What does Apple and Sun moving to x86 mean for Windows?

Now that the debate for the most part is over about the ability to run mission critical, consumer and business apps on x86 (some may have called it Wintel in the past which is a little unfair to AMD) is over, the industry has pretty much stated x86 is a viable platform for running mission critical applications and consumer applications (like video, animation, and other multimedia applications), the question is what does this mean for Windows?

Microsoft has had a tough up hill battle to fight moving into the enteprise space, especially the data center, for several reasons a lot has to do with the perception about the enterprise capability of x86 and Windows (a.ka Wintel) platform.  I think it will be important for Microsoft to attract mission critical customers and ISV’s during the Vista/Longhorn time frame.  There is a difference between enteprise applications and mission critical applications, how many insurance companies bet their large back end policy and record keeping applications completely on Windows, how many manufacturing companies bet the plant operations on Windows, and how many hospitals, banks, and major hotel chains put there most mission critical apps on Windows?

I understand a lot of these applications are 30-40 year old (sometimes older), mainframe or other host based systems that just keep running and running, it is sometime very hard to build a compelling case to get companies to spend several million dollars to move.  However the ones who have moved off generally move very little if any functionality of from the host based systems to Unix or other big blue platforms, such as the 400.  The ones that are consider moving have CxO’s not willing to place big bets on Windows because they do not wish to be the first in their industry to do so.

Now that Linux, Apple, Sun (and others) have validated the x86 platform as a enteprise player, Microsoft now needs to position itself to gain the trust of CxO’s to win the mission critical applications.  Let’s see what the Vista client and Longhorn server have to offer to mission critical customers.

Comments (1)

  1. Ron McMahon says:

    You are really comparing apples to oranges. The hardware platform on which ‘mission critical’ systems run is not and should not be bound to a software platform. After a quarter of a century in this industry, I can say that the fear of moving to Windows for any manager of a mission critical system is not who will be ‘first mover’, but rather; is the move worth the time and cost and risk?

    Hardware prices continue to plummet, even in the mainframe space – therefore improved application performance is simply a CPU / system hardware upgrade away.

    Rewriting application software on the other hand is a mixed-bag at best. When you do this you have an opportunity to eliminate design dead-ends. You may even have the potential to optimize a database design or workflow process, although the more change that is integrated with a migration, the greater the risk of failure.

    The real gotcha is that with any software platform migration you end up almost starting from scratch – and really risk introducing new bugs that do not exist in the ‘legacy’ application. Also, by moving in to a new technology you are walking away from the huge investment (in time, money and code quality) that you’ve made over what is often dozens of years.

    Hardware is easy to replace and upgrade, and 99% of the time its implementation is virtually invisible to most end users. An application upgrade, however, is a sea-change for everyone who interacts with the system (not just end users).

    The real question isn’t as simple as "Can Windows be trusted to replace a legacy system?", rather it is one that any CIO worth his salt will ask "Does such a move add a POSITIVE net result to my company’s bottom line?" For a very large number of companies, the answer is a resounding ‘No!’ and no amount of sexy buttons, translucent windows or even drag-and-drop capability will rescue such an analysis.

    This value calculus will only be overcome when migration of applications and all software between disparate operating systems is easy, automatic and transparent to an end user. Such system capability is so distant from what our computing infrastructure can accomplish today that such a suggestion may seem to you as impossible, wishful thinking or science fiction. Perhaps. I’m an optimist, and I believe that at some point in time in the next half-century we will see such portability – perhaps someday we’ll be able to move our software tools between OS and application platforms with the same ease and speed with which we can drag-and-drop data from one application to another today. No rewriting of legacy applications, just drag it from your old platform / OS to a new one and *presto* it just works.

    When this is achieved, Microsoft will have customer’s trust.