There has been a new development since I mentioned Wireless Application Protocol a few weeks ago that may signal significant changes in WAP service offerings. WAP, as originally envisioned by service providers, may even be on the ropes.
WAP was developed by the WAP Forum (www.wapforum.org), which got its start in 1997. WAP was developed to support IP-based services over low-speed wireless connections to cellular phones with very small displays, slow processors and limited battery lifetimes. The developers made some design trade-offs to support these constrained environments. The primary decision was not to support native Internet applications such as HTTP and Secure Sockets Layer, but to develop special versions that were better suited to the conditions in the wireless world of 1997. This decision meant that there had to be gateways between the WAP parts of the world and the rest of the Internet to translate between the WAP version of the protocols and the Internet-standard versions.
At the time this approach may have looked like the best - if not the only - way to bring Internet access to low-speed wireless devices. But one side effect of using gateways just landed France Telecom in court.
Although WAP does not require it, most wireless phone companies have assumed they would preprogram the address of one of their own gateways into the WAP phones they sell. There are good reasons for phone companies to want to do this. If they run the gateway, they get to charge the user for use of that gateway. If they don't, they risk becoming a commodity IP connectivity provider. There is even IP-telephony technology that would let the user bypass the phone company gateway for voice service, eliminating the major source of revenue for these companies. But a French court has just decided that locking the user into only using the provider's gateway violates competition rules, forcing France Telecom SA to let the users point to any gateway they want. This may have a profound effect on the business model needed to support this type of mobile service.
Another byproduct of the use of gateways is actually harder to deal with. No court can fix the problem that gateways by their nature inhibit the deployment of new applications. If a new application needs to transfer data in a way the gateway does not support, then the application cannot be deployed unless the gateway is updated. Gateways create walled gardens that block the view of the rest of the world.
These problems - along with newer integrated circuit and battery technology, and the higher bandwidth that will be available with the next generation of wireless devices - may make WAP a suboptimal solution without the benefit of user lock-in. We may just see real Internet service to new wireless devices.
Disclaimer: Harvard does not have to use lock-in to keep its customers, and the above is my observation.
Bradner is a consultant with Harvard University's University Information Systems. He can be reached at firstname.lastname@example.org.