article.
One industry pundit suggested that [HP VP] Markin Fink's reference to a "single platform" signals that the Intel Xeon chip family is going to win out in HP's near future, probably meaning the end of the Itanium developments from Intel after its next two processor rollouts become a reality.
I remember writing that suggestion on November 25, indeed. There were a few other articles that followed it, but I'm not going to cry out "misquoted out of context." No journalist should ever bark that out, although I invite you to read my other article that immediately followed my punditry. If you consider how long it's going to take Intel to do its next two Itanium rollouts, customers will be in the territory of 2016, or even later. (The last two rollouts took a lot more than two years each.) Nobody at HP has shown a roadmap on the future of HP-UX beyond that date.
If Xeon hasn't "won out" already -- and those are Dr. Bill's words, since I'm no fan of racing metaphors -- it will surely represent a walloping majority of Intel's energy in four years. The end of proprietary tech at a vendor can come silently and quickly for no good technical reason. The HP 3000 did not officially lose HP's favor until a blind-side announcement. Right up to the late summer before MPE/iX got its HP dismissal, HP was still encouraging customers to ride that racehorse.
Dr. Bill quotes Pauline Nist of Intel in her company blog as saying
Intel remains equally committed to the Itanium and Xeon platforms, both of which represent our portfolio approach to bringing open standards-based computing to the mission-critical environment.
The next thing you know we'll be hearing how Itanium is "strategic to HP." Lessons from the 3000 division -- whose final GM, Winston "Coup de Grace" Prather, has become the Tandem/NonStop GM -- should make you want to race for the doors if strategic ever gets used to describe a product's future.
"[A product marketing manager at BCS] said that 'HP is not now planning to port HP-UX to x86-based servers.' " But the article wonders, "Was the operative word 'now' intentional?" (I'm reminded of a semantic debate about the definition of "is.") Dr. Bill is ready with an answer, right now, from Mr. Fink.
Martin: [The HP rep] is basically saying, "Never say never." At this point there are no plans; and I predict that it will never happen. The big problem is the software support and the ISV support for the 5,000 current HP-UX ISV applications. The better model is to bring the HP-UX capabilities to Linux, rather than port HP-UX to x86.
These are all very sharp people. I wonder why everybody is missing this point: why you'd ever need to port HP-UX, or whatever this better model is supposed to do. My headline of Nov. 25 said HP was aiming Unix sites at x86 futures. It sure seems to me, as an industry pundit, that once you bring HP-UX's capabilities to Linux, those customers using the capabilities aren't going to be HP-UX users much longer. I might be confused -- because at the same time that one person in HP says "never say never," about a UX port to x86, another says "I predict it will never happen."
I prefer to take my lessons on soothsaying from Yoda. In The Empire Strikes Back, Luke asks him if his friends Leia and Han will die. "Difficult to tell," says Yoda. "Always in motion is the future."
All I know for certain is that 2016 looks like a lot more robust year for Linux apps than for HP-UX apps. And if you're still smarting from the knuckle-rapping HP gave you when you last paid the vendor to run a proprietary hardware and software platform -- well, you might not like what Highleyman assures you with. He quotes Martin: "IBM's strategy is not at all like Project Odyssey." But then the Doctor adds this.
IBM's proprietary operating system zOS has survived living alongside a hardened Linux. Hopefully this is an indication that the HP proprietary operating systems will survive alongside HP's hardened Linux and Windows.
If the demand for these operating systems declines, will Itanium survive? Martin: This is not an issue. Most planning cycles are five years or less. Changes like this take a long time. As long as you and I are around, we'll be supporting HP-UX on Itanium.
That sounds like some swell whistling past the graveyard, as our old friend Wirt Atmar used to describe optimism about HP's 3000 intentions. Martin Fink is predicting that UX is never getting to x86 and that HP will support HP-UX on Itanium as long as you and I will be around. I'm only 54. Unix is past 60 and starting to slow in the market. Father Time wins every race, to fall back on that hoary metaphor. Nothing lasts forever, but now HP has defined a new future where a "hardened Linux" somehow apes the best of HP-UX but it doesn't supplant it. Wirt had a professorship in evolutionary biology. He could see how any vendor's product has a lifespan that includes death.
If HP customers' best bet is to believe that IBM's mainframe zOS business will be a model for HP-UX -- well, good luck with that. HP's never liked doing things the IBM way until they're forced into it. I can't predict if there will be enough distubance in The Force of Itanium users to force HP to make HP-UX outlive us youngsters. I can only suggest.