Dissecting Java Portals
By Carlo Longino, Fri Jan 14 18:00:00 GMT 2005

Many content providers are looking to branded, dedicated Java applications as a way to increase their exposure on handsets and to avoid having to go through carrier portals. But there are plenty of pitfalls that must be avoided.

Lots of media companies see Java portals as a great way to establish a solid foothold in the mobile market. It sounds perfect: a branded application that sits on a user's phone, taking them to your content, where you've got total control over what they see and how to charge for it, with no interference from an operator and no revenue sharing.

But it's not a straightforward process, and Tom Hume does a nice job of pointing out some of the problems. His first point is a salient one -- can content providers convince users to put their brand on their phone? This might be easier for news outlets, where if I read a certain newspaper everyday, I'd be okay with having it on my device, if it makes getting to its content easier. But the phone is an immensely personal object, and one that people use to project an image of themselves, hence the market for ringtones, faceplates and backgrounds -- and how many brands have the cachet to speak for their users via their mobiles?

But Hume's main argument undermines a key assertion of content providers using Java portals and the companies selling them: that they offer a better user experience. He's pointed out before that a Java app can be better for the user than WAP, but that can is very different than will. He tried one particular application that's a Java storefront for games and ringtones and the like, which was beset with usability issues: it took three tries to open properly, it was slow and it had to connect to the network to display anything beyond the name of a Java game it was selling.

He adds that for applications being billed as offering greater speed and ease of use than WAP sites, it's not always the case, and serves to highlight the fact that usability is a fundamental aspect of mobile applications and services, no matter what a no-brainer some things are hyped to be. Hume sums up his experience by saying the app "just feels like a cheap attempt to squeeze out cash" -- and for all the good Java portals are supposed to do for brands, is that a good impression for the user to walk away with?

Just as Hume points out the difference between "can" and "will", there's a big difference between "can" and "should". There are instances and brands for which Java portals make sense, but there are still plenty of services -- most, even -- that would be better served by putting more effort into improving WAP or XHTML sites. But even in those instances where Java portals fit, there's a huge leap to go from concept to proper implementation.