VMS ecosystem revisited

vax 1.jpg

Dear Meg Whitman,

Do you take trains, drive cars? Do you ever need health services? Do you have good furniture? Sleep well near nuclear power plants because they are well contained? Do you buy things in supermarkets?

For every dependable service or product, and in many other contexts, somewhere there is a well-controlled software, running on a well-controlled Operating System – OpenVMS.

These software products are the result of decades of precise programming, inscribed in precise coding imperative for such functional necessities. A majority of them use functions specific to OpenVMS and still run on OpenVMS, as these custom features are hard to find elsewhere.

In June 2013, however, it was announced that OpenVMS would be reaching its end-of-life and no longer be used with new hardware, with all support ending around 2020; this is a nightmare for all of the companies using OpenVMS.

Hewlett-Packard has added to these difficulties by making the whole community wait for information for over a year. Rumors, small trickles of information to chosen customers, and insignificant steps announced as “last concessions” have characterized this period. Not only are big companies anxious about functional risks and huge investments, they have also been immobilized by over a year of expectation.

Why has this happened? In June 2013, HP made a decision that failed to comprehend the particularities of the OpenVMS market. The lifecycle for OpenVMS sites is ten or twenty years. Profits that were once high have declined in recent years. Product quantities have been suffering as well. HP, on the other hand, has been focused on laptop and desktop computers, on entering the “cloud,” and proposing new hardware to generations playing with large quantities on a very fast-paced market. OpenVMS has become an insignificant accessory by comparison, unprofitable and old-fashioned.

The IT world, however, is more complex than this picture leads us to believe. Dependable systems are not always the most profitable, and many important human and business needs depend on them. The foundations of a house are not as interesting as a beautiful facade; but without them, the facade cannot exist.

It is impossible to think that HP could not actively maintain support for dependable systems that are such an important part of its portfolio. OpenVMS is one of its most crucial systems in quality and industrial impact. Like other major IT players, HP must adapt itself to different speeds and varying policies between large, fast-paced markets and specialized, slower-paced markets.

It seems contradictory to the HP image to sweep dust under the rug in this fashion. Yes, the decision was a complicated one, as it pertained to one of the last OS systems running on Itanium –itself a source of problems – but the wrong decision was made. Why force hundreds of companies to wait for over a year for a decision so crucial to their futures, freezing industrial decisions in very important economic sectors?

A truly justified solution must be found and explained with transparency.

It seems HP inherited OpenVMS by accident. And, by the same sort of inadvertence, was prepared to let it die. HP should, on the contrary, integrate it into a special, innovative proposal for slower-paced ecosystems, thus preserving system health, security, and investments.

When you discarded the previous decision to leave behind laptop and desktop computers, everyone thought it was a realistic decision. The craziness of our times shows itself in ideas such as IT companies being maintained without computers, industrial profits being made without factories – some hyper cloud-based, brave new world. You, however, proved that HP is a concrete company.

The same goes for OpenVMS: although it seems merely to be a strange legacy from the East Coast and MIT, it is a concrete ecosystem that HP has a vested interest in maintaining. Please, accept the challenge of maintaining one of your best sustainable systems. Perhaps OpenVMS will ultimately outlive today’s modern operating systems, as it is robust, simple, easy to maintain, and dependable. Your present decision will be rewarded in the future.

In concrete terms: OpenVMS effectively functions on Itanium i4, so please validate it. OpenVMS is your future long-term Itanium i4 market, perhaps outlasting even HP-UX. Additonally, OpenVMS must be ported to future sustainable hardware created by Intel or its competitors. A realistic pool of competence for OpenVMS must be maintained either inside HP or outside HP with its help. A real effort to port modernization frameworks or interface frameworks to OpenVMS must be made.

We know that some of these components represent money and investments, and that neither quantities nor margins are currently very high for OpenVMS. But you cannot compare fast-paced (uncertain) markets with low-speed (certain) markets. Economics is a game where keeping some extra cards in your hand might appear unnecessary at the time, but prove worthwhile in the long-run and guarantee a huge final victory. As we know, you are an excellent player.

Don't hesitate to send us an answer, it will be an honour to publish it on our site.


Gérard Calliet, for HP-Interex France

[[To OpenVMS community : This Open Letter is an initiative of HP users club in France. Please send us your critics, comments, suggests and / or supports Cette adresse e-mail est protégée contre les robots spammeurs. Vous devez activer le JavaScript pour la visualiser. ]]