How Google is becoming an extension of your mind
July 18, 2012
Google is becoming is an extension of your mind, an omnipresent digital assistant that figures out what you need and supplies it before you even realize you need it, suggests CNET writer Stephen Shankland.
Think of Google diagnosing your daughter’s illness early based on where she’s been, how alert she is, and her skin’s temperature, then driving your car to school to bring her home while you’re at work. Or Google translating an incomprehensible emergency announcement while you’re riding a train in foreign country. Or Google steering your investment portfolio away from a Ponzi scheme.
Or your Google Glass automatically beaming audio and video to the police when you say a phrase that indicates you’re being mugged.
But it’s also, potentially, a profoundly creepy change. For a Google-augmented life, you must grant the Googlebot unprecedented privileges to monitor your personal information and behavior. And as Google works to integrate social data into its services, you’ll have to decide how much you’ll share with your contacts’ Google accounts — and the best way to ask them to share their data with your Google account.
Where your Google comfort zone ends
Shankland worries that handy new features will arrive in a steady stream of minor changes that are all but imperceptible until one day he wakes up and realizes that Google has access to everything that makes him who he is.
Google Now says it needs access to my calendar? Sounds useful. My Android phone needs to turn on my phone’s microphone so the Google Maps app can judge by ambient noise whether I’m indoors or outdoors? Well, that’ll help me get through the airport faster. My glasses need to identify the faces of people in my company so Google can deduce who gets consigned to the Google Voice answering machine and who gets through to my phone even at 3 a.m.? Well, I sure don’t want to have to set all that up manually.
Individually, those changes may be relatively benign, but collectively, they are a big deal.
Let’s face it, Google hardly has a spotless record here, he says. The company, which values pushing hard and apologizing later if necessary, launched Street View in 2007 but only started automatically blurring faces a year later; it let people remove their houses from Street View well after that.
It sorta kinda accidentally slurped up Wi-Fi data it shouldn’t have when gathering location data for Google Maps. It sidestepped pesky privacy settings with the Safari browser on iOS. It overreached with Google Buzz, a failed social networking project. It scanned millions of books without authors’ permission, presuming the activity was equivalent to indexing public Web pages. It collected location information about millions of laptops and mobile phones.
Google’s mission “to organize the world’s information and make it universally accessible and useful” is specific and direct where so many such statements are pompous and fatuous. But Google’s statement also is becoming obsolete, even if execs are slow to admit it.
“I’m not sure we have to rush out and change our mission statement,” co-founder Sergey Brin said in a conversation with a few reporters at the developer-oriented Google I/O conference in June. But today’s statement is too narrow, he indicated: “In general, I think our mission is to use technology to really change the world for the better.”
Some examples hr saw at Google I/O:
- Google Now is designed to draw on information such as calendar entries, Google Maps navigation, and search history to anticipate information that a person will find useful — something basic like a weather forecast or more sophisticated like a timetable to get to an appointment by foot, train, and bus.
- Better notifications in Android transform these alerts into graphically rich, miniature apps, letting people take actions such as approving a friend’s update on Google+ with a +1. Naturally, Google Now can send notifications, meaning that Google can nudge you when it’s time to leave for your doctor’s appointment.
- The tiny screen, camera, and speaker built into Project Glass’ computerized, networked glasses means electronic information can be woven directly into people’s interactions with the physical world. What sorts of information? Farther out, the glasses could snap a photo as you receive a business card, uploading the card image to your photo archive where its scanned contents would be added to your contact list along with a photo of the person who handed the card to you. They could recommend an appropriate wine from the 250 labels in front of you at the store. Or they could warn you when the person you’re talking to is getting angry.
Rewiring with Google+
Google+ is intended to help Google transform from a search engine to a service that lets people graft the Internet onto their social lives. Gundotra gave a specific example of how Google+ actions extend well beyond just the Google+ streams:
Endorsing a search hit — maybe you see an article on Parkinson’s disease and you want your family to know you’re endorsing that article or sharing that article — you can do that in a very, very easy way….
I don’t want the millions of followers I have to see an article on Parkinson’s disease because I injected it into my stream. But for my family, who may be dealing with that issue, that’s a great article, and I have the discretion of +1-ing it. That’s it. They see it when they do a search for “Parkinson’s disease.” Or I have the ability to push it into the stream, scope it to my family circle, and make a comment. We think this level of nuance, this level of dexterity and control, is exactly what users want.
Hardware time for Google
The company’s technology could have its finger on your pulse, literally. Its self-driving cars could give commuters another hour or two a day to be checking e-mail, watching video, and performing searches. Based on what you’re typing when preparing a presentation at work, Google could be finding speeches, news articles, and market data that’s relevant without you ever asking.
And if you’re wearing Project Glass glasses, it could be recording every word you say and sight you see. Wonder when your old college roommate’s child was born? Google indexed that conversation for you and knew who you were talking to. Concerned that you may have offended your date? Google can replay the video of his expression after you told the off-color joke. With Google’s glasses recording and processing what your senses capture, they could bring the perfect recall of computers to fallible human memory.
The more you share your personal life with Google, the better a job Google can do picking ads for you — and the more your privacy is being invaded. A company knowing you’re a 36-year-old male living in Miami is one thing, but how would you feel seeing ads based on a confidential conversation you had about your brother’s divorce within earshot of your Project Glass glasses? I’ll bet plenty of hackles will go up.
So Google should look at another option it can include in the revenue mix: selling services.
Google has for years charged $50 per person per year for use of Google Apps, its suite of online tools such as Gmail and Google Docs. More recently, it announced it’s charging for premium-tier use of its Google Drive online storage service, for example $5 a month for 100GB of capacity.
That could indicate Google might be looking beyond just advertising. Certainly the company is cautious about ads: there still aren’t any ads on the streams of Google+, where people’s comments could be used as keywords for ads the way Google does with Gmail messages.
Shifting the business gradually toward charging its users for services rather than selling ads might be a bitter pill, but Apple has shown that excellent products and marketing can easily persuade people to part with their money if they believe it’s worthwhile. The business might not grow as fast as giving stuff away for free, and Google certainly can’t be expected to move everything away from advertising, but it would sidestep a lot problems with privacy and trust.
Opting out of Google’s gaze
Google, evidently recognizing the risks to privacy missteps, tries to give people some control over their personal data.
Call me naive, but I think this is because company employees really do want to do the right thing. But even those with a more cynical interpretation of Google’s motives — that it’s granting the bare minimum of control to keep regulators and privacy activists at bay — must acknowledge that Google offers lots of opportunities to keep the Googlebot at arm’s lentgth.
One example: When Google unified its privacy settings so it could share user information among more of its services, it presented Google users with pop-up messages for weeks alerting them to the change.
Another example: Google’s Chrome browser and browser-based Chrome OS operating system can do all sorts of handy things if you grant it permission. But it does need your permission.
Many Google services, such as sending omnibox typing information to Google, are on by default. But Google often is more cautious. For example, the company won’t turn on its location-tracking Latitude service until people grant permission.
And Google Now also is off by default. When it’s first run, a full-screen page says, “Google Now is always working for you. It needs to: store your location periodically for traffic alerts, directions, and more [and] use your synced calendars and Google data for reminders and other suggestions.” You then have two choices: “Yes, I’m in,” and “No, maybe later.”
Behavioral ads present a new degree of intrusiveness. But if it irks you, Google gives you a advertising preference manager that lets you delete the categories Google has judged you’re interested in.
All these options are good — but I have a bigger concern.
Specifically, the system for granting Google its privileges is fundamentally broken.
The opt-in approach disables features by default until you specifically enable them. The opt-out alternative activates features but gives you the ability to shut them down.
This sort of user empowerment is a step in the right direction, but for a company of Google’s scale, neither approach works. The management problem is just too complex for ordinary mortals.
It’s like passwords today: as soon as we have too many to manage, we start getting lazy. When it comes to granting Google permission to rifle through your virtual desk drawers, few people have the patience to do more than accept the default suggestions or to re-evaluate their choices as terms and conditions change.
Look what happens with Android apps. Google created a system that explicitly requires a developer to tell a person what privileges an Android app needs — permission to access to the network, use the camera, store data, prevent the phone from sleeping, and monitor the phone status, for example. Most people might check a these the first few times they install an app, but for most people, it’s just like agreeing to terms of service — they click the “agree” button without a second thought.
In other words, if it’s a hassle, people won’t bother with it. And as the list of services multiply, users get lazier. It’s a problem that’ll get worse, not better.
Google’s prying eyes might not be too a big deal when it’s a matter of judging whether you want Google’s servers to know the Web addresses you’re heading to in Chrome. But what about when you’re talking about Google watching your medications, having your credit card number, and searching your recordings of your life’s soundtrack for relevant information?
As with passwords, better alternatives aren’t obvious. Shifting more toward paid services, though, could at least ensure Google is better motivated to please users rather than exploit their most personal information for the benefits of advertisers.
The more powerful Google’s services become, the more intrusive they become, too. Now is not a time to blithely grant Google whatever it wants. Perhaps one or two people will think about that in the coming years as “googling” means, well, living.