Why the pundits are wrong about Google I/O

Google’s personality can be hard to peg. Which might be why pundits and industry analysts are keen to try to affix a label of “the new Microsoft” or “the new Apple”, or other such nonsense to the strange machinations going on at Google.

Yes, Google sprawls like Microsoft. Yes, they are driven by a nearly-religious zeal like Apple.

But this week, they came into their own, and it’s very exciting stuff.

Everyone seems focused on the thought that Google stepped out at I/O as an alternate-universe Apple, with charts of Google’s anti-Apple stable of products. TV! FroYo Android phones! Mobile Ads! Music stores!

But developers saw a different story unfold at I/O.

Even though Steve Jobs is convinced that Google is trying to steal Apple’s lunch money, they’re not even at the same playground. While Apple is busy making technology desirable and friendly, Google is about the work of making that technology smarter.

Google’s I/O conference highlights

Here are a few examples of company-defining developments at Google IO:

Chrome App Store

They’re serious about the browser as OS.

What it means: Google sees operating systems as a dying art. To them, the OS is just a means of connecting together hardware so you can access cloud-based services. Unlike Apple or Microsoft, Google’s betting big on the shrinking footprint of OSX, Windows, and even Android.

If Google gets its way, the OS is going to be as obscure and irrelevant to you in the future as the underlying operating system of your DVR is now.

Google Storage for Developers

Pretty much anyone I’ve met using Amazon’s S3 storage has also used EC2 computing. And it’s a huge pain in the ass. So, although Google Storage for Developers is not too different from S3, its ties to Google’s App Engine are going to change things. Getting an application up and running used to mean setting up a machine with an operating system appropriate for your dev environment, installing database software, getting version control, blah, blah, blah…

Often, this involved hours (or tens of hours) of work before it was feasible to even start coding. Google App Engine is the opposite: Just submit your Java app (for me, JRuby), or Python (hello, Django) and boom, live application. And scalable, unlimited pay-as-you-go storage means that Google can now play the role of IT operations. This concept is not new (Heroku provides a friendly layer between Ruby and Amazon S3), but it’s going to become the basis for thousands and thousands of new applications, since any developer can now afford to turn an idea into reality.

Google Web Font API

Ask any CSS dev/designer about web font standards, and you’ll get a :rolleyes icon (given that you ask over chat, of course). Between variations in user-installed fonts and licensing issues in others, you wind up with about 4 or 5 fonts that are truly safe to use on the Web. That’s insane!

The Web Font API is a no-brainer, but only Google has got the cash and chutzpah to put it out there. Designers get a wider array of safe web fonts without a ton of work. Font designers can get exposure if Google decides to add their work to the library. Users can (potentially) get better-looking websites. This is so desperately needed and immediately useful that my former employer went from “never heard of it” to “essential to our site” in one day.

Prediction API

People are calling this “Skynet v.0.1”, and there may be good reason: it’s basically API access to learning computers. (Aside: this makes me think of Arnold Schwarzenegger saying “learning computer” and “neural-net processor”.)

It’s so out there that I still don’t know how it works: but as I understand it, it’s this: You feed it “Red, yellow, blue, red, yellow, blue, red, yellow”, and it’ll come back with “blue”. Or “one, two, three,” and it’ll come back with “four”.

This is scary stuff to me because it seems to represent the first step toward domination by our Cylon overlords, but there are lots of immediate applications. Historically, recommendation engines have been the exclusive territory of deep-pocketed e-commerce sites, but now any developer can have access to the thinking of the human-computer hybrids that are suspended in goo at Google headquarters.

Latitude API

There are other location-based services. And there are APIs to those services. But the Latitude API is so powerful and dead-simple to use that even I can do cool stuff with it, and I’m a neophyte. It’s a REST API that lets me feed in (or ask about) a person’s location, and get back everything from direction to altitude.

Location-aware was already a big deal (Apple, for their part, is doing great location-based API work for developers on the iPhone), but Latitude makes it universal.

What does all this mean for the future?

No one knows yet, since Google has a habit of throwing things to the wall to see what sticks. But you can look at the services above and mash them up with other apps: Using my location info from Latitude & Google Maps, an app checks Yelp for local listings, then references my Foursquare history and runs the results through the Prediction API to see which restaurant I’m most likely to want to eat at, then locates my friends in the area, and uses Twilio to auto-text them to ask them to lunch. If you had the desire and the chops, you could host that on Google App Engine for next to nothing.

With the depth and complexity of what Google churns out from day to day, I bet they have a tough time not looking down their noses at Apple’s work as child’s play. Oh, you have facial recognition in your photos app now? That’s adorable.

And before thinking, “of course Google can do all this stuff, they’re humongous”, think of the kind of rabid passion and laser focus that is required to pull off ideas of this magnitude, no matter the size of the company. Microsoft can’t stop screwing up its own best ideas (Courier, we hardly knew ye) due to infighting among its teeming middle-management ranks.

The biggest mistake Apple could make now would be to try to hunt on Google’s own turf (i.e. heading into the cloud). Google was “born in the briar patch”, so to speak. They and Apple could accomplish so much by cooperating instead of kicking dirt in each others’ eyes, but that’s a topic for another article.

Here’s how I would describe the 3 companies:

Apple is a pyramid, like the Luxor in Vegas, with a large base of people who taper to support one beam of light, one man’s vision. And that’s great (until a successor is needed).

Google is a series of pillars. Their biggest one supports most of the weight, but they are constantly building other pillars. They’re all building in the same direction and with the same purpose, but they’re more loosely federated.

Microsoft is just a mess. That’s all. It’s clear they don’t understand why they were ever successful in the first place, and unlike most who don’t understand their history, they’re doomed to not repeat it.

Incidentally… did you see Google’s “Pac-Man 30th Anniversary” homepage recently? A playable, fully-HTML-graphics, pixel-perfect version of Pac-Man is an astounding technical feat (certainly beyond Namco’s own capability), not to mention a lot of fun. It’s meticulously programmed: Pinky’s trickier, more aggressive personality is intact, and ol’ Pac is slower while chomping on dots than when the lane is clear. Google knew that this icon that shaped so many of us as young geeks deserved no less a tribute. That may be more telling of their corporate personality than anything at I/O.