HTML5 vs. Native Apps redux

A year ago, I wrote a piece on the tension between HTML5 and Native Apps, especially as it was playing out on mobile devices. The original is here. I found it interesting how the world flocked to the Web to get off of native apps, yet in mobile had flocked to native apps.

At least partially, I think that people were actually flocking to the idea of the cloud, rather than the no-distribution (or "No Software" as Benioff of salesforce.com would call it), which appealed primarily to IT departments. Nonetheless, it is ironic that the creation of truly portable devices - iPhones, iPads, Androids - meant considering being really offline on a regular basis, as opposed to mostly connected on a desktop/laptop. html5, with its offline capabilities, is meant (in part) to resolve that tension by providing the ease of development and distribution of a Web app, with the offline capabilities of a native app.

Pascal-Emannuel Gobry, who has quoted my writings before, wrote an extensive interview piece that largely agrees with my assertion: html5 apps will eventually dominate and then replace native apps. He asserts that it will take longer, and that it will have different economics and dynamics than the closed-wall garden of current app stores.

I would find it very interesting to apply the models of Clayton Christensen @claychristensen to the html5 vs native app discussion, especially in light of Gobry's description of native apps as very highly integrated (Christensen's terms) vs open and modular (again, Christensen's). Christensen's model would imply that in the early stages, the integrated, controlled native app model is likely to dominate, but over time the modular html5 model will take over, eventually disrupting Apple's and Google's position at least in the app store.

An interesting possible continuation is that without locking to native apps, both iOS and Android become just platforms. Much as I like my iPhone and iPad, will html5 not only reduce their clout in the app store distribution model, but possibly lead to openings to disrupt the very platform itself?