In fact, if you look at the issues being debated about what Microsoft can and can’t put into its operating systems, it’s only a comparatively minor skirmish in a major war that has yet to be fought over the role software companies will eventually play as we move toward next-generation web services.
If you look at the core concepts behind web services, it’s a fairly intriguing idea. Instead of a world dominated by packaged applications, software would become a series of components that developers could call across the internet. So instead of everyone building their own billing application, people would make use of a billing component that already exists on the internet, either by downloading that component or simply linking to it over the internet.
Conceptually, this is a great thing that has the potential to completely reinvent the software development process as we know it. But like all watershed moments in the development of technology, it creates opportunity for havoc.
For example, allowing developers to link components over the internet to build applications is only a good idea if some degree of trust is established concerning the quality of the components being linked. Truth be told, most developers think they can write something better themselves than other developers can.
If the whole concept of web services is going to work, there has to be a repository that helps people not only find components but also evaluate them. Now the industry as a whole is pushing the emerging UDDI (universal description, discovery and integration) specification as a solution for finding components of the web, but each individual vendor envisions creating a repository of certified third-party components that users of their application development tools can expect to be high-quality offerings.
This is a good and necessary thing, but it will also fundamentally change the nature of the relationship between companies that create software and companies that use software to run their business, because most of the web services pricing models that are being talked about are based on a usage model. So rather than buying an application development environment to create software that your company then owns, your company will be using software on an ongoing basis that was created by somebody else. So the companies that provide these components will be a lot more to you than just suppliers of software. Instead, they are going to be providers of ongoing services your business is going to depend on for every transaction. And the problem that raises is companies providing services tend to want a piece of the overall action rather than just a flat fee for using their software.
That is what’s troubling about Microsoft’s traditional sense of eminent domain, and specifically some of the wording in the pricing of its HailStorm offering based on Microsoft .Net. Given Microsoft’s tendency toward expansionism, HailStorm leaves open the possibility that Microsoft wants to be at the core of a set of consumer and business services that people would pay to use to conduct business. This is akin to General Motors saying that instead of just building cars and trucks, it now wants to build toll roads.
Previously, Microsoft said it would not require a vig (a gamblers’ term for taking a percentage of a bet to facilitate the placing of the bet) from people using Windows to access the internet for e-commerce. But at the same time, Microsoft is suggesting it will require a vig for using its services. So before your company signs up to participate in a federation of services managed by Microsoft, you’re going to need to figure out how comfortable you are with Microsoft as a business partner, rather than just a supplier of software products. After all, there’s a world of difference between the terms business partner and supplier.
Vizard is the editor-in-chief of InfoWorld. He can be contacted at firstname.lastname@example.org.