Open or Walled Garden: Cingular & DoCoMo points of view

The first session at Connect 2006 this morning was almost as controversial as the first session in Prague.  Bob Schechter (NMS CEO) led a panel with:

• Scott Swanburg, Consumer Data Marketing, Cingular Wireless
• David Berry, Sr. Product Marketing Manager, Sun MicroSystems
• Brad Blanken, VP Carrier Development, Third Stream Media
• Voytek Siewierski, Mitsui Ventures (but formerly ran DoCoMo's Java initiative)

discussing the status of the mobile industry in the US.  In Prague, the argument was between operators (Elisa & Vodafone) and Andrew Budd, Vice Chairman of the Mobile Applications Forum on where the new applications will come from – it boiled down to open versus walled garden.  Today's panel had a very different make up and a different starting point, but quickly came to the same argument.

Voytek started it by pointing out DoCoMo's incredible success in mobile data services and data revenue of any operator in the world, a success approached only by other operators in Japan and Korea.  He listed their key features:

• Application technology friendly to 3rd party developers
• Business model attractive for 3rd parties
• Targeting consumers (rather than business)

DoCoMo makes their money on data traffic and on billing services, but has no "On-deck" content of their own.  The result of this open model is 45 million active data users (90% penetration) and 100K+ distinct application sites available to DoCoMo users.

The Cingular point of view was roughly "we will never allow our 3G network to be commoditized."  Scott was polite and talked around a series of justifications for a walled garden (user experience, spam control, etc.).  The only interesting (to me) comment was near the end when he commented that they did have this discussion internally and there was one initiative in the works which he couldn't talk about but which might help.  Given the rest of his talk, I'm not hopeful.

The other key point raised (but not resolved) was the difference in clients – between a standard browser on every PC and the diverse handset interfaces found across the extremely broad range of mobile handsets.  David Berry put forward the thesis that Java-based IMS client frameworks would eventually solve this problem. Certainly Java is the most prevalent handset environment, but whether an IMS framework will be deployed widely enough to matter is an open question.

In any event, the controversy was good – far better than the boring panels at most conferences I attend.