As usual in the IT industry, hype is running far ahead of reality but that doesn't mean to say that we are not approaching a watershed from which good things will really advance. Let's go back a bit and recap on the whole Web services thing. What are they and what are they supposed to enable us all to do?

Single language 

For the uninitiated, Web services are business and consumer applications that use a set of shared standards and protocols allowing different systems to interact with each other, sharing data and services, without the need for manual intervention or translation. These data "conversions" allow applications to simply bypass operating systems, programming languages and middleware. 

But aren't we doing this already? Well yes, to a lesser degree. We have interfaced systems and sometimes they work well but often times they don't. With Web services, it will now be done using set standards, promoting much broader interoperability and platform independence. Steve Ballmer, chief executive of Microsoft says Microsoft's Web services technology will eventually make it possible to drag and drop data between unconnected software applications. So that ERP system which is not talking to your CRM system that in turn ignores the payroll system will soon be a thing of the past. 

The key to Web services is Extensible Markup Language (XML), a universal format that standardizes digital content on the Internet and promises to provide this wondrous single language that can be understood by all computer systems. 

Here comes the "but"...

But (and there's always a but) XML is no quick fix. It allows data to be tagged with descriptions. However, to make sense of those descriptions, they must conform to a schema. Getting schema to conform one with one another is a problem. It can come down to pure language. Take an invoicing form in XML: one of the tags might describe the amount payable. Say you call that tag "money" and I call it "cash." Suddenly the systems no longer understand each other. Unless we have agreed on the terms to be used in each case, one system will not be able to interpret another's terms correctly. Hence Larry Ellison likening Web services present capability to that of the mobile phone, which whilst making it simpler for people to communicate with each other, hasn't made it any easier for an English speaking person to hold a conversation with a French speaker. If you need to turn to a specialized language reconciliation software to iron out these problems, you can see why it could easily take a decade before these issues disappear. 

However, if it works as anticipated, Web services could not only shrink IT budgets (consultants McKinsey & Co. estimates Web services can cut the time and money needed for systems integration, the largest IT expense, by up to 20%) but create new interactions among businesses and make the Internet more user-friendly for consumers. For example, a customer could be provided with a single interface rather than having to go through multiple parts of an organization or group of organizations. Passwords and other information are stored in Universal User Profiles (UUPs), which give users this kind of access to multiple Web sites and services. Such a development could also crack the long-disappointing customer relationship management (CRM) nut, which has not realized its full potential at most companies and institutions.

Key players

Depending on which analyst house you listen to, fully integrated Web services are still anywhere between five and ten years away. Microsoft and Sun are the two biggest players in the Web services market. Microsoft's .net has become synonymous with the generic term Web services and is an entire strategy including a new programming language C#. But there is also J2EE, which stands for Java 2 platform, Enterprise Edition incorporating the Java programming language, central to Sun Microsystems' own Web services strategy Sun One.

Even Microsoft thinks it could be five to ten years before web services take-up reaches critical mass and last summer Bill Gates was quoted as awarding Microsoft a "C" grade for its development of .net so far. But after a sluggish start, some of the factors that have resulted in confusion and caution among potential users including vendor disputes, standards development and security have now started to die down.

The biggest problems are without doubt security and privacy. The idea of Web services relies on masses of data, usually customer data such as credit cards being shared in the same way across the Internet. Finding a way to prove securely that someone is who they say they are on the Web is probably the biggest challenge and so far the biggest cause of controversy.  Microsoft's controversial authentification engine Passport was to be developed as a single sign on point for a whole load of Web services, but their less than brilliant security track record meant that other parties felt that they should not end up providing the de facto standard for Web services authentification.

Narrowing the gap

The security issue spawned two consortia. The Web Services Interoperability Organization, operated by Microsoft and IBM, while Sun Microsystems' efforts are part of a broader consortium called the Liberty Alliance, alongside HP, Nokia, Sony and RSA Security, as well as some powerful user companies. Both consortia have been seen as being at odds with each other but the gulf appears to be narrowing, with Sun getting behind the Microsoft-led WS-Security initiative and Microsoft reviewing Sun's application to join the board of WSI.

There are other issues for IT mangers to consider. For one thing, sending and receiving XML messages takes up a lot of bandwidth, and requires parsing before processing. This can have a serious impact on the performance of companies' systems and could become a serious problem for the quality of Web services. There are also the upgrades that will be needed because many Web services tools will work only with the latest version of Windows, Internet Explorer and other environments. Food for thought.

That said, research commissioned by BEA Systems shows 54% of European companies have now adopted Web services. I don't know whether to be astonished, encouraged or just plain disbelieving of a figure like that?

Apparently most of the live projects out there are small, internal and still in the pilot stage. It seems sensible that the primary usage for Web services should be to break down siloed information and data within an organization and to shore up back office operations where the processes remain securely tucked behind a company's firewall. Nevertheless, users are getting their hands dirty and we may really be on the verge of the next big evolution in IT.