The theme for yesterday evening’s Mobile Monday London was ‘Mobile Entertainment Applications’. It turned out to be less about entertainment applications and more about applications in general, widgets and the perennial web vs app argument.
First of all, Patrick Mork from GetJar gave us some interesting statistics on GetJar…
- 1.2 million downloads per day
- 75% of users are under 25 yrs old
- Email and Instant messaging are the most popular categories, followed by maps (presumably Google Maps) and Opera Mini
- 70% of downloads are Java ME.
- The top growing category for popularity is Games
- GetJar doesn’t have an on-device app to discover applications but is working on this
Patrick played on the fact that 70% of downloads are still Java ME. However, I’d say that this isn’t surprising given that GetJar only had Java apps in its initial years and after all, it is called GetJar and not GetSIS, GetCab or GetApp.
Interestingly, GetJar now has a shortcut mechanism called a Web App, that’s a small application that just launches a given web site in the mobile web browser. The argument is that this aids discoverability and that users don’t really care or sometimes even know if they end up using an application or web site.
A panel session followed with Patrick Mork from GetJar, Andreas Constantinou from Vision Mobile, Joachim Hoffman from Fjord, Mark Curtis from Flirtomatic and Daniel Appelquist from Vodafone.
Dan revealed that Vodafone will soon be providing the capability to sell widgets and also to sell within widgets. Despite the fact they will be widgets, they will be marketed to consumers as applications. Now that was two instances in one evening where I had heard of non-applications masquerading as applications!
There followed a discussion on whether the future would be apps, widgets or the web browser. This tied in well with Google’s Vic Gundotra recent comment that browsers are the Future.
A few years ago we were told that widgets or the web would solve application fragmentation problems. At the time, I commented that the majority of applications I create need facilities that will never be available from a web runtime. Furthermore, even if these facilities were available (incidentally, 3 years on they still aren’t), exposing them would expose differences in the underlying capabilities that themselves would cause fragmentation.
I have commented on inherant fragmentation within frameworks and backward compatibility of widgets. Three years on and a Vodafone Widget isn’t compatible (not packaged the same, different phone APIs and not interchangeable across app stores) with a S60 DevKit widget and Samsung have very recently released yet another widget format.
However, we can hope that the state of widgets will improve with major players’ (including Vodafone’s) participation in W3C and OMTP BONDI (API) widget standardisation. The big question is whether the existing widget runtimes will migrate to a common format and API.
However widgets and browsers evolve, I still believe there will always be applications, widgets and the web, all used according to their respective strengths and capabilities.