Take a closer look at the buzzword of the month. Enterprise Pursuits 4/10/01 Nelson King Net Services
Here we go again. The buzzword of the month is “Net services” (or “Web services” or “e-services”). I actually attended a meeting where a board member told the CIO that the company had better check out Net services. The CIO didn’t look offended, but he did look uncertain about how to proceed with the board member. I think a lot of people feel that way about Net services.
It’s very easy to think that Net services are nothing more than old wine in new bottles. Take any old application service provider, bust up the applications into some smaller bits, and provide an interface so end-users or programmers can use the pieces–presto changeo, Net services. Sorry to say, a lot of Net services will be just about that creative. But this is not the real story.
How do you spell I-N-T-E-R-O-P-E-R-A-B-I-L-I-T-Y? It’s been one of the most ungainly–and unattainable–words in the IT lexicon. Simply put, life would be a lot better if all of our hardware and software could freely interoperate. Of course, life is not simple and perfect interoperability isn’t even a dream. But what if we could really get Net services to interoperate? That would mean, for example, that an IT shop could go to six or seven Web sites, sign up for various application components–a billing component there, a customer data component here–put them together, and lo! They become a new–working–application.
The cynics among us (and we know who we are) will start clucking about “pie in the sky” dreams for components. All we have to do is remember various experiences with Java applets, ActiveX and CORBA components to realize that we’ve been promised things like this before, and that the reality has been something less than scintillating.
I’ll admit that there’s plenty of room for cynicism about Net services, but a cynic may be too busy looking for worms to see the apple pie. There are some things that are different about Net services that may be tentative or even fragile; they still could mean success. Let’s start with who’s doing it.
At the head of the parade is Microsoft. This is unusual. Like the IBM of old, Microsoft never leads into a new market. Except this time they are leading big time. The instrument of their involvement is the .Net initiative, which for the most part is about developing, deploying, and utilizing Net services (Microsoft calls them Web services). IBM, Oracle, HP, and Sun Microsystems have joined the parade. It’s an impressive amount of muscle and it makes you wonder if they don’t really see something in Net services that they like.
What they see may well be standards that they can stick with. It’s an alphabet soup, but ever so tasty: SOAP, XML, UDDI, HTML, and WSDL. Each piece of this obviously complex set of standards and protocols contributes to the possibility that Net service components will, in fact, interoperate. If this works as it’s supposed to, then consumers will be able to rent various pieces of software over the Internet, instead of buying large software programs that have to be upgraded almost yearly. Corporations can rent or lease specialized applications for almost any business purpose imaginable. Corporate developers can access a vast wealth of services that can be cobbled together into new applications.
I’m not sure that Net services qualify as the greatest thing in software since sliced Wonder Bread, but the scale at which it’s being approached is unprecedented. Certainly for the software vendors, Net services are a Good Thing. It means they’ll finally get free of the cycle of update and have a much steadier revenue stream from rental and licensing. Whether there’s a flip side of savings for their customers remains to be seen. I have a feeling that this month’s buzzword will be worth revisiting in future months.
Editor at Large Nelson King also writes Pursuits monthly for ComputerUser magazine.