Mobile application testing

According to the statistics generated early this year, there were 222 different mobile devices which were being sold by the various carriers in the US in February 2008. These do not include the ones which were discontinued by the various carriers and the ones which were no longer being advertised in the carrier websites. The number of mobile subscribers increasing from less than 100,000 in 1985 to close to 3 billion in mid 2008 has obviously got in the need to keep the market up-to-date with new devices. As we can see, the competition among the different device manufacturers and the carriers in releasing attractive devices is at the peak these days.

People who are most affected with this device boom are the engineers who develop and certify their mobile applications. If the QA team has to test their applications and certify them on all the available devices or even the most popular ones in the market, it could take months and months before you can release an application to the market. The most common way adapted for testing is by making categories of the devices in the market and grouping the available devices into these categories. QA team then picks the most popular devices in a particular category and certifies the applications on those devices. There are risks involved in this method as the applications behave differently from one category to another and even from one to device to another in a particular category. Identifying these defects in the different devices remain a big challenge.
Will there be an easy solution to this in the near future?