Around two years ago I asked why there weren’t that many successful Java ME consumer applications that weren’t games. I went on to conclude that the few that were successful were just a window on more open, capable and consistent processing somewhere else.
It’s interesting to see that a number of development frameworks for several development platforms have evolved around this model of having a view on what’s essentially mainly happening at a server somewhere. In fact, even without these frameworks most platforms (except Java ME) have some kind of webView that can be embedded within an application.
WebView apps give many of the advantages of full native applications (use of phone APIs/features and discoverability via app stores) with the main disadvantage that the application itself needs a data connection at the time the user uses the application. Also, using a web oriented interface sometimes isn’t as usable as an equivalent interface made up of the phone’s native controls.
Sometimes it’s possible to use a webview instead of a plain text view to show application information. This allows information such as help or troubleshooting to actually reside online somewhere where it can be changed easily. Alternatively, it allows advertisements to be easily embedded in applications results because most ad systems are oriented towards embedding code in web pages.
I have done a few feasibility studies recently where a webView type application made a lot of sense. They tend to be applications that people will use for the specific unique content as opposed to those that need to compete with other applications on features or user experience.