Yahoo! has done a lot of things right with its oneSearch,
but it should be cautious with sponsor messages.

The 2008 edition of the annual CTIA wireless show is history. As expected, there was a lot of buzz regarding 4G: Is a new technology holy war looming between WiMAX and LTE? The results of the 700 MHz auction (with the non-traditional players again being pretty much non-factors) also dominated many animated discussions among exhibitors and attendees.

But if there was a theme to this year’s event, I’d have to say it centered on the perceived opportunities and potential pitfalls of the emergence of “open” wireless networks. I’m not suggesting that everybody was talking about the dawn of a new era for the industry, but the tone and scope of exhibits and flashy demos suggested that many companies are thinking that way.

In particular, I noticed two things at the show that focused my attention. One was the size and eye-popping appearance of the exhibits promoting handsets and other user devices. Open networks suggest a trend away from the traditional role of wireless operators as the dominant customer for handset vendors, and it’s pretty clear that the big names like LG, Nokia and Samsung are gearing up to market more directly to end users.

The other factor I particularly noticed was that some big players have obviously spent serious money on development of new mobile data applications. That’s not something that they were as willing to do in an environment where carriers might arbitrarily decide to act as gatekeepers.

HELP FROM ONLINE
One benefit of this new investment in open network opportunities is that highly experienced providers of online services are contributing their expertise to addressing the two key factors that have heretofore frustrated the quest for mobile data “killer apps.”

The first, and perhaps most important of these is the user interface – or more accurately the lack of a robust user interface – in mass-market handsets. Higher functionality devices like Blackberrys and iPhones may grab all of the attention, but more “vanilla” phones with basic user interfaces still account for the overwhelming majority of devices in service. Clearly, you’re not going to get 100 million users for an application that only works well on a few million iPhones.

The second problem that has been the undoing of many mobile applications is that the throughput performance of mobile data networks has been, and likely will continue to be, highly variable. A 3G network that delivers download speeds of “up to” perhaps 1 Mbps may provide only a small fraction of that performance with the sort of coverage reliability that users are accustomed to for voice service. That really puts a crimp on applications that require a high volume of data transfer between the handset and an application server – something that many developers have failed to realize.

SEARCHABILITY
Perhaps no single exhibit or demo at CTIA’s Wireless 2008 more fully captured the new thinking about open network opportunities, and how best to address them, than did Yahoo! with its new “oneSearch” voice-activated search engine.

Voice recognition has been a mainstay in wireless handsets for a number of years, but Yahoo! has taken it to a whole new dimension with oneSearch. The system recognizes not just a limited vocabulary but entire spoken phrases, which form the input to the search engine. Then, results are presented in a format that works well within the constraints of the tiny screen of a typical handset.

If the desired information is among the offered selections – and based on my demo of the system, it usually will be – the handset’s up/down keys can be used to highlight it as the starting point for obtaining further details. Operation is fast, intuitive and most important, does not require any of the agonizing 10-key “typing” that is the most frustrating element of basic handset user interfaces.

DATA LOADS
Yahoo! also has demonstrated with oneSearch that it understands the limits of wireless data network performance. The enormously complex processing required for the voice and phrase interpretation that drives the search engine can’t practically be loaded into a handset, so the voice signals themselves have to be transported over the data network to a server. But software resident in the handset does provide data compression so that the required throughput on the uplink is minimized. After that, downlink delivery of search engine results and uplink requests for more information will generally require relatively small amounts of transmitted data.

The search engine part of oneSearch is oriented toward providing information of particular value to mobile users. For example, say “United Airlines 256,” and oneSearch will offer the departure and arrival status of United Flight 256. Say “Mexican food close,” and if the handset is enabled with GPS autolocation, oneSearch will offer a list of nearby Mexican restaurants.

As with other Yahoo! Services, oneSearch will be offered free to users with revenues derived from advertising. It’s here that some issues may come up. Contextual ads associated with a search, for example where sponsors’ information is displayed first, should be fine. But ads with more content or interactivity could run afoul of the same limitations that oneSearch has seemingly addressed so well.

That’s something that Yahoo! will have to be wary of. Developers of mobile data applications have learned, sometimes painfully, that user satisfaction is a fragile thing, with the data pipe and the user interface easily strained. These are realities that the opening of mobile data networks will not change.

Drucker is president of Drucker Associates.
He may be contacted at edrucker@drucker-associates.com.