Archive for November 18th, 2009
On the surface these are happy days in Android-land.
Going into the key Christmas selling season, Android is eating Windows for lunch. New (non-Google) development centers are continuing to open, new manufacturers are coming on stream.
What could possibly go wrong?
Knowing that rising markets need a wall of worry to keep going up, here are some possibilities:
- Momentum must be maintained. Once you start gobbling market share you have to keep doing it. Even a slowing of momentum can be read as failure.
- Developers must be kept happy. Some are complaining they’re working full-time getting apps written for the Android operating system running on multiple phones.
- The Android app store has some catching-up to do, especially in the user experience area.
- When will Android get a “killer app” that the iPhone can’t match, or one it hasn’t already matched?
- Can Google ride herd on its complex ecosystem? Everyone knows who the boss is with the iPhone. Not so with Android. (UPDATE: Rich Sands writes to say this article from his site is more to the point.)
We shouldn’t get too excited about Android’s early success. A 3.5% market share is still a gnat on Apple’s elephant. Early buzz does not make for victory — as President Howard Dean will tell you. (Or President Huckabee, if you prefer.)
Google has set itself a more complex task than that which faced Apple when it introduced the iPhone a few years ago. Google is using an open source approach, which means there are more hands on the steering wheel. And Google is trying to overcome an established leader, leading to charges of me-tooism.
A good start is not the race.
![]()
![]()
At 46, Larry Augustin is much too young to be the grand old man of anything.
But he is one of the grand old figures of open source. He was in the group that coined the term back in the late 1990s.
Larry was Sourceforge, he was VA Linux, back during the dot-boom 10 years ago. He rode the stock to $240/share, then watched it plummet to nearly nothing in the dot-bomb.
Once had had a Web site, made it run. Made it race against time. Once he had a Web site, now it’s done, buddy can you spare a dime?
But Larry kept his hand in. He became an angel investor and adviser, a “go-to” guy for any open source start-up looking for some street cred. His current bio has him on 9 different corporate boards, topping the list of the most influential people in open source a few months ago. (Our own Matt Asay was number two.)
The news today is that Larry is the permanent CEO of SugarCRM. Appointed on an interim basis in May to replace co-founder John Roberts, he has pointed the software into the cloud, adding a business model to its large community.
What has he learned in that time?
- Influence is not a contest among bloggers. Augustin’s last blog post is dated July.
- Companies grow through teams. The next negative word about SugarCRM’s people I hear from Larry will be the first.
- The future of open source is in SaaS, in the cloud.
Larry Augustin’s story is proof that second acts in business are possible. Most of those who boomed during the dot-boom were never heard from after the dot-bomb. But not Larry.
His story reminds me of the man who was managing my Atlanta Braves when I first moved here in the early 1980s. He won a pennant, got fired, got kicked around. He got a few gigs, did some broadcasting, advised a few people here and there.
Then he got another shot, in New York, and he took it. His second chance made Joe Torre a sure Hall of Famer.
I can’t guarantee Larry Augustin the same success, but after a decade in the wilderness he has a team again. Hard for me not to root for the “old guy.”
[Sponsored]
Back in 2001, to give people a new, quicker way to find images, we launched Image Search. When you do a search for [eiffel tower] you’ll find an array of images of the tower in the daytime, in black and white, at sunset and more. With Similar Images, which recently graduated from Google Labs, you can click “Find similar images” to narrow your search to, say, pictures of the Eiffel Tower lit up at night. Today, we’ve launched an experimental feature in Labs called Google Image Swirl, which builds on new computer vision research to cluster similar images into representative groups in a fun, exploratory interface.
For example, if you search on Image Swirl for [washington], you’ll see 12 image thumbnails including President Washington, the Washington Monument, a map of Washington D.C. and the Capitol Building:
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. For example, here’s what you’ll see if you click the image of the Washington Monument:
You can then further explore additional sub-groups within any cluster. The interface may look familiar to those of you who have tried Google’s Wonder Wheel available in the Search Options panel, which enables you to explore related search queries.
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.
Image Swirl currently works for more than 200,000 queries and we plan to include more queries in the future. Available queries will auto-complete as you start to type in the search box, similar to Google Suggest.
You can try out Google Image Swirl in Google Labs today. The feature is experimental and the underlying technology is a work in progress, so please share your feedback and let us know how we can make it more useful.
Posted by Aparna Chennapragada, Product Manager, and Yushi Jing, Google Research
As many of us recall from our civics lessons in school, the United States is a common law country. That means when judges issue opinions in legal cases, they often establish precedents that will guide the rulings of other judges in similar cases and jurisdictions. Over time, these legal opinions build, refine and clarify the laws that govern our land. For average citizens, however, it can be difficult to find or even read these landmark opinions. We think that’s a problem: Laws that you don’t know about, you can’t follow — or make effective arguments to change.
Starting today, we’re enabling people everywhere to find and read full text legal opinions from U.S. federal and state district, appellate and supreme courts using Google Scholar. You can find these opinions by searching for cases (like Planned Parenthood v. Casey), or by topics (like desegregation) or other queries that you are interested in. For example, go to Google Scholar, click on the “Legal opinions and journals” radio button, and try the query separate but equal. Your search results will include links to cases familiar to many of us in the U.S. such as Plessy v. Ferguson and Brown v. Board of Education, which explore the acceptablity of “separate but equal” facilities for citizens at two different points in the history of the U.S. But your results will also include opinions from cases that you might be less familiar with, but which have played an important role.
We think this addition to Google Scholar will empower the average citizen by helping everyone learn more about the laws that govern us all. To understand how an opinion has influenced other decisions, you can explore citing and related cases using the Cited by and Related articles links on search result pages. As you read an opinion, you can follow citations to the opinions to which it refers. You can also see how individual cases have been quoted or discussed in other opinions and in articles from law journals. Browse these by clicking on the “How Cited” link next to the case title. See, for example, the frequent citations for Roe v. Wade, for Miranda v. Arizona (the source of the famous Miranda warning) or for Terry v. Ohio (a case which helped to establish acceptable grounds for an investigative stop by a police officer).
As we worked to build this feature, we were struck by how readable and accessible these opinions are. Court opinions don’t just describe a decision but also present the reasons that support the decision. In doing so, they explain the intricacies of law in the context of real-life situations. And they often do it in language that is surprisingly straightforward, even for those of us outside the legal profession. In many cases, judges have gone quite a bit out of their way to make complex legal issues easy to follow. For example, in Korematsu v. United States, the Supreme Court justices present a fascinating and easy-to-follow debate on the legality of internment of natural born citizens based on their ancestry. And in United States v. Ramirez-Lopez, Judge Kozinski, in his dissent, illustrates the key issue of the case using an imagined good-news/bad-news dialogue between the defendant and his attorney.
We would like to take this opportunity to acknowledge the work of several pioneers, who have worked on making it possible for an average citizen to educate herself about the laws of the land: Tom Bruce (Cornell LII), Jerry Dupont (LLMC), Graham Greenleaf and Andrew Mowbray (AustLII), Carl Malamud (Public.Resource.Org), Daniel Poulin (LexUM), Tim Stanley (Justia), Joe Ury (BAILII), Tim Wu (AltLaw) and many others. It is an honor to follow in their footsteps. We would also like to acknowledge the judges who have built this cathedral of justice brick by brick and have tried to make it accessible to the rest of us. We hope Google Scholar will help all of us stand on the shoulders of these giants.
Posted by Anurag Acharya, Distinguished Engineer
Google usually shows a green web address, or URL, at the bottom of each search result to let you know where you’re headed. Today we’re rolling out an improvement that replaces the URL in some search results with a hierarchy showing the precise location of the page on the website. The new display provides valuable context and new navigation options. The changes are rolling out now and should be available globally in the next few days.
Some web addresses help you understand the structure of the site and how the specific page fits into the site hierarchy. For example, consider a search for the biography of Vint Cerf (Google’s Internet Evangelist). The URL for one result, “www.google.com/corporate/execs.html,” shows that the page is located in a page about “execs,” under “corporate,” which is on the “google.com” site. This can provide valuable context when deciding whether to click on the result.
Often, however, URLs are too long, too short, or too obscure to add useful information. For example, consider this result from ProductWiki for the query [spidersapien reviews]:
The URL of this result is “www.productwiki.com/spidersapien,” which doesn’t provide much additional information about the site or this result. Now take a look at the result with the new site hierarchy display:
The new text provides useful information about the page. You can tell that the ProductWiki site has information about many different products, organized in different categories, and you can even tell that Spidersapien is a robot toy. In addition, each phrase in the green line is actually a link. For example, clicking on “Toys & Games” takes you to ProductWiki’s listing page for all toys, and clicking on “Robots” takes you to a list of their robot toys. This way if you realize that you’re interested in a more general category than this specific product (there are a lot of cool robot toys out there) you can easily access information on broader topics.
The host and domain for the site (in this case www.productwiki.com) will always be shown, so you always know what website you’re going to before you click. There’s not always enough room to show the complete hierarchy, so sometimes we use ellipses to replace some of the intermediate levels, like in this result for [how to make granola]:
The information in these new hierarchies come from analyzing destination web pages. For example, if you visit the ProductWiki Spidersapien page, you’ll see a series of similar links at the top, “Home> Toys & Games> Robots.” These are standard navigational tools used throughout the web called “breadcrumbs,” which webmasters frequently show on their sites to help users navigate. By analyzing site breadcrumbs, we’ve been able to improve the search snippet for a small percentage of search results, and we hope to expand in the future.
When we design the way results appear on google.com, our goal is to get you to the information you’re looking for as quickly as possible. Sometimes that means improving how we represent websites, and other times that means giving you new ways to explore content. We’re always happy when we can introduce a feature, like site hierarchies, that does both!
Posted by Harvey Jones and Daniel Rocha, Software Engineers, Result UI Team
