May 30, 2008

New Google Favicon

Many people noticed a small change: Google has a new favicon. A favicon is a small image that is generally used to reflect the identity of a web site. You can find it in the address bar, in your browser's tabs, in the list of bookmarks or in some feed readers like Bloglines.

Google's favicon is hosted at google.com/favicon.ico and it's a 16x16 pixels image, a standard size for favicons. Google replaced the upper-case "G" in blue border, green and red borders with a lower-case purple "g" in a rounded corner rectangle.


The first time when I saw the new favicon at Google Image Search, I thought there was something wrong with my browser's cache or I typed an incorrect address. Google's new favicon is less cheerful and comforting, but it makes a lot of sense: the small g is a symbol for infinity (∞). A googol (10100) is just a poor approximation for the huge amount of information that needs to be indexed, organized and made useful by Google.


{ Thanks for the tip, Louis, John and Aleksandr. }

Update: If you want the old favicon back, try this Greasemonkey script (requires Greasemonkey for Firefox, Opera or a userscript plug-in for other browsers).

Update 2: Search Engine Land has an official position from Google. "We recognized there was a need for a Google icon that would better work across multiple applications including web, mobile and client applications. We felt the small 'g' had many of the characteristics that best represent our brand: it's simple, playful, and unique. We will be looking to improve and enhance this icon as we move forward."

Update 3: Google says this is not the final favicon. "We tried in total more than 300 permutations [some of them are displayed below]. It was much harder than we thought at first. We wanted something distinctive and noticeable, so we aimed toward transparency or semi-transparency, so the image would have a more distinctive noticeable shape than just a block. We wanted something that embraced the colorfulness of the logo, yet wouldn't date itself." If you have some suggestions for Google's favicon, send them here.


Update: The favicon was updated in January 2009.

May 29, 2008

Gmail Labs?

Google pre-announced a mysterious new feature for Gmail. "On Thursday, June 5th we're going to be launching a new Gmail feature that we like to think of as a next evolution of 20% time. It's a change in our development process and in the way users will be able to influence Gmail's design."

20% time is one of the many peculiar things about Google. "The 20 percent time is a well-known part of our philosophy here, enabling engineers to spend one day a week working on projects that aren't necessarily in our job descriptions. You can use the time to develop something new, or if you see something that's broken, you can use the time to fix it."

Google's engineers may work on many side-projects related to Gmail and this could be an opportunity for users to try them, send feedback or influence which features should be further developed and added to Gmail. It's not clear if that's the right interpretation for Google's intentionally-ambiguous text, but we'll find out next Thursday.

It might even match one of my predictions for this year: "Gmail will launch a Google Labs-like site with experimental features that could be added by those who are curious to see the next features before they are officially launched." Instead of releasing entirely new applications on Google Labs, it's more practical for Google to create mini-labs for each important service and expose experimental features that could be added or removed, like in Google Experimental Search.

Update (June 5th): TechCrunch covers Google's announcements. "At 6 pm PDT Gmail will launch Gmail labs under a Gmail tab - all users will have access to it. The first batch will have 13 features." Here are the features:

* a module for quick links (saved searches, important messages)
* superstars (additional star icons)
* pictures in Gmail chat
* fixed width font
* custom keyboard shortcuts
* mouse gestures
* random signature
* signature tweaks (places the signature before the quoted text)
* custom date format
* muzzle (hide the status of your contacts in Gmail chat)
* play Snake inside Gmail
* email addict (blocks the screen for 15 minutes)
* hide unread count

You'll be able to find them in a new tab from the Settings section. Probably the most useful features are: quick links, superstars and custom shortcuts. More about Gmail Labs in a future post.

Google App Engine, Open for Developers

Google announced that App Engine, the service that allows you to create web applications and host them on Google's servers, can be used without an invitation. Unfortunately, you need to verify your account by entering a code sent to your mobile phone.


The only programming language supported at the moment is Python. Google offers 500 MB of storage and enough bandwidth for 5 million pageviews per month for free, but in the future you'll be able to purchase more resources. Here are the expected prices, according to the App Engine Blog:

* $0.10 - $0.12 per CPU core-hour
* $0.15 - $0.18 per GB-month of storage
* $0.11 - $0.13 per GB outgoing bandwidth
* $0.09 - $0.11 per GB incoming bandwidth

For comparison, Amazon SimpleDB costs $0.14 per machine hour, $1.50 per GB-month of storage (8-10 times more than Google App Engine), $0.1 per GB incoming data transfer and $0.1-$0.17 per GB outgoing data transfer. Except for the storage cost, the pricing is somewhat similar to Amazon's offering, even if the services are quite different.

If you need ideas for an application, browse the gallery - you'll find a lot of simple apps for URL shortening, reading lists, unit conversion, time management or just for fun. You should also check Niall Kennedy's article about App Engine before reading the documentation.

"Google App Engine is a new and interesting solution for Python developers interested in adding features, not servers. Google spends hundreds of millions of dollars developing its custom infrastructure with 12-volt power supplies tapped into a hydro-electric dam next door and fat fiber pipes owned by local governments carrying requests and responses to their proper home. Google's physical infrastructure is vast array of highly optimized web machines, and we'll now be able to see how such infrastructure performs across more generic applications on App Engine," writes Niall.

{ Thank you, Tim. }

orkut Themes

When orkut doesn't borrow features from Facebook, it gets inspired by other Google services. This time, orkut added themes, but not all the users can see the new option. The themes should be familiar, since most of them were created for iGoogle. When you select a theme, only you and your friends can see it in your profile page.

The unofficial Orkut Plus blog has some Greasemonkey scripts that could be used to see the new themes if the feature is not yet enabled in your account. If you notice a link in the header that lets you "change themes", then you have the new feature.

Since iGoogle's themes are really appreciated by users, maybe it makes sense to add themes to all Google services and change them from a central location.


{ Thanks, V-render Studio. }

May 28, 2008

Google Earth Browser Plugin

Google Earth can now be used from a browser, without having to install the full application. Instead of the application, you need to install a plug-in that only works in Firefox and Internet Explorer 6/7 on Windows.

Google also launched an JavaScript API that lets you interact with the globe, draw markers, add layers or integrate with Google Maps. "The Google Earth Plug-in and its APIs let you embed the full power of Google Earth and its 3D rendering capabilities into your web pages." Google LatLong blog announced that each Google Maps mashup can take advantage of the new 3D view by adding a single line of code. "Our goal is to open up the entire core of Google Earth to developers in the hopes that you'll build the next great geo-based 3D application, and change how we view the world."

The samples look pretty promising, but I find it hard to understand why Google didn't use the API to create a better experience and bring more features from Google Earth (the search box, the list of overlays, the navigation controls) in a single interface.


Before downloading the plugin, please note that Google installs it in Firefox and Internet Explorer, along with a system service called "Google Update Service". The plug-in uses a lot of memory (around 100 MB just for loading the initial view and 300 MB for the Monster Milktruck demo) and, for each embedded object, you're running an instance of the Google Earth application.

Find the Font from a Logo

WhatTheFont is a site that lets you upload a logo or any other image that contains text and shows you a list of fonts that are likely to be used in that image. WhatTheFont supports some common image formats like GIF, JPEG, TIFF, BMP, but you can't upload images bigger than 360 x 275 pixels.

After uploading an image, you may need to enter the corresponding letters from the text, since the system can't always detect them accurately. In the example below, the site recognized the font used in Google's logo: Catull.


The site seems to be pretty old, but I couldn't find a similar online font recognition system. Identifont asks you some question about the font's appearance, so it's more difficult to use, while FontExpert only handles individual characters.

{ via Daring Fireball }

May 27, 2008

Google Street View Cars in Spain

The Spanish newspaper El Mundo (English translation) reports that Google's Street View cars have been spotted in many Spanish cities, including Madrid and Barcelona. Apparently, the local police wasn't very impressed and Google was fined for parking in a restricted area.


There's even a custom Google Map that includes many locations where Google's cars were spotted in Spain and Google Blogoscoped points to a video that shows a Street View car on the streets of Seville.


Google will add Street View imagery for Australia, Canada, France, Germany, Spain, Italy and other European countries in the coming months.

{ Thanks, wkh wkh. }

Google Hosts Popular JavaScript Libraries

If you've ever wanted to use a JavaScript library like script.aculo.us, Prototype, jQuery or Dojo, but you couldn't upload its files to a site or you found it difficult to manage all the different versions, Google hosts them for you.

"The AJAX Libraries API is a content distribution network and loading architecture for the most popular open source JavaScript libraries. By using the Google AJAX API Loader's google.load() method, your application has high speed, globally available access to a growing list of the most popular JavaScript open source libraries. (...) The AJAX Libraries API takes the pain out of developing mashups in JavaScript while using a collection of libraries. We take the pain out of hosting the libraries, correctly setting cache headers, staying up to date with the most recent bug fixes".

In addition to Google's APIs for search, maps, feeds and translations, you can now access AJAX libraries that add useful features like drag-and-drop, controls, animations or easier DOM manipulation. You can use Google as a fast proxy, but Google also caches the files related to iGoogle gadgets, the feeds that are served by Google Reader, and we should expect to see more content that can be accessed from Google faster and more reliably. In the future, Google could provide even more tools for a site: stats, monetization, search are useful, but what about using an unified Google API that lets you find facts, get spelling corrections, add ratings and forms, store data, add social features and code applications that run on Google's platform?

May 26, 2008

Google Shows Additional Information for Forums

After displaying the published date for search results, Google experiments with showing more information about forum threads. Below the title, there's a new line with the following format:

Discussion forum: number of posts - number of authors - date of the first post


The extended snippet categorizes the search result and indicates its potential usefulness: for example, a thread that has a single post or a single author is not very useful.

Other search results that include special metadata:

* videos show a thumbnail, the duration and sometimes the average rating (on a related note, Google no longer lets you play videos inline)


* Google Books results show the author, the published date, an important category and the number of pages

Google slowly categorizes the documents from its index and starts to show additional information relevant for each category. We can expect to see extended snippets for blogs, news sites, shopping sites, scholar papers and even a way to restrict the results to a certain category. An experiment from last year grouped the search results in different categories: references, reviews, stores etc.

For now, the extended snippet for discussion boards is not live for everyone, but you can change your Google cookie to be a part of the experiment.

Google Video's Player Detached from Search Results

The default view for Google Video, TV view, has been updated and you can now search for videos without interrupting the video that is already playing. You can go to the next page of results or reformulate your query, while still watching a video.

You'll also notice that Google Video can play videos inline not only from Google's sites, but from many other popular video sites (some notable exceptions: AOL Video, MySpace, Yahoo Video).


Google Video started by close-captioning TV shows, then moved to hosting videos, searching videos on the web and now the natural evolution is to come full circle back to close-captioning videos. Maybe in the future Google Video will bring back playlists and we'll be able to watch a selection of videos, like in YouTube Fast Search.

The Importance of Being Included in Google's Index

Last Friday, Google Analytics reported a surprisingly small number of visitors for this blog. According to Search Engine Roundtable, Google removed from the search results all the blogs hosted at blogspot.com and some of the .info domains for about 8 hours. It probably was an accident or a web spam filter that learns from my posts.


Many sites receives most of the traffic from search engines (especially from Google). Ryan Stewart's blog has been delisted from Google and he describes the consequences.
Yesterday was one of the scariest days I've had in a long time and it put in total perspective the power that Google wields. (...) I was working on a blog post and was searching for something I'd written before so that I could reference it. Instead of using my blog's search, I of course used Google's site feature, like I always do. Only this time, I got nothing. (...) I was confused, angry, and I felt completely helpless. Everyone uses Google. It was a big referrer to my blog and when I posted tips and tricks for Flash/Flex/AIR I did so knowing that someone could Google the problem and find my blog.

In his case, it wasn't Google's fault. "It turns out that Ryan's blog had been hacked, and Google does remove hacked sites from our index to protect our users," explains Matt Cutts. But whether a site is removed from Google's index accidentally or on purpose, it's interesting to quantify the importance of a popular search engine like Google for both users and publishers.

May 25, 2008

Microsoft Thinks Short Term

I was a little surprised to see that Microsoft decided to end the book scanning initiative and close its book search engine. But I was even more surprised to read the motivation:

"Given the evolution of the Web and our strategy, we believe the next generation of search is about the development of an underlying, sustainable business model for the search engine, consumer, and content partner. For example, this past Wednesday we announced our strategy to focus on verticals with high commercial intent, such as travel, and offer users cash back on their purchases from our advertisers."

In other words, the book search engine didn't make enough money and Microsoft decided it's better to focus on areas that are more profitable. Instead of improving their search engine with valuable content from books and offering better search results, Microsoft chose to make decisions based on the short-term profits.

This is also visible in the Cashback offering: "Microsoft Live Search cashback is The Search That Pays You Back. Find great deals on millions of products from hundreds of brand name stores that you know and trust. You will be able to earn cashback savings based on a percentage of the product price. Your savings will be paid to you via your choice of a deposit to your PayPal account, direct deposit to your bank account, or a check in the mail."

Microsoft hopes to attract more search users, but this will probably work in the short term. People will only go to Live Search for the rebates and they'll return to Google to find information on the web. If you pay your users and define your service as "The Search That Pays You Back", your brand's value diminishes because the search should pay off through its own merits.


Google has a long-term strategy for its search engine and for the other services. "Our goal is to develop services that significantly improve the lives of as many people as possible. In pursuing this goal, we may do things that we believe have a positive impact on the world, even if the near term financial returns are not obvious. For example, we make our services as widely available as we can by supporting over 90 languages and by providing most services for free," wrote Google's founders in the IPO letter.

To offer a compelling alternative to Google, you need to offer services that are more valuable to the user, even if there's no profit involved. Google thinks you first need to create a great service and then start to find the business model. Microsoft still has to work on building a great search engine, so discontinuing the book search engine is not a step in the right direction.

May 24, 2008

Filter and Sort Tables from Web Pages

TableTools is a Firefox extension that lets you sort, filter and export tables from any web page. After installing the extension and restarting Firefox, find a web page that includes tables (this search might help) and right-click on a table to find a lot of interesting features:

* sort a column, depending on the type of elements (text, numbers, dates)
* filter the rows of a table by selecting elements from a drop-down or by entering your search terms, like in the screenshot below. You can even enter regular expression to filter more complex patterns. After typing the query, press Enter or click outside the active input box.
* export the entire table as tab-delimited text, HTML code or just the filtered rows


This might be helpful if you want to insert some partial data from a table in a document, a site or in Google Spreadsheets, which only supports filtering using a limited gadget. For Google Spreadsheets, choose the option to copy the rows as tab-delimited text.

May 23, 2008

Google Anti-Malware Diagnostic Pages

ZDNet's security blog points to an update to Google's malware warnings. Like McAfee SiteAdvisor, now each web site has a special diagnostic page that lists answers to four questions:

1. What is the current listing status?
2. What happened when Google visited this site?
3. Has this site acted as an intermediary resulting in further distribution of malware?
4. Has this site hosted malware?

Here's, for example, the diagnostic page for google.com: http://www.google.com/safebrowsing/diagnostic?site=google.com, which lists some interesting facts.

"Of the 274621 pages we tested on the site over the past 90 days, 4 page(s) resulted in malicious software being downloaded and installed without user consent. The last time Google visited this site was on 05/22/2008, and the last time suspicious content was found on this site was on 03/13/2008. Malicious software includes 4 scripting exploit(s), 4 trojan(s). Successful infection resulted in an average of 10 new processes on the target machine. Malicious software is hosted on 4 domain(s), including 58.65.239.0, truemaybe.com, abc-powers.com. 5 domain(s) appear to be functioning as intermediaries for distributing malware to visitors of this site, including xtraff.biz, x-traffic.ws, smartvideochannel.com."

Despite all of these findings, google.com is not listed as suspicious, probably because the domain is whitelisted or the suspicious content is not very significant. It's likely that the domains listed above are from Google's search results, so that means the anti-malware system doesn't respect robots.txt.

TinEye - Upload an Image and Find It on the Web

I've always thought that an image search engine should accept as an input images and list identical or similar images from the web. This is useful if you have an image, but you don't remember what it depicts or if you want to find a higher-quality version of an image.

TinEye tries to do that and the best part is that it mostly succeeds. The new image search engine, powered by Idée's technology and currently in private beta, has an index of 487 million images (Google's index is at least 12 times bigger) and manages to find identical versions of an image or alterations. According to the FAQ, "TinEye frequently returns image results with colour adjustments, added or removed text, crops, and slight rotations. TinEye can also detect images that are part of a collage or have been blended with another image."

And the FAQ doesn't lie: I uploaded a screenshot of Flickr's homepage that included a Flickr image in the top-left corner. TinEye returned 6 results: 5 of them were different versions of the featured image (including the original image hosted by Flickr) and another result showed Flickr's homepage with a different featured image.


Then I uploaded an image from my computer that shows fingers in a book scanned by Google and TinEye pointed to me to a TechCrunch article that included that image:


TinEye doesn't do a good job at ranking images, as it orders the images "by relevance i.e. how well the result image matches your query image". It can't figure out the most-likely original source of an image, so TinEye's algorithms could be combined with a traditional image search engine like Google's in order to determine the authority of each image. TinEye also doesn't recognizes faces or objects in an image, so it just looks for similar images.

How does it work then? "TinEye uses sophisticated pattern recognition algorithms to find your image on the web without the use of metadata or watermarks. TinEye instantly analyzes your query image to create a compact digital signature or 'fingerprint' for it. TinEye searches for your image on the web by comparing its fingerprint to the fingerprint of every single other image in the TinEye search index."

The search engine is in private beta, but you can request invite or watch this screencast:


{ Thanks, Life Tester. }

May 22, 2008

Print Layout in Google Docs

As anticipated in March, Google Docs has a new way of displaying documents: fixed-width page view (or print layout in Microsoft Word). Documents are more readable and look closer to the way they appear when printed. The previous mode (plain view, also called web layout in Microsoft Word) is available in the new View menu.

The OpenOffice wiki explains the advantages and disadvantages of different word processing layouts: the web layout is useful for documents that are written for presentation, is suboptimal for editing because of the long lines and acts as a final preview. "The Print Layout implements WYSIWYG and tries to come as close as possible to the printed document. However, this layout is not particularly suited for numerous use cases. In many circumstances, more specific layouts fare definitely better." In addition to the web layout and the print layout, Microsoft Word 2007 includes a full-screen view for reading documents, an outline view for creating the document's structure and a draft layout that focuses on content, not on formatting.


{ Thanks, Alexander. }

Google Sites Available Without Google Apps

As promised when the service was launched as part of Google Apps, now you can use Google Sites without having a domain. "A few months ago we launched Google Sites exclusively as part of Google Apps for companies and organizations that wanted to use the service on their own domains. Now we've made it easy for anyone to set up a website to share all types of information -- team projects, company intranets, community groups, classrooms, clubs, family updates, you name it -- in one place, for a few people, a group or the world," says Andrew Zaeske on the Google Blog.


The sites are available at: http://sites.google.com/site/SITENAME and there doesn't seem to be a limit for the number of sites you can create. You can create as many web pages as you like, but each site has a storage quota of only 100 MB.

By default, each site is public, but you can make it private in the settings or when you create it. The same as in Google Docs, you're able to invite people as collaborators or viewers, but a site can have more than one owner.

Google Sites offers the same basic customization options like Blogger (themes, layout editor) and a rich text editor similar to the one from Google Docs, except that you can also embed a small number of whitelisted objects (Google Docs documents, spreadsheets, gadgets, YouTube videos).

In addition to web pages, you can also create simple blogs, lists, file cabinets and iGoogle-like dashboards. Each page can be arranged in a hierarchy and has a revision history that allows you to revert to earlier versions of a page. Google automatically creates a sitemap and can notify you when someone makes changes to the site or to an individual page.

For now, Google Sites looks pretty basic and doesn't include all the powerful features from JotSpot, the service acquired by Google and transformed into Google Sites. But that shouldn't be surprising, if you take into account that all Google services started small and gradually became more powerful and useful. I think the future of Google Sites is to combine Google's collaborative services so you can share more documents in a single page or to create blogs the same way you create calendars, to-do lists and photo galleries.


If you find Google Sites underwhelming, there are a lot of free or paid alternatives, including PBWiki, Wetpaint, Wikia and Wikispaces. There's also a nice video from Common Craft that explains "wikis in plain English":

May 21, 2008

Future Updates for Google Image Search

At the recent "Search Factory Tour" event (slides, YouTube video), Google announced that its image search engine has received a lot of attention from users lately and it intends to dramatically improve it. People no longer use Google Image Search only to find celebrity pictures or pretty images, they started to use it to compare products, to choose vacations or visualize unfamiliar situations.


To cope with the increasing number of images from the web and to provide better answers for the new use cases, Google promised that will start to add features that use complex image analysis.


One of the new features will allow you to find similar images, given a selected image. Since it's difficult to describe pictures using words, you will be able find a a group of images that illustrate the same situation.

After adding face detection as a restriction for image search, Google prepares to expand it and actually recognize faces. This will improve the quality of results for searches that include person names. Google doesn't intend to limit image recognition to people faces: finding objects in pictures is a difficult task, but it's a reliable way to filter irrelevant pictures.

Image search engines don't use the information from EXIF tags, that could offer a lot of interesting contextual details about location, date, image quality. Google will start to add information about geolocation from digital images.


The most controversial new feature tested by Google is the addition of display ads next to image results for commercial queries. Google's previous experiments with text ads weren't very successful, so adapting the ad format to the content could be a better idea. It depends on their usefulness and their prominence: the second mock-up displayed above puts too much emphasis on the image ads. For now, Ask.com Image Search is the only important search engine for images that displays ads, but they're text-only.

If we take into account that, in addition to all these enhancements, Google developed an improved algorithm for ranking images (VisualRank), we can expect an entirely new image search engine from Google in the near future.

Google News Layer for Google Earth

There's a new layer for Google Earth that shows Google News stories related to a location. At the recent "Factory Tour of Search" event, Google explained the difficulties of automatically identifying the locations of a news story. For example, just because a news article includes "Paris" doesn't mean that the article talks about France's capital. It could be about the Texas city or Paris Hilton, so the algorithm needs to disambiguate names, identify complete addresses and determine the importance of an alleged location in a text. Google News also uses its automatically-generated clusters to validate locations and their importance to a news story.

"The launch of Google News on Google Earth is a milestone in the evolution of the geobrowser. By spatially locating the Google News' constantly updating index of stories from more than 4,500 news sources, Google Earth now shows an ever-changing world of human activity as chronicled by reporters worldwide. Zoom into areas of personal interest and peruse headlines of national, regional and, when fully zoomed in, even the most local of interest," says Brandon Badger, Product Manager of Google's Geo team.

To enable the Google Earth layer, go to the Layers sidebar, expand "Gallery" and select "Google News" from the impressive list of overlays. Another news-related layer that has been recently added to Google Earth is for New York Times, but it's likely that the news are geo-coded manually.


If you want to read news related to a location in your browser, add a local section to the personalized Google News homepage. "Adding a Local News section allows you to track news stories from and about a particular city or region. While this function is currently only available in our English language editions, we hope to add more languages and regions in the near future," explains the Google News help center.

May 20, 2008

Videos Taken Down from YouTube


YouTomb is an interesting project by MIT Free Culture that collects YouTube videos taken down because of copyright infringements. "More specifically, YouTomb continually monitors the most popular videos on YouTube for copyright-related takedowns. Any information available in the metadata is retained, including who issued the complaint and how long the video was up before takedown. The goal of the project is to identify how YouTube recognizes potential copyright violations as well as to aggregate mistakes made by the algorithm."

Since YouTube operates under Digital Millennium Copyright Act, it's obliged to take down content if it receives a notification claiming infringement from a copyright holder. In some cases, videos are wrongly taken down because YouTube is in no position to judge the validity of a claim.

According to YouTomb's stats, the companies that have recently taken down the biggest number of popular YouTube videos are: TV TOKYO, Viacom, Warner Bros, World Wrestling and other media companies. "YouTomb is currently monitoring 157340 videos, and has identified 4389 videos taken down for alleged copyright violation and 13330 videos taken down for other reasons."

{ via friendfeed/imma }

FeedBurner FeedFlare for Blogger Comments

Now that FeedBurner is owned by Google, it makes sense to integrate with other Google services. One of the most requested features, a FeedFlare that shows the number of comments for each post from a Blogger blog, has been recently added. The FeedFlare links to Blogger's (ugly) comments page so you can easily add a comment if you read the post in a feed reader. Note that the comment feed has to be enabled in Blogger's settings page.

FeedBurner offers a FeedFlare API, but you need to do some server-side coding to create dynamic FeedFlares like the one that shows the comment count.



{ Thanks, Pat Hawks. }

May 19, 2008

Google China's Homepage in Mourning

The homepages of Google China, Yahoo China and other sites turned black today to commemorate the victims of last week's devastating earthquake. "Construction workers put down their tools, drivers stopped suddenly in the street, and rescuers briefly paused in their increasingly vain search for survivors amid the rubble of China's earthquake devastation. China stopped for three minutes to mourn the estimated 50,000 people killed by the earthquake exactly one week earlier," notes The Press Association.


Google China links to a custom search engine for sites that include information about missing persons. Kai-Fu Lee, President of Google China, told Google Blogoscoped: "[Engineering lead Harry Ke] and two other engineers came up with the idea of helping people search for lost relatives or friends. Given there are... tens of thousands still trapped, and some 200,000 wounded, and many more homeless. Also, cell phone and land lines are mostly not working. Transportation is difficult because the disaster area is on mountains and valleys. So, there are many people frantically looking for lost relatives and friends."

There's also a page for offering donations that mentions "Google will donate $2 million, including $1.7 million from Google.org, to help assist in relief and rebuilding efforts". A special Google Earth layer shows recent high-resolution imagery from the areas affected by the earthquake.

{ Thanks, Matthew Chan. }

Google Health Launches

After a year and a half since the first announcement, the much-anticipated Google Health has been released at Google Factory Tour of Search. "Patients need to be able to better coordinate and manage their own health information. We believe that patients should control and own their own health information, and should be able to do so easily," said Adam Bosworth in November 2006.


Here's what you can do in Google Health:
* create a health profile with information about your health conditions, medications, allergies
* import medical records from US hospitals that use Google's APIs to make the conversion possible. Unfortunately, the list of partners is almost empty.
* read medical resources, information about diseases


* find a doctor using Google Local Search
* use other health services that integrate with Google Health and can can import your data securely and use it for different purposes: calculate the heart attack risk, print your health history or share it with doctors. According to the FAQ, "Google Health is a PHR (Personal Health Record), but it is also a bit of a different model. We believe it's not enough to offer a place where you can store, manage, and share your health information. You need to act on your health information to better manage your health needs on a daily basis. This is why we provide a directory of online health services to you. You must elect to sign up with a service and decide what level of personal data you want to share in exchange for the customized services those companies offer."

Google Health wants to become the central place where you organize your health information and share it with people or services you trust. Since this information is very sensitive, Google takes a lot of precautions by using SSL connections and a separate privacy policy that clearly states: "You control who can access your personal health information. By default, you are the only user who can view and edit your information. If you choose to, you can share your information with others."

This is a big test of trust for Google and probably the most personal service ever offered by the company. While you can also enter your credit card information in Google Checkout, your location in iGoogle or Google Maps, personal information in orkut, Blogger and YouTube, Google Health is about your existence.

"Health information is very fragmented today, and we think we can help. Google believes the Internet can help users get access to their health information and help people make more empowered and informed health decisions. People already come to Google to search for health information, so we are a natural starting point," says Google. But there's a big difference between offering general health information and storing health records, so it will be interesting to see if Google manages to convince users that this shift is beneficial to them. Google could integrate more health information in the search results, the same way Microsoft shows information from HealthVault in Live Search and lets you collect it in your account. TechCrunch says that "Google promises never to advertise on Google Health".

(Small tidbits: Google Health is one of the very few Google applications created using Google Web Toolkit and its codename seems to be Weaver.)

May 16, 2008

Set Basic HTML as Default Gmail View


If you use Gmail and you'd like a simplified interface that doesn't use AJAX, loads fast and works well in most browsers, try the basic HTML view. Gmail links to this version at the bottom of the page, but you can also access it if you go to http://mail.google.com/mail/h/.

Until recently, Gmail didn't have the option to set the basic HTML version as the default interface and you had to bookmark its URL or click on the link from Gmail's footer. Now there's an option to always go to the HTML version, every time you load Gmail.

Please keep in mind that the simplified version lacks many useful features:
* integration with Google Docs and Google Calendar
* keyboard shortcuts
* integrated chat
* composing options: spell checker, rich formatting, address auto-complete, custom From address
* ads (useful feature?), related pages, tracking packages and addresses
* contact management
* web clips.

Google recommends to use the basic HTML view for slow Internet connections, although you may find it useful when you use exotic browsers or when the standard version doesn't work. If you change your mind and you want to go back to the AJAX version, click on "Standard View". Apparently, the standard view loads much faster thanks to some aggressive optimization.

May 15, 2008

Product Images in Google Shopping OneBox


Google started to display a thumbnail in the Product Search OneBox, as you can see by searching for [usb flash]. The image illustrates the top search result, but links to the list of results. As part of the Universal Search, this OneBox can be displayed at the top of the search results page or at the bottom of the page, depending on its relevance to the query.

Another change is that the OneBox groups identical tech products and shows a range of prices. For queries that include clustered listings, Google no longer shows a link that restricts the results to products that can be bought using Google Checkout.

Here's an old screenshot of the Froogle/Product Search OneBox:

Google Doctype, an Encyclopedia for Web Developers

Google released Doctype (HTML version), an encyclopedia of the open web. "The open web is the web built on open standards: HTML, JavaScript, CSS, and more. The open web is a beautiful soup of barely compatible clients and servers. It comprises billions of pages, millions of users, and thousands of browser-based applications."

Google Doctype is an encyclopedia that can be edited by anyone who has a Google account and wants to keep it up-to-date or add new articles. The encyclopedia contains articles about web security, DOM manipulation, CSS, HTML best practices, references for HTML, DOM, CSS, complete with browser compatibility information. There's also previously-unreleased code used internally by Google that is now documented and available for anyone to use.

You could call it Wikipedia for web developers or cross-browser MSDN, but Google Doctype is a clear sign that Google wants to foster a community of developers and encourage building web application using open standards.


Mark Pilgrim, the author of "Dive into Python" and now a Googler, explains more about Google Doctype in the following interview:

May 14, 2008

Google Spreadsheets Can Be Edited by Anyone

Google Spreadsheets added an option in the sharing dialog that allows anyone to view or edit the spreadsheet just by knowing the URL. Until now, you had to send an invitation URL that contained a secret code and the people you invited had to login using a Google account. If you click on the Share tab and enable "Anyone can edit this document WITHOUT LOGGING IN", your spreadsheet becomes a wiki that can be edited by anyone.


To experiment the new feature, I set up a spreadsheet that compares the features of two desktop office suites (MS Office, OpenOffice) and two online office suites (Google Docs, Zoho). The spreadsheet has three sheets for: word processing, spreadsheets and presentations.

In other spreadsheet-related news, you can now embed forms in a web site by just copying some code, there's a new option to duplicate questions and users can add their own answer to a multiple-choice question.

Update: You can view a snapshot of the spreadsheet after the first day of editing. The comparison has been extended with information about voice recorders, clay tablets, smoke signals, telepathy, Jedi mind control and more.

Update 2 (May 26): You can no longer edit the spreadsheet.

Use iGoogle and Google Reader in the Sidebar

A simple use for Google's iPhone interfaces is to add them as sidebars in Firefox or Opera. I mentioned last year some Google gadgets for Google Notebook, Google Talk, Google Docs, that could be displayed in a permanently-visible sidebar. Here are two interfaces optimized for iPhone that have permalinks:

iGoogle - all of the gadgets are displayed in a single column and you can switch between tabs at the bottom of the page.


Google Reader - a beautiful interface that lets you read posts inline, star them, share them and browse by tags or feeds. "This new version is designed to offer many of the same features as the desktop, while making it quick and easy to act on items," says Google Reader blog.


Normally, if you click on the two links for iGoogle and Google Reader in Firefox or Opera, you should be asked if you want to bookmark the page. To open it in the sidebar, go to the Bookmarks menu and click on the corresponding item or press F4 in Opera. If the links don't automatically create a sidebar:

* bookmark them and select "Show in panel" (for Opera)

* bookmark them, then go to the Bookmarks menu, right-click on the bookmarks, select "Properties" and enable "Load this bookmark in the sidebar" (for Firefox).

Google Shows Blurred Faces in Street View to Protect Privacy

Google prepares to launch Street View in Europe, Canada and Australia, where local laws regarding privacy in public places are less permissive than in the US. As promised, Google will blur faces for all the Street View imagery not just because of local laws, but also because the purpose of Street View is to show places, not people that happen to be there when Google's Street View cars go by.

To test the face detection algorithms, Google started with Manhattan, where Google replaced the Immersive Media imagery with newer and better panoramic images. "This effort has been a year in the making - working at Street View-scale is a tough challenge that required us to advance state-of-the-art automatic face detection, and we continue working hard to improve it as we roll it out for our existing and future imagery." The results are pretty good, although not all the faces are correctly detected and people can be identified even when their faces are blurred.


As Avi Bar-Zeev suggested, a better idea would be to remove people and cars from the images. "The main reason for removing people and cars is the same reason you'd want the base Google Earth imagery to lack clouds. These things tend to block your view. You can't really look behind them in a simple (essentially 2D) panoramic image. They only represent one snapshot in time vs. a broader/more virtualized essence of the place. They make it confusing to add dynamic versions of the same things on top of what's permanently baked into the imagery."

May 13, 2008

Google Maps Adds Real Estate Search

As a Google employee recently said, "Google Maps is evolving from a driving directions and business search tool, to a comprehensive representation of all the world's information, on a map." That's why Google Maps started to integrate different layers of information when you search for an address and it added a new "More" button to enable layers for photos and Wikipedia articles. Google Maps now includes in search results personalized maps, geolocated content from the web and mapped web pages.

There's a new option to search for real estate: click on "show search options" and select "real estate" from the drop-down. The search results don't seem to be powered by come from Google Base. Google shows structured information about houses and lets you refine the results by price, number of bedrooms and bathrooms. Even if there aren't too many advanced features, it's interesting to see that Google Maps wants to index all the information that could be displayed on a map.

Many real estate sites use Google Maps API and the first Google Maps mashup was HousingMaps, a site that displays Craigslist housing listings on a map. Last month, Trulia was one of the first sites that integrated Google Street View "to add efficiency to the real estate search experience and help home buyers discover more information about particular neighborhoods".


(The post was updated to reflect that the data is obtained from Google Base.)

May 12, 2008

Spreading OpenSocial Across the Web

Any web site can be a container for OpenSocial, any web site can add social features even if it's not a social network - that seems to be the idea behind Friend Connect, a new piece from Google's social puzzle. Friend Connect will allow the users of a site to add profiles, to import their friends from other social networks, to use social applications in the context of a site.

Paul Buchheit wrote last year that "there's no such thing as a social network". The social aspect of a site is just one of its many features. "Real products need more functionality in order to somehow deliver value to their users. It is this other functionality that defines the real purpose of a product, not the social network, which exists only to enable or enhance the core purpose."

Friend Connect is an enabler for making web sites more social, since the barrier to entry is really low. "First, many website owners want to add features that enable their visitors to do things with their friends, but the technology and resource hurdles have been too high. Second, people are tiring of needing to create new logins and profiles and recreate their friends lists wherever they go on the web." Google will use OpenID or Google Accounts for authentication, OAuth or APIs like Facebook Connect, MySpace Data Availability to find your friends from other social networks and OpenSocial gadgets to interact with your friends.

"Social is in the air. It's the blossoming of a lot of work by a lot of people. We don't move in lockstep and don't need to. We converge on interoperable technology. There's more than one way to connect a site to the social Web. With Friend Connect, we're confident it's a good step forward. I'm sure there will be more ways to do that than what Friend Connect does. We wanted to start with easiest and safest starting point," said Google's David Glazer in a conference call.

A preview of Friend Connect will be available later today at http://www.google.com/friendconnect (update: the page is live) and the service should be launched in a couple of months.



May 11, 2008

Tips for Linking to Google Presentations

If you publish a presentation at Google Docs, you'll receive a simple URL that can be used to view the presentation online. Unfortunately, if you go to that URL without being logged in to a Google Account, Google will ask you to log in:


The explanation is that Google Presentations offers some advanced features that require authentication: chatting with other people that view the presentation and joining a presentation that's already in progress. To view the presentation without logging in, click on the small link from the bottom of the page: "View published presentation in a new window".

If you want to link directly to the presentation and skip the authentication page, just add "&skipauth=true" to the URL provided by Google and replace "Presentation" with "Present". Here's an example.

URL provided by Google:
http://docs.google.com/Presentation?docid=ID

Modified URL:
http://docs.google.com/Present?docid=ID&skipauth=true
(replace ID with the presentation's ID)

Most people will also want to download the presentation, but Google doesn't offer this feature in the view-only interface. You can link to the PPT file by using this format:

http://docs.google.com/MiscCommands?command=saveasdoc&exportFormat=ppt&docid=ID

Yahoo Search's Differential Features

Yahoo's strategy to increase the search market share is to add features that can't be found at Google or somewhere else. The problem is that these features need to be distinctive and useful enough to attract the attention and make people switch to Yahoo or at least use it a secondary search engine.

The first innovative feature added by Yahoo was Search Assistant, an integrated pane that combined autocomplete and related searches. Search Assistant was heavily inspired by Ask.com's left sidebar, but it included a distinctive feature that made it less obtrusive: the pane is only displayed if you stop typing for a couple of seconds or when your typing slows.

Google also tests a query suggestion feature and places a list of related searches at the top of the page, but Yahoo's implementation is more interesting.


This week, Yahoo started to add SiteAdvisor's warnings next to search results. "Safety ratings from McAfee SiteAdvisor are based on automated safety tests of Web sites and are enhanced by feedback from volunteer reviewers". Yahoo only shows warnings next to sites that use browser exploits, offer malicious software or send spam. Google also shows warnings next to web pages that may install malicious software, but McAfee SiteAdvisor seems to offer a more comprehensive protection and more information about the potential threats (you can also install a plug-in for IE or Firefox that works with the most popular search engines or manually find the testing results for a site).


Probably the most impressive new feature in Yahoo Search and the only one that's not yet live is SearchMonkey (an unfortunate play on GreaseMonkey), a way for site owners to enrich the snippets with structured information. "Site owners will be able to provide all types of additional information about their site directly to Yahoo! Search. So instead of a simple title, abstract and URL, for the first time users will see rich results that incorporate the massive amount of data buried in websites -- ratings and reviews, images, deep links, and all kinds of other useful data -- directly on the Yahoo! Search results page."

Yahoo uses semantic web standards to retrieve structured information from web sites, but users are the ones who decide if they want richer search results from a site. Yahoo will support a small number of microformats (hCard, hCalendar, hReview, hAtom, XFN), "some of the vocabulary of Dublin Core, Creative Commons, FOAF, GeoRSS, and MediaRSS, as well as RDFa, eRDF, and the OpenSearch specification".

Google chose a different approach - plus boxes that show additional information automatically detected: addresses, stock symbols, products etc. Google also lets you add subscribed links to search results pages, but very few sites took advantage of this feature.


If Yahoo manages to promote these features and site owners build interesting applications for SearchMonkey, people might discover that Yahoo has a pretty good search engine and search is not synonymous with Google. Exploring different ways to present search results will lead to a better user experience and to an improvement for all search engines, since the best features are usually copied by all of them. Yahoo Search hopes to become a serious alternative to Google by having a distinctive voice, but the history of Ask.com or Opera shows that being innovative is not the only necessary ingredient for becoming popular.

May 10, 2008

Reorder iGoogle Tabs

iGoogle finally offers a way to reorder tabs: just go to google.com/ig/settings and use the arrows displayed next to each tab name in the Content section. This solution is not very elegant, compared to other services like Netvibes, where you can reorder the tabs using drag and drop.


If you're wondering how to access the settings page from iGoogle, click on the small arrow displayed next to the name of the current tab and select "Edit this tab". The settings page is also the place where you can select your location, choose to automatically open links in a new tab, change the name and the layout of a tab or backup the iGoogle page. Oh, and don't forget to click on the barely noticeable Save button every time you change the settings of a tab.

It's amazing to see how simple and intuitive Netvibes' interface can be: the tabs can be reordered using drag and drop, tab names can be changed using a simple click and switching between different tabs doesn't reload the page. iGoogle still has a lot to learn from Netvibes in terms of user friendliness.

May 9, 2008

Using Google's N-Gram Corpus

Two years ago, Google released a collection of n-grams from web pages and made it available on Linguistic Data Consortium's website. "We processed 1,024,908,267,229 words of running text and are publishing the counts for all 1,176,470,663 five-word sequences that appear at least 40 times. There are 13,588,391 unique words, after discarding words that appear less than 200 times." Here are some examples of 3-grams, followed by their frequencies:

ceramics collectables collectibles 55
ceramics collectables fine 130
ceramics collected by 52

While this huge corpora is useful to build linguistic models, there are other ways to use it. Chris Harrison created some visualizations for bigrams and trigrams that start with pronouns. "These visual comparisons allow us to see differences in how the two subjects are used - both where they are similar and diverge. For example, among the top 120 trigrams, 'He' and 'She' have many common second words. However, they differ on some interesting ones, for example, only 'he' connects to 'argues', while only 'she' connects to 'love'."


Chris DiBona from Google works on IsolWrite, a word processing program that will include a text prediction option. "I gotta get my greasy hands on an open version of our published n-gram data (which is ranked) and incorporate that, if it makes sense."

{ via information aesthetics }

Picasa's Hello Discontinued


Hello, Picasa's obscure instant messenger, has been discontinued and will be completely shut down on May 15th. Hello's last major update was released in January 2005, six months after Picasa was acquired by Google.

According to its site, "Hello is a program that lets you connect directly with your friends to share your digital pictures. (...) With Hello, you just pick what pictures you want to show off, and click Send. That's it. Hello takes care of all the hard work. And you and your friends can download full resolution, print-quality pictures from each other, while you're doing more important things, like talking about your pictures. (...) With Hello's file sharing technology, [your friends] only have to download high quality versions of the [pictures] they really like. Everything else comes in at a smaller size, optimized for viewing on-screen."

"Hello enables users to instantly share images securely over a peer-to-peer network and chat about them. Hello currently has more than 250,000 users and can be used with or without Picasa, the company's flagship digital photo organizing software," mentioned Picasa in a press release from May 2004. It would be interesting to know the number of active users as of today, but it's probably much lower than 250,000.

The instant messenger was integrated with Picasa, so you could select the photos from Picasa and send them to Hello, but also view in Picasa the pictures you receive. Like Picasa, Hello was easy to use and shared the same clean interface. Despite its usefulness, Hello should have been a feature in a more powerful IM client, not a stand-alone application.


If you haven't tried Hello, you can still download the application from http://updates.picasa.com/hello/651/Hello651.exe. You and all your contacts need to have Picasa accounts. According to hello.com, you can still use the application one more week.

Here's Hello's goodbye:

"We originally embarked on a mission to make photo sharing easier and more fun with Hello. We plan to keep carrying that torch in new projects to come. We hope that you continue to enjoy the other sharing products Google offers including Picasa, Picasa Web Albums and Google Talk."

and here are some reactions:

"I live on Hello because it allows me as a designer to show my clients the screen I am working on real time. It also allows me to communicate with other designers who work for my business and mentor them through the design process, or get myself some mentoring!"

"i credit hello for leading to some very good friendships of mine all over the US, and i credit hello for my transition from crafter to artist. i do not want to lose this program."

"PLEASE reconsider the decision to shut down Hello. It is an invaluable tool for the digital scrapbook community, especially the designers. I can't imagine what I would do without the ease Hello provides to show off screen shots and get critique on my Designs."

"I use Hello daily in conversation and to share photos with family members out of state. It has also been very useful in learning and teaching applications by using real time screen shots. There just isn't anything else that compares that is so easy to use!"

"My mother who has no idea what IM even stands for, uses Hello everyday. We send her pictures of her grandkids, she shows us pictures she's taken, its awesome! She only uses it because it is simple and user friendly. IM programs like Google Talk, Yahoo, MSN or Goober are way too complex for her to figure out and I doubt she is the only one in this situation."

{ via Google Blogoscoped }