Just about everyone is talking about Web services. Microsoft, as usual, heads up the lobby but even the likes of IBM and Sun are joining the movement.
And it is not only limited to the traditional "big boys" when it comes to Web services. Just a month ago I spent five days in Chicago with Progress software developers who could talk about little else.
Web services will survive in existing trusted relationships only
Alastair Otter, Journalist, ITWeb
However, despite the overwhelming lobby in favour of Web services, there is what appears to be a complete lack of understanding as to what they are. Perhaps a few of the initiators of these all-capable pieces of code have a better idea than most of what they are expected to achieve. But the marketing departments of most large software corporations have been running wild with the concept for the best part of a year now and the result is a garbled definition that grows daily to assume an ever-increasing role in the future of computing.
Take for example the idea that Web services are small pieces of code that do just one thing, and that together a collection of these "services" are able to make up a full-blown application that does exactly what you want. Farfetched at best, this is the kind of definition that makes Web services less believable than they should be. After all, if you`re running a business, are you going to trust your essential business processes to these services, downloaded off the Internet on the fly? Not likely. Particularly because when something does go wrong, who are you going to call? Who`s actually responsible for the corruption of your data? The creator of the Web service? Good luck.
Of course there is an argument that there will be trusted repositories of services that can be relied upon. Great, except for the issue of who controls the repositories? Recent history suggests that Microsoft would, which is not going to be popular with other software vendors. And if we spread these repositories around the major players, we`re back where we started, so this is a no-win situation.
Perhaps it is time to be realistic about what Web services can deliver. For a start, they are not a panacea for industry-wide computing. In all honesty they are initially likely to survive only within existing trusted relationships between vendors and clients. This way, if there is a problem, there are existing channels for follow up. They are also very unlikely to be generally available for consumption by general computer users. And in fact they may never reach this stage, despite what the marketing machines would have you believe.
No solve-all
Taking off another layer, it is also possible to argue that Web services are not small, task-specific applications at all, but are in fact snippets of XML-based text floating around the Internet. I would argue that this is a simplistic but far more accurate description of Web services.
Interestingly, despite the fact that some of the standards underpinning the Web services strategy are not yet in place, including the UDDI standard for locating Web services, most software vendors are already fully subscribed to the Web services bandwagon. There is, however, still some way to go and still not a lot of agreement on standards.
Stripped of the marketing jargon, Web services may still make some impression on the world of computing. Particularly because Microsoft does not throw R&D dollars down the drain and will no doubt make sure that Web services work, no matter what. But in the short-term they are not the solve-all for current computing problems that the heavyweights would have us believe.
Share