It might be almost as old as Unix but OpenVMS is no lumbering dinosaur, not when it can continue to run rings around Unix, Linux and Windows for security and clustering - which brings OpenVMS a whole range of advantages that others can't match, not even the IBM behemoths. And yet for all its quality, HP has continued to leave it in the cupboard.
Shortly after the merger, Carly Fiorina made a visit to OpenVMS headquarters and to an enthusiastic crowd she professed her admiration for the efforts of Ken Olsen and Digital. As far as I can ascertain she has not been back there in the last 12 months, and HP has been almost silent about OpenVMS.
Some say that the recent set of management changes at HP bode well for OpenVMS but in reality all it has done is only put those who were responsible for it at Compaq back in charge of the product at HP. They did almost nothing at Compaq and now it looks like they will do nothing once again - nothing that is, except continue to test the patience of OpenVMS owners and see them move to other platforms.
Until the management reshuffle, Scott Stallard was responsible for Business Critical Systems, the section which includes OpenVMS, but a search of the web indicates that not once in the 12 months of his tenure, did he say anything positive about OpenVMS. In fact, he rarely mentioned it except in composite statements with other operating systems from HP.
In a report here in the INQUIRER about the consequences of the merger, he speaks at length about the other operating systems but all he can manage to say about OpenVMS is that it will move from Alpha to Itanium and that tools will be available for any OpenVMS site that wants to move to HP-UX. He was also absent from the major European user forum in Lyon in May 2002, sending only a video-taped interview where he assured customers that the new HP intended to continue to aggressively promote and sell OpenVMS-based AlphaServer systems as long as customer demand exists (my italics), something that OpenVMS is still waiting for.
So much for the support of OpenVMS from senior management in HP!
Following the reshuffle earlier this month, Peter Blackmore has taken over the role from Scott Stallard but this move brings no confidence either because Blackmore and CTO Shane Robison - he of a finger in every technology pie at HP - seem to have been largely responsible for OpenVMS going nowhere when it was under the stewardship of Compaq.
While one could make allowances for their lack of visible effort at PC-centric Compaq, the fact is that these two have held influential positions at Hewlett Packard for 12 months and in that time have done nothing for OpenVMS.
Robison is on record in a recent interview as saying "We have a very good, three-operating-systems strategy, with NT, Linux, and HP-UX." So much for NSK (aka Tandem) and so much for OpenVMS. And probably too bad for the sales prospects of NT and HP-UX and their associated software -- it has all been sacrificed for Linux, which itself brings little or no continuing revenue stream.
Linux would be great for HP - but only if all that HP ever wants to be is a box-shipper. Robison needs to get out of his "Internet-boom mentality" and realise that profit is not a dirty word and that OpenVMS generates decent profits.
Blackmore is no better. In a press release from HP, he appears to be the source of the comment "HP attributed its market-share leadership to continued customer loyalty and a powerful line-up of cross-platform offerings that provide customers a choice of Windows, Linux, UNIX and specialized operating platforms." (my italics)
I don't know where Blackmore has been for the last 25 years or how little he knows about the products that he deals with, but OpenVMS has always been a general-purpose operating system ever since its release in 1978. It has been used for a wide variety of commercial, industrial and scientific applications - or does Blackmore believe that this breadth of applications still makes it "specialised"?
The only area that VMS could even remotely be said to specialise is in providing high-availability and reliability. It is a sorry world if Blackmore and those associated with him believe that this justifies its marginalisation by Compaq and now HP.
He is consistent if nothing else. In an interview in in Computerworld in March of this year, Blackmore said: " We will continue to support it on IA-64. It is a very special class of operating system. We are not planning to do anything else other than to continue supporting this environment, as well as the 450,000 users around the world that are on this." (My italics).
I am afraid that his 450,000 users is nothing more than a wild dream. About 12 months ago we were told of a 10% increase in OpenVMS sales, predominantly in the USA. What was not detailed was the number of customers moving away from OpenVMS -- or preferring to expand with other platforms -- both in the USA and Europe. Let's face it, those customers have given up waiting and new customers are about as rare as hen's teeth.
And if the customers aren't investigating OpenVMS for their applications, then the major blame can be assigned to Compaq's failure to promote it during the few years that it owned OpenVMS.
It is not simply the sales that matter, it is getting information out to the potential customers - and in that respect Compaq was an abject failure, caused in part by the Capellas mantram of not confusing customers by giving them a choice.
The paucity of information about OpenVMS is reflected in its scant appearances in the various technology publications. Use the search utilities at a number of major IT publications that have a web presence and you will see just how little information about OpenVMS gets into the public domain. VMS was renamed OpenVMS about 12 years ago but very few web publications mention it more than 35 times, which is an average rate of about three times per year.
Compaq dictated that OpenVMS would be aimed only at very specific target markets but the search tools at the websites of a number of appropriate publications for those markets returned almost no references to OpenVMS. So much for market penetration and so much for establishing a powerful technology presence in them - but these are the markets for OpenVMS that HP has persisted with in its half-hearted manner.
It is clear that OpenVMS does not have a future at HP, at least not a future that does it any credit.
As Peter Blackmore has made abundantly clear, HP is busy concentrating on Linux, Unix and Windows. Never mind that these are low-margin products that channel money into other pockets and that produce little on-going income. Never mind that Linux takes sales from HP's own variant of Unix as well as from its other software and operating systems offerings.
The just-announced quarterly financials show the folly of this kind of thought. OpenVMS is part of the Enterprise Systems Group and this was the only product group to increase revenue in the last three months and the increase in profits was the greatest by a long shot. On the other hand the Personal Systems products returned a profit of less than 0.5% for the more than $5 billion spent on them, which less than the interest rates at most banks. With their marketing emphasis on the low-profit profit lines, HP is looking a lot like Compaq Mark II - and that is not at all a good sign for OpenVMS.
With this kind of attitude in HP it is high time that they decided what to do with OpenVMS - whether to drop the sophistry of market niches and sell it to the IT community at large, or to sell the OpenVMS division in its entirety.
I would be very surprised if HP really decided to put effort into selling the product to the broad IT community. For far too long Blackmore, Robison and others have had the effrontery to assert that there is no potential market worth exploring outside the niches that they target it at, niches that Compaq and HP have discovered are highly vulnerable to competition and to shrinkage through company takeovers, moreover niches that Compaq simultaneously targeted with Unix and NSK products even while claiming that these were the markets for OpenVMS.
I believe that it would be expecting too much of Blackmore and Robison to suddenly change their tune now.
The only sensible option for all concerned is for HP to sell OpenVMS - and in a single step reduce its costs, gain some money to waste on low-profit ventures and keep the stockholders happy.
The question is of course, who would buy it and who has the kind of support to ensure that the underlying hardware continues to function?
Dell might like it as an entry ticket into the high-end of business computing. Unfortunately for Dell, its hardware support at that level is still growing and it would be hard-pressed to have credibility in the commercial enterprise sphere.
Sun might be interested if its own financial position was more favourable because it seems to have a sneaking admiration for OpenVMS. Sun has all its eggs in very few baskets and it are under a lot of pressure at the moment - perhaps even themselves a take-over target - so its thoughts are likely to be elsewhere.
The most obvious buyer for OpenVMS would be IBM because its acquisition would let it stitch up the high-end commercial market as well as make useful gains in other markets. Make no mistake, OpenVMS has features that IBM systems still cannot match, and I am sure that IBM would take advantage of them, and it is not scared of proprietary operating systems. IBM also has the required level of hardware support and has good credibility in the business world, as well as in the important world of the US Defense department.
Sure, such a sale would probably require modifications to run on specific types of processor, which would most likely be the Power4, but perhaps with either Intel or AMD as wildcards. Unlike the transition from Vax to Alpha, these changes would be relatively simple because as part of the move from Alpha to Itanium, changes have been made to reduce the effort of working with processor dependencies.
Whether HP would actually sell it to these competitors is another question but arguably if HP sees so little value in retaining OpenVMS - and its unwillingness to attempt to increase their profits from it suggest this is true - then it should have no compunction about taking the buyer's money.
And after about 10 years of half-hearted support from the various vendors, I guess the OpenVMS community would be more than willing to take the risk. µ
Thanks to a hard-coded Nvidia Tegra X1 flaw
Time's up. Me too. Not him
Redmond says 'the fix is more complex than initially anticipated'
And, yep, they're really expensive