Google I/O 2012, the company's annual developer conference, offered a glimpse of the company's future and a reminder of the company's problems. None of Google's peers, allies, or competitors--Amazon, Apple, Microsoft, Mozilla, or Samsung--have a vision for technology that is quite as ambitious or socially transformative. And neither do they strive to realize their respective visions in such an incautious way.

Google's demonstration of Project Glass, its augmented reality glasses, exemplifies its approach. In six weeks, the company conceived, coordinated, and executed an impressive stunt twice, flawlessly--skydivers wearing Project Glass prototypes jumped from an airship over San Francisco onto the roof of the Moscone West convention center while streaming video from their glasses to the conference.

That's a lot of risk to call attention to a product that won't even be available in prototype form until next year, is of unproven value, and is sure to provoke privacy concerns. But then what would you expect from the company that pushed ahead with Street View despite the privacy controversies and scanned libraries full of books despite the copyright problems? Google doesn't play it safe.

[ Read Google I/O Day 2: Chrome Hits iOS, IaaS Play. ]

Project Glass could redirect the mobile computing revolution, or it could turn out to be Google's Ginger, a.k.a. the Segway Human Transporter, impressive engineering overshadowed by absurd hype. Ginger, Time suggested a decade ago, might be bigger than the Internet. Don't expect Project Glass to be that big.

At least Google is gambling. It made a credible foray into the hardware business with its Nexus 7 tablet and an iffier attempt to enter the consumer electronics market with its Nexus Q streaming media device. One of these will probably flop--the one priced at $299. But having acquired Motorola Mobility, Google can't back out of the hardware business now. Nor should it: With everyone emulating Apple's model--wedding hardware and software--Google faces a landscape of diminishing opportunities if it doesn't do the same.

As if to prove that point, Google launched Chrome for iOS, free of the primary things that make Chrome meaningful. Google says Chrome's core principles are speed and security, yet Chrome for iOS comes with neither.

Google itself admits that Chrome for iOS might run more slowly than Apple's mobile Safari browser. "Rendering and the Javascript engine are provided by iOS through UIWebView," Google explains on its website. "Because Chrome [for iOS] does not have access to Safari's Nitro engine, Chrome might have slower Javascript performance." The need to accept the restrictions in Apple's UIWebVIew API also affects the privacy afforded by Incognito mode in Chrome for iOS. Google uses an asterisk to refer to Chrome's browser privacy option, calling it Incognito* mode, to call attention to the API limitations that prevent it from handling HTML5 local storage as securely as the desktop version of Chrome.

Apple enforces its API restriction as a matter of security, a rule that, coincidentally, ensures that Safari's competition can't really compete. Being handicapped on another company's hardware explains why Google might want to get into the hardware business, whether it's ready or not.

Yet, Google appears to be learning from Apple about the value of an iron hand. Its Android Platform Developer Kit represents an attempt to get its hardware partners and carriers to march with more coordination. Of course, we've been here before: At Google I/O 2011, Google launched the Android Update Alliance to combat Android fragmentation. It didn't work. Perhaps the sequel will find an audience.

Android 4.1, known as Jelly Bean, offers further proof of Google's penchant for recklessness. One of Jelly Bean's primary features is Project Butter, a UI fix. A smooth, responsive UI is the sort of thing Apple would have made a priority. Google, having iterated as rapidly as possible to catch up with Apple, is now getting back to basics.

But Jelly Bean isn't just sanding Android's rough edges. Smart app updating, improved notifications, and Android Beam enhancements show innovation in Android is alive and well. Perhaps most impressive is Google Now, which represents a bold attempt to re-imagine search on a mobile device. It's search combined with just-in-time delivery: Google Now will do things like alert you to traffic conditions before you leave for work. If you've grown jaded after years of hearing about the coming of intelligent agents, pay attention to Google Now.

Google tried to encourage people to pay attention to Google+, which can now help organize events. But Google's "social spine" isn't nearly as interesting as its commitment to cloud computing.

Google's decision to enter the infrastructure-as-a-service business with Google Compute Engine might seem inevitable in retrospect, but it will still shake things up. Massive computational power on demand--that's what change is made of.

As Google continues its mission to organize the world's information, it's reorganizing the social, technical, and commercial landscape in the process.

New apps promise to inject social features across entire workflows, raising new problems for IT. In the new, all-digital Social Networking issue of InformationWeek, find out how companies are making social networking part of the way their employees work. Also in this issue: How to better manage your video data. (Free with registration.)