Just as Sun Open Net Environment (Sun ONE) was launched in response to Microsoft's .Net initiative, the Sun-led Liberty Alliance was set up to counter Microsoft's HailStorm initiative, which has now been officially dubbed .Net My Services. Like My Services, the Liberty Alliance (that's ProjectLiberty.org, not LibertyAlliance.org, Jerry Falwell's site) proposes a standard format in which to store users' personal information--addresses, credit-card numbers, calendars, and more. This format would be independent of any single computing device and the information stored in it would be used for all manner of Internet transactions.

My Services and the Liberty Alliance are crucial to Microsoft .Net and Sun ONE, these bitter rivals' parallel Web services initiatives. Microsoft seems ready to roll out My Services in relatively short order, while the Liberty Alliance has yet to decide on the shape of the conference table. Yes, this is another sign that Microsoft is ahead in Web services--but in the long run, I don't think that matters much. We're looking at two different worlds, with Java programmers creating XML-wrapped Enterprise JavaBeans on the Sun side, just as Visual Basic, C++, and soon-to-be C# programmers build managed components on the Microsoft side. Not many developers are going to switch loyalties. And each crowd needs a HailStorm-like scheme to get Web services off the ground, because instant authentication and enhanced e-wallet functionality are the building blocks of the next wave of e-commerce.

The degree of interoperability between these two worlds is the pivotal question--one accentuated by the My Services-Liberty Alliance dichotomy. With either one of these schemes, a user-controlled "identity" service discloses and gathers data in a specific format on the user's behalf. It would be tragic if the Internet split into sites that support only My Services and sites that support only the Liberty Alliance. But it's also highly unlikely. Instead, think Visa and MasterCard, with the vast majority of venues supporting both.

Yet the level of interaction between identity services and e-commerce sites promises to be much more complex than credit checks and payments. Arrive at a music site, for example, and if you've given the authorization, you'll be logged in automatically and your stated musical preferences as well as all the music you've purchased online will be fair game, resulting in a list of appealing titles you've never bought. If there are two different identity types with two different ways of storing that history, then next-generation e-commerce sites will have to shoulder the big technical burden of supporting both.

Presumably, both Microsoft and the Liberty Alliance plan to extend their identity schemes beyond the consumer sector to the corporate one, so employees can conduct online transactions with customers, partners, and suppliers. Here again the stage is set for parallel worlds. Through its Active Directory Services Interface, Microsoft plans to hook into its Active Directory server software for local area networks. For its part, we can assume that the Liberty Alliance will ignore Active Directory and use Java Naming and Directory Interface to hook into Sun's iPlanet Directory Server or Novell Directory Services.

It's tempting to ask why Microsoft and Sun can't just get along and agree on the same identity standard. Each has made the grand gesture of inviting the other to join its own e-commerce "federation"--but those were merely schoolyard taunts. The incentive is too great to add proprietary advantages to each scheme. After all, consumers and businesses will have to pay something to use identity services and vendors need some way to differentiate.

We can't realistically expect Microsoft and its rivals to shun competitive advantage by not using different schemes. But we can demand that as these schemes progress, standards emerge that guarantee a base level of interoperability. On the corporate side, an XML-based interoperability scheme is already emerging in the Directory Services Markup Language, a standard developed by Bowstreet that promises to establish a standard format for exchanging local directory information. The consumer identity concept embodied by My Services and the Liberty Alliance is too new for an XML committee to tackle yet. But let's keep the pressure on. Any enterprise with a consumer-facing e-commerce venture should be excited by the customer relationship benefits of identity services. And interoperability standards that make interfacing with those services as smooth as possible should be part of the deal.

Are identity services on the radar in your enterprise? E-mail Eric or Talk Back below.