There is no such thing as “killer application for HTML5″. And the “owner” of operating system or ecosystem is not the one that needs to create most popular apps. Best iOS apps are not done by Apple. Best Android apps are not created by Google (with exception of Google Maps or Google Now, but those are very specific data-intensive applications relying on completely separate data aggregation ecosystem, based on e.g. location and mapping data).
HTML5 has its perks. You don’t need to install it as native, it is cheaper, and HTML5 standard freeze will eventually ensure that it will be performance-wise on par with e.g. cloud-based apps. But, at the same time, it is similar to other, competing ecosystems, since they all support it. Sencha recently demoed HTML5-based Facebook imitation on iPhone to show that performance-wise there was actually no difference between native app and HTML5. Don’t build your hopes on a dedicated ecosystem for you which is based on HTML5; that’s not going to happen.
I have been studying for some time the concept of a Design Ecosystem, which is related to operating system / software ecosystem in terms of how things are built for them. While Apple/iOS still leads the game, Android’s power comes from a vast mass, and they can more quickly integrate design innovations from external applications to core design style. Apple does this too (e.g. rubber band effect in lists), but their update cycle on design and HMI guidelines for iOS is much slower. Thus, Android may catch up in 2014 with iOS in terms of Design Ecosystem competition. I think Apple knows this and Jonny Ive’s nomination to not only ID, but UI/UX chief relates to this.
How to differentiate an HTML5 app? Differentiation can be positive or negative; performance being the usual metric. Best opportunities are in cloud-based, SNS integrator applications that have the massive data advantage, and can leverage “smartness” being outside. Think applications like “Path”, or those News or internet compilation/parser apps, which are in dire demand to be updated to next level (Zite, Flipboard). Also cloud-based productivity apps aimed at consumers are good focus areas (Zoho, photo sharing apps). But I would not try to aim for differentiation as such, because that is old thinking for competitiveness and, as said, other platforms can have the same easily. HTML5 apps that can have push notification support from the operating system are also something interesting.
There are always significant UI innovations such as Ixonos SuperApp, which can be based on HTML5 only. We have really one the most beautiful HTML5 apps around made for KONE, how to design an elevator. That kind of “Catalogue App” would be good to be HTML5 based.
The device manufacturers are not the ones who create the ecosystem, they enable it for external companies. This might be hard to understand and accept. BUT what they can do is to build a smart way to focus on certain kind of apps and do those really well, formulating winning style guides and accumulate an easy global distribution channel (for example, based on Ixonos Experience Store). Then have the analytics data constantly monitored, and basically pay first apps to be done specifically for their specs in HTML5, and let somebody else, not the manufacturer, to facilitate this. From strategy to action.
Sami Paihonen is Vice President of User Experience Design at Ixonos.