joi, 21 aprilie 2011

Official Google Blog

Official Google Blog


Oklahoma, where the wind comes sweepin’ down the plain

Posted: 21 Apr 2011 08:20 AM PDT

Rodgers and Hammerstein weren't kidding when they wrote what is now Oklahoma's official state song. The gusts on the plains are fierce, which makes the Sooner State a great place to harness clean, renewable wind energy. Our commitment to greening our energy supply is also strong, which is why we've just signed a power purchase agreement (PPA) for wind energy—our second in less than a year—in Oklahoma.

The purchase is similar in size and structure to the agreement we signed last July for wind energy in Iowa, but this time we will be applying the power to our Mayes County, Okla. data center, which will be fully operational later this year. We've agreed to purchase all of the energy from NextEra Energy Resources' Minco II wind facility in Oklahoma for the next 20 years, through Google Energy LLC, an entity that enables us to participate in the wholesale energy market. This 100.8 megawatt facility will be built as a direct result of our financial commitment and should be operational in late 2011.



We've made the commitment to be a carbon neutral company, and this purchase is part of our effort to minimize our impact on the environment. We've managed to reduce our energy consumption by over 50 percent by building highly energy-efficient facilities, but we know that efficiency alone isn't enough to eliminate our carbon footprint. We've been exploring ways, such as this PPA, to reduce emissions further by increasing the amount of renewable energy we use to power our operations; we purchase high-quality carbon offsets for any remaining emissions.

If you're interested in learning more about the whys and wherefores of our renewable energy purchases, we've just published a white paper (PDF) on the topic. Our hope is that by laying out our reasoning and methods we'll make it easier for others in the industry to explore similar arrangements.

These purchases represent long-term, meaningful actions to reduce our carbon footprint and power our operations with clean electricity. Our infrastructure team will continue to seek similar opportunities globally as Google's businesses continue to grow. As a company we hope that purchases like these, plus the additional $350 million we've invested in renewable energy projects, support the market and drive down the cost of clean energy. This will enable even more companies to invest in sustainable energy solutions.

More predictions in autocomplete

Posted: 20 Apr 2011 01:12 PM PDT

I work on a team that develops autocomplete—the feature that provides predicted searches while you type. When you combine autocomplete with Google Instant, you can really accelerate your searching. Because it is so important to your search experience, we've been looking for ways to provide predictions for even more queries. Today we're improving the predictive powers of autocomplete, helping you search for things even when no one else in the world has.

One of the main ways autocomplete works is by looking at the most popular searches on Google. For example, most people who type "w-e-a" are going to search for [weather], so Google can make that prediction. What's tricky is that a huge percentage of the queries we get have almost never been typed before, so this makes it difficult to provide predictions based on popularity. For example, very few people have searched for [florida state senate building], so until today, even when you typed nearly the whole search query, you wouldn't get a prediction.

Now what we're doing is making predictions based on only part of your search—specifically, the last word or words. While few people have searched for [florida state senate building], many more have searched for [state senate building]. By looking at just the last part of what you've typed into the box, in this case "state senate bui," we can generate a prediction for "building." You'll see a dropdown box below the end of your search with predictions for just that word.


As before, to search for the predicted query you simply click the prediction or arrow down and hit enter. The feature can be particularly helpful for long queries, since the query is likely to be more unusual. For example, if you're trying to figure out [how many stairs to climb the arc de tri]... now you'll actually get the prediction for "triomphe" (a good thing, too, because there's no way I'd spell that right). Or, if you're looking for an [online store with underwater gad], you can save that extra second while you're shopping for "gadgets."

We've been experimenting with this change for a couple weeks and it's currently rolling out to all users on google.com in English. As we continue to improve the feature and test additional languages and locales, I can predict with high probability (pun intended) that we'll be expanding globally.