Gartner Group Compaq assessment codswallop

TERRY SHANNON, publisher and editor of Shannon knows Compaq has written an assessment of Compaq's rebuttal to the Gartner Group, and has kindly given us permission to run the story here.

This follows a leak we published yesterday. µ

"Compaq appears to have a slight disagreement with the Armani Analyst crowd. It seems that the Gartner Group, who last year penned a paean to OpenVMS which was subsequently posted on the CNET.com Web site, now has some second thoughts.

On 20 December 2001, Gartner Group published "The Future of OpenVMS," a research note which presented a bleak assessment for OpenVMS customers regarding the proposed merger of HP and Compaq, and specifically Compaq's plans to port OpenVMS to the Itanium Processor Family.

Not suprisingly, Compaq's OpenVMS Group wasted no time responding to a draft copy made available for review, and subsequently talking with Gartner directly to ensure that the firm's analysts were working from the latest information and facts. Electing not to be confused with the facts, Gartner issued a final report that ignored most of the points articulated by the OpenVMS Group.

The bottom line, Gartner claims, is that users should not feel compelled to follow the Compaq's IPF road map. Instead, they should seriously consider alternative migration plans from OpenVMS to other OSes, preferably Unix, but not Tru64 UNIX. Such customers are advised to have most, if not all, of the migration complete within three years of the final Alpha generation. Those customers who fail to heed this advice should not expect Compaq to deliver a full OpenVMS port with all layered software. So sayeth the Software Savants from Stamford.

Codswallop, sayeth SKC, which laments the fact that Gartner apparently didn't have anyone on hand at the VX Company's VMS: The Facts seminar or any of the many VMS-centric events such as CETS2001, Compaq's Platinum Forums and Technical Update Days, Encompass LUG meetings, and Compaq IPF teleconferences that have been delivering the True Facts about OpenVMS and IPF since 3FQ01.

Otherwise they wouldn't be making unsupportable claims such as "Grafting OpenVMS onto IPF is like trying to fit a square peg in a round hole," and "Compaq will fail to port the entire OpenVMS operating environment to IPF by 2003/2004 (Probability Factor 0.8)."

The fact of the matter is, VMS-to-IPF is a port, not a graft, and the procedure is well underway. At least one cross-compiler has been up and running for several months, OpenVMS was successfully ported from VAX to Alpha more than a decade ago, and all subsequent development work has been conducted in the C language to enhance portability. Unlike the VAX to Alpha port, the IPF port will affect one-tenth as many OS code modules and require as little as one-tenth the dedicated engineering resources as did the first port of the OpenVMS operating system.

The "square peg in a round hole" canard is particularly specious, as OpenVMS today has no knowledge of the underlying Alpha architecture, PALcode acts as the go-between. The current porting effort will incorporate the features and intrinsics now extant in PALcode directly into the base OS, rendering OpenVMS significantly more amenable to future ports should an IPF alternative emerge sometime down the road.

Accordingly, SKC strongly advises OpenVMS customers to obtain a second-and perhaps a third and even a fourth-opinion before accepting the flawed Gartner assessment at face value. And we'll stick by the prediction we made earlier this year (see "Fearless Prognostications: Compaq's Changes in 2002," SKC V9N1, January 3, 2002) that an initial OpenVMS-on-IPF release for early adopters could materialize as early as early as mid-2003. And based on our two decades of experience with OpenVMS, we'll award our prediction a Probability Factor of 0.99! µ