I used to think voice input was going to be huge some day, but now I sort of hope not. Can you imagine sitting in a cafe with everyone talking to their smart phones? UGH! Sure the "background noise" tech for speech recognition may be improved to the point where that would work in a technical sense, but in a humanistic sense, it's not a world I want to live in.
Ultimately it's also not just about input devices either (though that's a big part of it). I guess maybe it's a feedback loop where the limited input controls, screen size, and processing power (compared to a desktop app) influence the app developer's programming and UI development to the point where consumption is the easiest and most well realized use case. The only way that's going to change is if these portable app-centric devices gain more CPU, more display size/resolution, and better input precision and speed. The latter is probably the most challenging, though there will be limitations on smart phone display size for the foreseeable future just as a matter of practicality (until they can beam it right into our brains

).
Tablets are a bit different, and that's where I'd expect to see some divergence from the typical consumption-oriented app development paradigm, if at all. Certainly that has shown to be somewhat the case already, but I wonder how much that will really grow. There seems to be a practical limit to just how productive most people will be able to be on tablets without significant changes to the input systems and/or form factor.
- Oshyan