something in the way

WDYL.com - When searching for a dirty word or phrase, instead of giving a safe-search error the website shows results for the word “kittens.”

/via Zachary Reese

(Cross-posted on the Inside Search Blog)

Today we’re announcing schema.org, a new initiative from Google, Bing and Yahoo! to create and support a common vocabulary for structured data markup on web pages. With schema.org, site owners and developers can learn about structured data and improve how their sites appear in major search engines. The site aims to be a one stop resource for webmasters looking to add markup to their pages.

Search engines have been working independently to support structured markup for a few years now. We introduced rich snippets to Google search in 2009 to help people find better summaries of reviews and people, and since that time we’ve expanded to new kinds of rich snippets, including recipes and events. We’ve been thrilled to see content creators across the web—from stubhub.com to allrecipes.com—add markup to their pages, and today we’re able to show rich snippets in search results more than 10 times as often as when we started two years ago.

We want to continue making the open web richer and more useful. We know that it takes time and effort for webmasters to add this markup to their pages, and adding markup is much harder if every search engine asks for data in a different way. That’s why we’ve come together with other search engines to support a common set of schemas, just as we came together to support a common standard for sitemaps in 2006. With schema.org, site owners can improve how their sites appear in search results not only on Google, but on Bing, Yahoo! and potentially other search engines as well in the future.

In addition to consolidating the schemas for the categories we already support, schema.org also introduces schemas for more than a hundred new categories, including movies, music, organizations, TV shows, products, places and more. As webmasters add this markup to their sites, search engines can develop richer search experiences. With webmaster feedback, we’ll be able to regularly publish new schemas for sites to use and, in turn, expand the list of queries with rich results. For webmasters who have already added microformats or RDFa currently supported by rich snippets, their sites will still appear with rich snippets on Google. You can learn more on our Webmaster Central Blog, Help Center and on schema.org.

While this collaborative initiative is new, we draw heavily from the decades of work in the database and knowledge representation communities, from projects such as Jim Gray’s SDSS Skyserver, Cyc and from ongoing efforts such as dbpedia.org and linked data. We feel privileged to build upon this great work.

We look forward to seeing structured markup continue to grow on the web, powering richer search results and new kinds of applications.

Posted by Ramanathan Guha, Google Fellow

(Cross-posted on the Inside Search Blog)

Today we’re announcing schema.org, a new initiative from Google, Bing and Yahoo! to create and support a common vocabulary for structured data markup on web pages. With schema.org, site owners and developers can learn about structured data and improve how their sites appear in major search engines. The site aims to be a one stop resource for webmasters looking to add markup to their pages.

Search engines have been working independently to support structured markup for a few years now. We introduced rich snippets to Google search in 2009 to help people find better summaries of reviews and people, and since that time we’ve expanded to new kinds of rich snippets, including recipes and events. We’ve been thrilled to see content creators across the web—from stubhub.com to allrecipes.com—add markup to their pages, and today we’re able to show rich snippets in search results more than 10 times as often as when we started two years ago.

We want to continue making the open web richer and more useful. We know that it takes time and effort for webmasters to add this markup to their pages, and adding markup is much harder if every search engine asks for data in a different way. That’s why we’ve come together with other search engines to support a common set of schemas, just as we came together to support a common standard for sitemaps in 2006. With schema.org, site owners can improve how their sites appear in search results not only on Google, but on Bing, Yahoo! and potentially other search engines as well in the future.

In addition to consolidating the schemas for the categories we already support, schema.org also introduces schemas for more than a hundred new categories, including movies, music, organizations, TV shows, products, places and more. As webmasters add this markup to their sites, search engines can develop richer search experiences. With webmaster feedback, we’ll be able to regularly publish new schemas for sites to use and, in turn, expand the list of queries with rich results. For webmasters who have already added microformats or RDFa currently supported by rich snippets, their sites will still appear with rich snippets on Google. You can learn more on our Webmaster Central Blog, Help Center and on schema.org.

While this collaborative initiative is new, we draw heavily from the decades of work in the database and knowledge representation communities, from projects such as Jim Gray’s SDSS Skyserver, Cyc and from ongoing efforts such as dbpedia.org and linked data. We feel privileged to build upon this great work.

We look forward to seeing structured markup continue to grow on the web, powering richer search results and new kinds of applications.

Posted by Ramanathan Guha, Google Fellow

Google recaps search trends for the year in Google Zeitgeist 2010, from the World Cup and the Olympics to the earthquake in Haiti and the BP oil spill. Above is relative search volumes around the world during the ash cloud in Iceland. You can browse the interactive map, or use the timeline to watch changes over significant events during the year.

A video (below) also accompanies the interactive, showing how the physical world and digital are melding ever so nicely.

• Watching Our Twitter World – twittervision Redux
• A Couple New Looks for Google News
• A world of tweets

Search as you type. It’s a simple and straightforward idea—people can get results as they type their queries. Imagining the future of search, the idea of being able to search for partial queries or provide some interactive feedback while searching has come up more than a few times. Along the way, we’ve even built quite a few demos (notably, Amit Patel in 1999 and Nikhil Bhatla in 2003). Our search-as-you-type demos were thought-provoking—fun, fast and interactive—but fundamentally flawed. Why? Because you don’t really want search-as-you-type (no one wants search results for [bike h] in the process of searching for [bike helmets]). You really want search-before-you-type—that is, you want results for the most likely search given what you have already typed.

As you can imagine, searching even before someone types isn’t easy—which is why we are so excited today to be unveiling Google Instant. Google Instant is search-before-you-type. Instant takes what you have typed already, predicts the most likely completion and streams results in real-time for those predictions—yielding a smarter and faster search that is interactive, predictive and powerful.

Here are a few of the core features in Google Instant:

• Dynamic Results - Google dynamically displays relevant search results as you type so you can quickly interact and click through to the web content you need.
• Predictions - One of the key technologies in Google Instant is that we predict the rest of your query (in light gray text) before you finish typing. See what you need? Stop typing, look down and find what you’re looking for.
• Scroll to search - Scroll through predictions and see results instantly for each as you arrow down.

Here’s a video that explains Google Instant in greater depth:

To bring Google Instant to life, we needed a host of new technologies including new caching systems, the ability to adaptively control the rate at which we show results pages and an optimization of page-rendering JavaScript to help web browsers keep up with the rest of the system. In the end, we needed to produce a system that was able to scale while searching as fast as people can type and think—all while maintaining the relevance and simplicity people expect from Google.

The user benefits of Google Instant are many—but the primary one is time saved. Our testing has shown that Google Instant saves the average searcher two to five seconds per search. That may not seem like a lot at first, but it adds up. With Google Instant, we estimate that we’ll save our users 11 hours with each passing second!

As part of our current rollout, Google Instant will become the core search experience on Google.com for Chrome, Firefox, Safari and IE 8. We’ll also be offering Google Instant to our users in France, Germany, Italy, Russia, Spain and the U.K. who are signed in and have Instant-capable browsers. Over the coming weeks and months, we’ll work to roll out Google Instant to all geographies and platforms.

We’re very excited about today’s announcement and hope that you are too. Give Google Instant a try and let us know what you think!

Posted by Marissa Mayer, VP Search Products & User Experience

When I was growing up, my dad had the best job I could possibly imagine: he was an arcade game and pinball technician. For me, that meant summer trips through Poland’s coastal cities with their seasonal arcade parlors; peeking inside cabinets to learn programming and engineering secrets; and—of course—free games!

One of my favorites was PAC-MAN, whose popularity transcended the geopolitical barriers of that time. During the heyday of space shooters, Tōru Iwatani’s creation stood out as one of the first video games aimed at a broader audience, with a cute story of pizza-shaped character gobbling dots in a maze, colorful (literally!) characters, friendly design, very little violence and everlasting fun.

Today, on PAC-MAN’s 30th birthday, you can rediscover some of your 8-bit memories—or meet PAC-MAN for the first time—through our first-ever playable Google doodle. To play the game, go to google.com during the next 48 hours (because it’s too cool to keep for just one day) and either press the “Insert Coin” button or just wait for a few seconds.

Google doodler Ryan Germick and I made sure to include PAC-MAN’s original game logic, graphics and sounds, bring back ghosts’ individual personalities, and even recreate original bugs from this 1980’s masterpiece. We also added a little easter egg: if you throw in another coin, Ms. PAC-MAN joins the party and you can play together with someone else (PAC-MAN is controlled with arrow keys or by clicking on the maze, Ms. PAC-MAN using the WASD keys).

PAC-MAN seems like a natural fit for the Google homepage. They’re both deceptively straightforward, carefully hiding their complexity under the hood. There’s a light-hearted, human touch to both of them. And we can only hope you find using Google at least a quarter as enjoyable as eating dots and chasing ghosts. You know, without actually needing any quarters.

Posted by Marcin Wichary, senior UX designer and developer

Using Google today, you may have noticed that something feels slightly different — the look and feel of our search results have changed! Today’s metamorphosis responds to the increasing richness of the web and the increasing power of search — revealing search tools on the left and updating the visual look and feel throughout. While we are constantly rolling out small changes and updates, today’s changes showcase the latest evolutions in our search technology, making it easier than ever to find exactly what you're looking for.

What’s new and what’s changed?
We’ve added contextually relevant, left-hand navigation to the page. This new side panel highlights the most relevant search tools and refinements for your query. Over the past three years, we've launched Universal Search, the Search Options panel and Google Squared, and it’s those three technologies that power the left-hand panel.

Universal Search helps you find the most relevant types of results for your search. The top section of the new left-hand panel builds on Universal Search by suggesting the most relevant genres of results for your query and letting you seamlessly switch to these different types of results. The “Everything” option remains our essential search experience with different types of results integrated into the main results, but now you can also easily switch to just the particular type of results you are looking for.

Our expandable Search Options panel launched last spring brought many rich slice-and-dice tools to search. The new left-hand navigation showcases these tools and enables you to get a different view of your results. Perhaps you’d like to see images from each of the results or just the newest information? These options are all on the left, and our technology will suggest the tools that are most relevant and helpful to your query.

Google Squared (available on Google Labs) helps you find and compare entities. Our “Something different” feature builds on the technology in Google Squared to find other entities that are related to your query, so you can easily explore not only the results for your current query but other related topics.

In addition to the left-hand side changes, we’ve updated our look and feel in terms of our color palette and our logo. These changes are slight, keeping our page minimalist and whimsical, but make our overall look more modern.

The new design refreshes and streamlines the look, feel and functionality of Google, making it easier to pinpoint what you’re looking for. It’s powerful, yet simple. Today’s changes are the latest in our continuing efforts to evolve and improve Google. We've been testing these changes with users over the past few months, and what we're launching today reflects the feedback we've received.. We want to ensure that the Google you use today is better than the one you used yesterday, and these latest changes open up many possibilities for future features and enhancements.

To hear more about our new design, check out this video:

Our new interface begins rolling out today globally across 37 languages. We are also launching the mobile version concurrently for English interfaces in the United States. Search on!

Posted by Marissa Mayer, VP Search Products & User Experience

Search is a natural starting point for discovering the world's information, and we strive to bring you the freshest, most comprehensive and relevant search results over an ever expanding universe of content on the multitude of devices you use to access it.

That's why today, at the Computer History Museum, we're excited to share a few new innovations in the areas of real-time, mobile and social search that we feel are important steps in the evolution of information access.

First, we're introducing new features that bring your search results to life with a dynamic stream of real-time content from across the web. Now, immediately after conducting a search, you can see live updates from people on popular sites like Twitter and FriendFeed, as well as headlines from news and blog posts published just seconds before. When they are relevant, we'll rank these latest results to show the freshest information right on the search results page.

Try searching for your favorite TV show, sporting event or the latest development on a recent government bill. Whether it's an eyewitness tweet, a breaking news story or a fresh blog post, you can find it on Google right after it's published on the web.

Here's how it looks:

Our real-time search enables you to discover breaking news the moment it's happening, even if it's not the popular news of the day, and even if you didn't know about it beforehand. For example, in the screen shot, the big story was about GM's stabilizing car sales, which shows under "News results." Nonetheless, thanks to our powerful real-time algorithms, the "Latest results" feature surfaces another important story breaking just seconds before: GM's CEO stepped down.

Click on "Latest results" or select "Latest" from the search options menu to view a full page of live tweets, blogs, news and other web content scrolling right on Google. You can also filter your results to see only "Updates" from micro-blogs like Twitter, FriendFeed, Jaiku and others. Latest results and the new search options are also designed for iPhone and Android devices when you need them on the go, be it a quick glance at changing information like ski conditions or opening night chatter about a new movie — right when you're in line to buy tickets.

And, as part of our launch of real-time on Google search, we've added "hot topics" to Google Trends to show the most common topics people are publishing to the web in real-time. With this improvement and a series of other interface enhancements, Google Trends is graduating from Labs.

Our real-time search features are based on more than a dozen new search technologies that enable us to monitor more than a billion documents and process hundreds of millions of real-time changes each day. Of course, none of this would be possible without the support of our new partners that we're announcing today: Facebook, MySpace, FriendFeed, Jaiku and Identi.ca — along with Twitter, which we announced a few weeks ago.

The new features will be rolling out in the next few days and will be available globally in English. You can try them out today by visiting Google Trends and clicking on a "hot topic," which in most cases will bring you to a search results page with the new real-time feature.

Here's a first look at our real-time search:

We have also made some new strides with mobile search. Today's sensor-rich smartphones are redefining what "query" means. Beyond text, you can now search by a number of new modes including voice, location and sight — all from a mobile device. So we've been working to improve technology that takes advantage of these capabilities.

Starting today, we're extending our voice search capabilities on Android devices to recognize Japanese. In addition, we're using the location of your mobile phone to launch some helpful features, like showing you "what's nearby." Finally, at our event this morning, we demonstrated Google Goggles, a visual search application that lets you search for objects using images rather than words, using your camera phone. For more information on these mobile innovations, check out the Google Mobile Blog.

As we've written before, search is still an unsolved problem and we're committed to making it faster and easier for people to access a greater diversity of information, delivered in real-time, from across the web. I'm tremendously excited about these significant new real-time search features.

Update on 12/8: Check out the video from our Search Event:

When we first launched public data on Google.com, we wanted to make statistics easier to find and to encourage debate based on facts rather than intuition. The day after we launched, a friend who worked at the World Bank called me, her voice filled with enthusiasm, "Did you know that the World Bank also just released an API for their data?" Excited, I checked it out, and found an amazing treasure trove of statistics for most economies in the world. After some hard work and analysis, today we're happy to announce that 17 World Development Indicators (list below*) are now conveniently available to you in Google search.

With today's update, you can quickly access more data with a broad range of queries. Search should be intuitive, so we've done the work to think through queries where public data will be most relevant to you. To see the new data, try queries like [gdp of indonesia], [life expectancy brazil], [rwanda's population growth], [energy use of iceland], [co2 emissions of iceland] and [gdp growth rate argentina]. For example, if you search for [internet users in the united states], you will see the following chart at the top of the results page:

Clicking on the result will bring you to an interactive chart where you can compare the United States with other regions around the world. We've also added a new feature to enable you to embed these charts in your own website or blog by clicking on the "Link" button in the upper right-hand corner of the chart page. You have the option to either embed the chart with static data, or you can also set the chart to update dynamically when new data becomes available. To give you a sense of what these charts look like, we've embedded the chart below comparing Internet users in the United States and South Korea:

We hope this new data and our new embedding feature will help facilitate quick and easy access to public statistics. There are still many other data sets and sources out there, and we're excited about the possibilities for the future. If you're a data publisher interested in making your data more easily discoverable in Google, please contact us.

* Complete list of World Bank indicators currently available: CO2 emissions per capita, Electricity consumption per capita, Energy use per capita, Exports as percentage of GDP, Fertility rate, GDP deflator change, GDP growth rate, GNI per capita in PPP dollars, Gross Domestic Product, Gross National Income in PPP dollars, Imports as percentage of GDP, Internet users as percentage of population, Life expectancy, Military expenditure as percentage of GDP, Mortality rate, under 5, Population, and Population growth rate.

Update at 11:26am: We've updated this post to include a better example.

Posted by Chung Wu, Software Engineer

Science fiction books and movies have long imagined that computers will someday be able to see and interpret the world. At Google, we think computer vision has tremendous potential benefits for consumers, which is why we're dedicated to research in this area. And today, a Google team is presenting a paper on landmark recognition (think: Statue of Liberty, Eiffel Tower) at the Computer Vision and Pattern Recognition (CVPR) conference in Miami, Florida. In the paper, we present a new technology that enables computers to quickly and efficiently identify images of more than 50,000 landmarks from all over the world with 80% accuracy.

To be clear up front, this is a research paper, not a new Google product, but we still think it's cool. For our demonstration, we begin with an unnamed, untagged picture of a landmark, enter its web address into the recognition engine, and poof — the computer identifies and names it: "Recognized Landmark: Acropolis, Athens, Greece." Thanks computer.

How did we do it? It wasn't easy. For starters, where do you find a good list of thousands of landmarks? Even if you have that list, where do you get the pictures to develop visual representations of the locations? And how do you pull that source material together in a coherent model that actually works, is fast, and can process an enormous corpus of data? Think about all the different photographs of the Golden Gate Bridge you've seen — the different perspectives, lighting conditions and image qualities. Recognizing a landmark can be difficult for a human, let alone a computer.

Our research builds on the vast number of images on the web, the ability to search those images, and advances in object recognition and clustering techniques. First, we generated a list of landmarks relying on two sources: 40 million GPS-tagged photos (from Picasa and Panoramio) and online tour guide webpages. Next, we found candidate images for each landmark using these sources and Google Image Search, which we then "pruned" using efficient image matching and unsupervised clustering techniques. Finally, we developed a highly efficient indexing system for fast image recognition. The following image provides a visual representation of the resulting clustered recognition model:

While we've gone a long way towards unlocking the information stored in text on the web, there's still much work to be done unlocking the information stored in pixels. This research demonstrates the feasibility of efficient computer vision techniques based on large, noisy datasets. We expect the insights we've gained will lay a useful foundation for future research in computer vision.

If you're interested to learn more about this research, check out the paper.

Posted by Jay Yagnik, Head of Computer Vision Research