Google found a great way to use the data from the people who uploaded songs to Google Play Music. Google Music Timeline "shows genres of music waxing and waning, based on how many Google Play Music users have an artist or album in their music library, and other data (such as album release dates). Each stripe on the graph represents a genre; the thickness of the stripe tells you roughly the popularity of music released in a given year in that genre."



The timeline starts in 1950, data is is normalized by the total number of albums from a certain year and there's no classical music.

Mouse over a genre to find popular albums, search for an album or artist.

{ via John Mueller }

If you like the full-screen Google Trends visualization that shows the latest hot searches, there's a cool project that uses the same animations to show Google Analytics searches in real time. It's made possible by the recently launched Google Analytics Real Time Reporting API and it shows your site's internal searches, as well as some of the Google searches that lead to your site.


"You can setup your own site visualization in only two steps: first authorize access to your Analytics data, select your site, and set the query parameter for your site searches, usually the letter q. Click save, and see your visitor's searches appear live as a beautiful visualization," explains Yvo Schaap, who developed the service.

Here's a video that shows how to use this addictive mashup:


According to the site, Live Site Search Visualisation is "a dashboard visualizing live search activity on your site using your existing Google Analytics tracking." It also shows Google search referrals. Just like in Google Trends, you can click the top-left corner to select the number of keywords that are displayed at a time.


Just in case you're worrying about this, the site does not "save any data coming from your Google Analytics account on its servers. All data requests are made client-side only over https."

{ Thanks, Yvo. }

This video from Google lets you visualize the evolution of Android activations from October 2008 to January 2011. HTC Dream/G1, the first Android device, was launched in the US and the UK in October 2008 and it became available in other countries in 2009, but it wasn't until the launch of Motorola Droid in November 2009 that Android became popular.


Android Central says that this is "a staggering reminder of just how far Android has come in a relatively short amount of time". Eric Schmidt has recently mentioned that the average number of daily Android activations is now 350,000, up from 300,000 in December, 200,000 in August and 60,000 in April 2010.

Google Suggest Venn Diagram Generator is a visualization tool for popular Google searches. Enter an incomplete query, replace one of the keywords with "X", add three options for "X" and the tool will find the overlapping suggestions.

For example, if you enter "How can I get my X to", where "X" is: "wife", "cat", "dog", you'll find overlapping suggestions like: "How can I get my {wife, cat, dog} to lose weight", "How can I get my {cat, dog} to gain weight".


Hacker News readers came up with a lot of funny examples:

* Why is {America, Canada, Europe} so ...
* My {friend, wife, husband} is addicted to ...
* Why are {Chinese, Korean, Japanese} so ...
* Why is {C, PHP, Java} so ...

You should keep in mind that Google's suggestions are determined algorithmically based on the popularity of the search terms, so you'll find many stereotypes.

What Do You Suggest? is a site that lets you visually explore how Google Suggest autocompletes a query. "What Do You Suggest takes a seed from you, then guides you on a journey through language and the collective lives of Google users."



Simon Elvery, who created the site, found a lot of interesting patterns:

There are some recurring themes which have emerged as I've been playing with the site during its creation. Among other things, it seems people do rather a lot of searching on the topics of music, religion and relationships.

Sometimes the results are sad, like the people searching for free ebooks on relationships for dummies, and sometimes hopeful like looking for love.

Health is also a recurring theme with the names of medicines, diseases and symptoms appearing over and over again.

{ via Information Aesthetics }

Google Browser Size is an experimental service that shows if a web page has interface elements that can't be viewed by a significant amount of people. "Google Browser Size is a visualization of browser window sizes for people who visit Google. For example, the 90% contour means that 90% of people visiting Google have their browser window open to at least this size or larger."

The service can be used for any web page, but the data is obtained from the visitors of google.com. As you can see from the screenshot, Google's top result can be viewed by more than 99% of the visitors if no ad is displayed above the results.


Google Browser Size is one of the many Google tools that help you optimize web sites:

* Google Website Optimizer - testing and optimization tool
* Google Analytics - web analytics
* Google Webmaster Tools - site performance, crawl errors, top search queries
* Page Speed - an open-source Firebug add-on that helps you evaluate the performance of a web page.
* Speed Tracer - a Chrome extension that helps you fix performance problems in your web applications.
* Closure Compiler - a tool for making JavaScript download and run faster.
* "Let's make the web faster" tutorials.

{ Thanks, Kevin. }

Google released a new Google Labs experiment related to image search: Google Image Swirl. The service "organizes image search results based on their visual and semantic similarities and presents them in an intuitive exploratory interface."

Google Image Swirl clusters the top image search results for more than 200,000 queries and it lets you explore the clusters and the relation between images.

"Once you find the group of images you're interested in, you can click on the thumbnail and a cluster of images will "swirl" into view. You can then further explore additional sub-groups within any cluster. Image Swirl expands on technologies developed for Similar Images and Picasa Face Recognition to discern how images should be grouped together and build hierarchies out of these groups. Each thumbnail on the initial results page represents an algorithmically-determined representative group of images with similar appearance and meaning. These aren't just the most relevant images — they are the most relevant groups of images," explains Google.

Try queries like: jaguar, flowers, van Gogh and keep in mind that this is an early demo, so not all the queries will return results.


Google has two other visualization tools: "wonder wheel" for exploring related searches and YouTube Warp, an interactive way to visualize related YouTube videos.

Gone are the days when search engines provided mostly textual information, a list of 10 blue links that invite you to find the answers on your own. Search engines are now focused on extracting useful information from web pages and surfacing data in interesting ways.

In June, Google launched Squared, a Labs project that generates lists of entities related to your query and finds values for the associated attributes. Search for [dog breeds] and you'll find a list of breeds, suggestive images, information about the average weight and the country of origin.

Bing, Microsoft's search engine, has just launched a similar feature: visual search. It's just an early demo and it requires Siverlight, but it looks more impressive than Google Squared. Probably because Microsoft built the interface around images and used the attributes to refine the results.

"Visual Search allows you to quickly scroll through the galleries or do a one-click refinement using the quick tabs on the left, which are specifically relevant to the type of results you are browsing through," mentions Bing's blog.


Bing's visual galleries are more polished than the results generated by Google Squared, but Bing only shows galleries for a small number of queries like "popular TV shows" or "world leaders" and the information is obtained from structured data sources.

It remains to be seen if Microsoft manages to extend the visual search engine and dynamically generate results for any query. For now, you can play with the demo galleries.

YouTube added the "wonder wheel" visualization tool for exploring related searches interactively. The feature has been launched last month for Google search and it's now available in a different context.

Type a general term in YouTube's search box, click on "Search" and then click on "Wonder wheel" to find some suggestions. As you click on the suggestions, YouTube shows the search results next to the wheel and it lets you go back to the initial search terms.

"The Wonder wheel visually presents connections between related searches and your search term as an interactive diagram. Click the different nodes in the diagram to see how searches can branch out," explains Google.


While the idea seems interesting, I didn't find the "wonder wheel" very useful. It only works for some searches and the suggestions aren't diverse enough.

Google Labs has a new service for visualizing data: Google Fusion Tables. Unlike Google Spreadsheets, the service is optimized for large data sets, so you can import tables of up to 100MB.

"The goal of Fusion Tables, as with other database systems, is to manage larger amounts of data than spreadsheets typically do. This size difference leads to a focus on a different set of functionalities. For example, Fusion Tables focuses more on bulk operations on the data (filtering, aggregation, merges). These operations are typically not necessary for smaller collections of data stored in spreadsheets. In contrast, spreadsheets preserve complete flexibility in managing data -- you can put any value you want in any cell and work carefully to format your spreadsheet to make it look nice," explains the FAQ.


You can import spreadsheets from your computer, select one from Google Spreadsheets or open a table from the gallery.


One of the samples, titled "Popular Baby Names -- 1880 to 2008", has 129,000 rows, but Fusion Tables displays 100 rows at a time. To manage a such a large amount of data, you can apply filters, show aggregates, create views that only include some of the columns.

There are many options to visualize the data: on a map or intensity map, using lines, bars, pies, scatter plots, motion charts or on a timeline. For example, you can visualize the popularity of the name "John" in the US, which has slowly declined since 1924.


To see which are the most popular male names in the US since 1990, I added two filters: year >= 1990,rank <= 3 and then I aggregated the data by male name. The bar chart shows that Michael is the most popular male name.


The service has rough edges and there aren't many options to perform computations: there's no support for SQL or GQL queries, the sorting options are limited and you can't add data using forms.

Even if it's just a pre-alpha version, Fusion Tables has built-in collaboration features: you can invite people as viewers, contributors (they are allowed to add columns, but not to edit the existing columns), collaborators or owners. Anyone who is invited can add comments to a cell or a column.

{ Thanks, Surendra. }

Interesting timing. On the same day when Stephen Wolfram shows a demonstration of his knowledge engine Wolfram | Alpha, Google launched a new OneBox that visualizes public data. If you search for [Florida Unemployment Rate], Google shows an answer ("9.5% of the labor force - Not seasonally adjusted - Mar 2009"), the source and a graph that illustrates how the unemployment rate changed over time.



For now, you can only search for data related to the US: for example, California population. When you click on the result, the graph lets you compare California's population with the population for other states.


"Public statistical data, such as unemployment rates or population numbers, doesn't need to be hard to find or, more importantly, hard to understand. Google is making it easier to find and use important public statistical data from governments and other sources. This data is included in Google.com search results and displayed as interactive graphs that you can customize and share with others. If you've ever done research involving large datasets, you know that it can take hours to comb through databases in order to find and analyze trends. Anyone who searches for U.S. unemployment rates or population numbers on Google.com will see relevant statistics and graphs included in their search results. You can even search by state or county (but not by city)," explains the help center.

Google News Timeline is the second new services released today in Google Labs. According to Google, it's "a web application that organizes information chronologically. Google News Timeline allows users to view news, scanned newspapers and magazines, blog posts, sports scores, and more on a zoomable, graphical timeline."

The search box is very powerful, but you first need to pick a category from the drop-down. Google lets you add different sources: blogs, newspapers and it shows structured information from Wikipedia and Freebase.

Here's a simple way to visualize the list of posts from two Google-related blogs:


Using data aggregated by Freebase, Google can show some famous paintings by Claude Monet:


You can create interesting timelines using this services and it would be nice to save them or to have a permalink for future reference. Hopefully, the service will also add the option to use custom data sources.

Google Cache Mapper is a Greasemonkey script that enhances Google's cached pages by adding a virtual scrollbar with all the matches of your keywords. Google highlights your search terms using different colors and this script shows the position of the matches. The virtual scrollbar is clickable and it shows snippets for each match of your search terms.


Google Chrome has a similar scrollbar that uses yellow markers to locate the matches for the "find in page" feature.

The user script requires Firefox and the Greasemonkey extension.

Related:
Browsing the web using Google Cache

Information Aesthetics points to an interesting visualization of the "informational" relation between cities. Two cities are "informationally" related if they are often mentioned together, so the visualization uses the number of Google results to approximate the distance:

Gdistance(w₁,w₂) = (#(w₁)+#(w₂)) / (#(w₁+" and "+w₂)+#(w₂+" and "+w₁)),
where #(w) is the number of Google search results for the query w enclosed in quotes.

This approximation could be improved by replacing "and" with "*", so that the words aren't necessarily separated by the conjunction "and". The Google distance is multiplied with the physical distance between cities to increase the connection between cities that are far away.

Among the cities that have a small "informational" distance: London and New York, Tokio and Sydney, London and Singapore City.

Another way you can use the number of Google results is to calculate the mindshare of a word or name within a domain. If you divide the number of search results for [nokia mobile phone] by the number of results for [mobile phone] you can find Nokia's Googleshare within the mobile space.

Google Spreadsheets, the most mature application from Google's online office suite, has suddenly become much better.

Now you can get email notifications when your collaborators make changes or edit some specific cells. You can also get notifications when someone submits new data using a form.


Google Spreadsheets autocompletes the value from a cell so it's much easier to enter repetitive data in a column.


If the existing features aren't enough for you, there's a new directory of gadgets that can be added to a spreadsheets and use existing data. You may remember the data visualization gadgets I found last month: they're part of this directory, which includes many other interesting gadgets. You can add interactive time series charts, Gantt charts, funnel charts, timelines, tables with filters and grouping, pivot tables, maps, search results and you can also create your own gadget that adds other missing features. As with any beginning, not all the gadgets work very well and the pivot table gadget created by Panorama doesn't seem to work at all.

To add a gadget, click on the "Insert" dropdown and select "Gadget". Each gadget can be embedded into web pages or added to iGoogle, which is extremely cool because the data is updated automatically.




Google Docs help center mentions a new visualization API connected with the new gadgets, but the documentation is not yet available. "The Gadgets-in-Docs for spreadsheets API should be used when you want to create user-facing features which are accessible from within the spreadsheet editor of Google Docs itself. This approach combines the Google Gadgets API with the Google Visualization API, to allow the developer to access data on the spreadsheet for use or presentation in practically any form they choose. Developers using this method should already be familiar with the development of Google Gadgets, and then only need to learn some basic additions provided by the Visualization API. Note that this approach currently only supports one-way interaction with the underlying spreadsheet (reading data), however, it has the advantage of enabling the publishing of gadgets created in a spreadsheet to other gadget-enabled sites, such as iGoogle."

If you find some interesting uses for the new gadgets, publish your spreadsheet and post a link in the comments.

{ via Blogoscoped Forum }

Update. A cool visualization gadget based on GapMinder World (you'll find in Google Docs as "motion chart"):

Last year, Google acquired a data visualization technology from Gapminder, a Swedish non-profit organization. Gapminder's technology, Trendalyzer, was used to visualize data from United Nations Common Database in an intuitive way. To quote from Gapminder's site, "this software unveils the beauty of statistical time series by converting boring numbers into enjoyable, animated and interactive graphics." For example, you could see the evolution of carbon dioxide emissions in a country over time or find correlations between CO2 emissions and the income per capita.

Google probably intends to use this technology to visualize custom data. One of the interfaces that will allow you create your own Trendalyzer visualizations is an iGoogle gadget which creates a Flash multi dimensional chart. The input is a file with the following structure: the first column is the item's name, the second is time, the third is for x-coordinates, the forth is for y-coordinates, and the fifth is the size of a ball. I tried to upload a file that has this structure, but the gadget didn't accept it. The gadget is probably a part of a new service that also hosts your data.


This is not the only data visualization gadget developed. Another gadget creates an interactive table that supports filtering and grouping, the simple table gadget lets you perform queries, while the heat map gadget "displays a map with color intensities that match given values". There are also gadgets for pie charts, column charts, bar charts, area charts, image lines, scatter charts, organizational charts, time series.


Hopefully we'll be able to actually use these gadgets and Google will create a service around visualizing data.

Update: The gadgets are a part of a new Google Spreadsheets feature.

YouTube has a very cool visualization that lets you discover related videos. Just go to a YouTube video, click on the full-screen button and then click on the small button that shows a network. A lot of video ballons will pop up and the configuration will change once you hover over a button. It's an interesting way to discover videos and it reminds me of visualization sites like Liveplasma and Musicovery.




Update: You can find the visualizations at www.youtube.com/warp_speed.

Besides the standard Google News page, there are other ways to explore the news automatically clustered and ranked by Google.

The image version of Google News shows pictures related to the most important news and lets you explore a gallery of images that illustrate a news. This is a great way to find the key elements from a news at a glance, without even reading the text.


Newsmap is a brilliant visualization for the headlines that reflects the importance of a news. Everything is displayed in a single page that is automatically updated. "Newsmap is an application that visually reflects the constantly changing landscape of the Google News news aggregator. A treemap visualization algorithm helps display the enormous amount of information gathered by the aggregator."



Google News Cloud displays the most important keywords from today's news in a tag cloud. When you hover over a keyword, other related keywords are highlighted.


Google Trends checks how frequently a word was mentioned in the article indexed by Google News. The results are displayed at the bottom of a chart in the "news reference volume" section.

Buzztracker shows the relation between news and locations. Every day you can see the top cities mentioned in the news on a map. There's also an archive that goes back to 2004. Baghdad, Washington, Gaza, New York seem to be the most frequently mentioned cities.

An interesting way to explore the evolution of a person, company, idea or event is to use the timeline view from Google News Archive. Even if most articles require paid subscriptions, the snippets provided by Google are really helpful.

Blogger launched a new feature that shows the latest uploaded photos in a never-ending slideshow. Blogger Play is addictive, like many other real-time visualization tools, and could be a pretty good screensaver. The slideshow leaves out vulgar pictures, so the screensaver would be safe-for-work. "Blogger Play is (...) a great snapshot of what people are thinking and posting about, right now" and it expands the stream of recently-updated blogs from Blogger's homepage.

To create a screensaver from a web page, you can use RunSaver for Windows and IdleWeb for Mac (both are free, but there are more polished commercial alternatives).

Blogger Play was developed two years ago, but it remained a cool toy for Google employees. "Shortly after Blogger launched photo uploading two years ago, one of our engineers whipped up a web page that would show us the pictures that were being uploaded in real time. The result was fun, often beautiful, but above all, compelling. We couldn't stop watching. Over the years we've kept this photo scroller as part of the Blogger offices, on a monitor or projector, as an interesting (distracting?) slideshow, and a reminder of the diversity and vivaciousness of Blogger blogs."

Technorati has recently launched a real-time visualization of the most recent posts from the blogosphere, but it only shows posts from popular blogs. Now that Google indexes web pages really fast, it would be interesting to search for something and see the new web pages added at the top of the search results as soon as they're indexed.

SearchCrystal is an interesting way to visualize results from different search engines. The results common to more search engines are placed closer to the center of circle and have special icons that indicate the search engines. You can use this visualization to compare search engines, to find the rankings of a site or just to discover less common web pages.

SearchCrystal also lets you view image results, videos, news and blog posts from the most popular specialized search engines, but the overlaps are less frequent here. Unlike most meta-search engines, SearchCrystal lets you view all the top results in a single page and you can also embed it as a widget in your blog. Unfortunately, the site is pretty slow, so I preferred to include a screenshot instead.