The Google I/O keynote this year started off in an almost pedestrian, passé note – instant messaging? Seriously? Even with a dollop of machine learning hype, the subject felt stale.
The only thing that held my attention was the Instant Apps demo, which besides bringing to mind XKCD #1367 posed an interesting technical conundrum – after all, how do you slice your
.apk bundle and make sure distinct activities can not just be dynamically loaded but actually work separately from the “whole” app? More to the point, how do you get that working on KitKat as well?
After watching the keynote I scanned the schedule for more interesting announcements, and found nothing much. Obviously, I missed the interesting bit: Official Android app support in Chromebooks, available this summer to developers and going mainstream this fall.
I agree with both Ars Technica and The Verge that this is big. Exactly how big, though, it’s too soon to tell. This because even though people are going ballistic about Chromebooks outselling Macs in the US, they tend to overlook the “in the US” bit.
Obviously, I don’t have hard figures on Chromebook sales outside the US, but if they’re happening anyplace else then they must be shipping inside unobtainum crates to the rest of the world, because it’s still almost impossible to get hold of one by normal means in my neck of the woods. I’m looking at you too, HP.
That said, the implementation looks solid (at least as far as The Verge’s short demo goes), and given my lasting desire to have something much like it, I’m definitely going to keep an eye on things. Their inclusion of corporate administration policies is a nice gesture and a hint of how serious they are about the whole thing, but on the other hand Google still hasn’t figured out enterprise sales, so… It’s a toss-up.
But there is one thing I keep thinking about: Android Instant Apps on a Chromebook could be an absolutely killer business platform – heck, it even has the potential of finally living up to the network computing hype and providing people with a truly usable interface that is loaded piecemeal off the network and is able to run on any device.
(The whole thing is actually kind of ironic considering that the Oracle vs Google trial is ongoing as I type this, but I digress.)
And it would span phones, tablets and laptops, starting from an installed base (and an app catalog) that despite not being high on quality (in any aspect), is almost certainly the largest in terms of quantity (either way).
Which gives it a pretty big market reach, and hence more appetizing to developers.
On the other hand, there are other takes on what a “universal” platform should be like.
Now, I haven’t been keeping track of UWP (and my current role at Microsoft doesn’t intersect with that space, so I have no stake in it either), but you have to wonder about where computing as a whole is going…
It’s going to be an interesting year. In fact, it’s already too interesting in many regards – but that, I will leave for another time.