Yesterday I briefly posted about a Firefox experiment with integrating multi-touch into web pages. We’re not talking about the multi-touch in the browser here. We’re not talking about page-wide zooming or panning gestures. No, we’re talking about actually using multi-touch gestures to manipulate the elements in a page. Think in terms of multiple users being able to simultaneously drag around jpgs on a page. Or a user zooming in on one image using a pinch gesture, not not any of the other images on that page. Or think in terms of supporting painting using more than one finger at a time. The key here is that in each of these cases the user is manipulating the page objects directly, not the page as a whole as the iPhone or Windows 7 hosted versions of IE can do.
The question is should browsers be supporting this type of manipulation? Should HTML 5, for instance, support multi-touch object manipulation?
I can hear some arguing No. The main concern would be over whether we really understand the need yet. Maybe there isn’t one. So why build in a feature that we don’t understand yet. That’s not a bad argument. It makes a strong case for at least building in multi-touch as an add-on, which would at least force the browsers to support multi-mice or multi-object manipulation hooks, whether they actually provide standardized APIs is another story. This would at least give developers something to work with.
Some others might argue No, in part because they don’t believe multi-touch is compelling. I’d disagree a 100% here. The reason it is compelling is because it enables new interaction models. You only have to check out a Surface Computer to get your brain moving. But innovation doesn’t stop there. Think in terms of a whiteboard where more than one child could be working on it at a time. Or think in terms of an onscreen multi-touch keyboard all written in “HTML” that would make Google proud. Sure, for desktop displays and classic notebooks, there’s not much that multi-touch can offer, however, why do we want to continue to limit our innovation to a mouse/trackpad and keyboard–especially when it’s so obvious to see such dramatic innovations taking place in hardware–right now.
There’s another reason some people might argue against multi-touch at the browser page level. The might say that multi-touch in itself is fine, but it should instead be supported in something like Silverlight or Flash. To me, I’m not against these run-times offering multi-touch, however, the problem is that these run-times are not yet proving to be great matches for all classes of devices. It’s not just the iPhone that doesn’t support rich versions of Silverlight or Flash. There are very good engineering reasons why these tools are lagging in smartphones or e-Readers and the like. And if the market continues to head in the direction it currently appears to be–namely towards more and more smaller devices–there’s going to be a growing discontinuity between desktop browser experiences and device experiences should these run-times not find their way into more resource constrained hardware. Now, yes, multi-touch pages could have the same performance problems, but if multi-touch is supported as a standard with different conformance levels depending on the hardware, you could have experiences that web page designers could better design for. Theoretically you could do the same for Flash or Silverlight if only you always knew it was to be there. My guess is also that if you support multi-touch within the browser itself that you could do it more efficiently than let’s say within a plug-in’s runtime–though maybe it’s comparable. I’ll have to think about that a bit more. I’d still vote for trending towards a standardized approach though.
I’m sure there are a bunch of other engineers that might drag their collective feet and not want to implement multi-touch in web pages. All I’ve got to say to this is: Fine. Let someone else do it. Just don’t stand in the way. It’s theoretically not that difficult. And where there are issues that exist they are there because all designs up to this point have been so single mouse or single cursor biased. We really need to have them come to an end.