Welcome to Technology News

Google’s translation center: Language lessons for the Googlebot?

Autor admin | 05.08.2008 | Category Google, Internet News

Google looks set to launch a beta test of a document translation service, a new move in the company’s efforts to break down language barriers.

 

Court orders YouTube to give Viacom video logs

Autor admin | 05.07.2008 | Category Internet News, YouTube

NEW YORK - Dismissing privacy concerns, a federal judge overseeing a $1 billion copyright-infringement lawsuit against YouTube has ordered the popular online video-sharing service to disclose who watches which video clips and when.

The Yahoo + Google - Microsoft spin room

Autor admin | 15.06.2008 | Category COMPANIES, Google, Internet News, Microsoft, Yahoo

With the Microsoft/Yahoo/Google triangle taking a new shape as Microsoft exited and Yahoo and Google connected, the analysts covering tech industry sports are weighing in with their opinions.

Some Wall Street analysts believe Microsoft will take another run at Yahoo if the company can’t get back on track or Carl Icahn wins his proxy fight to control the Yahoo board. That may be wishful thinking. Kara Swisher reports that Microsoft is done with its courtship of Yahoo and nothing will bring them back to the negotiating table.

Mike Arrington of TechCrunch called the Yahoo-Google deal a massive destruction of shareholder value, employee morale, and the Interent balance of power:
Yahoo’s hatred of Microsoft runs so deep that they were actually, in the end, willing to destroy the future of their company just to keep it independent for a short while longer. They’ve ignored the wishes of their shareholders, employees and many now former key employees in killing that deal. And apart from Google, CEO Jerry Yang, President Sue Decker and possibly Tim O’Reilly, I don’t believe there is anyone in the world that is happy with what has happened.

In a further lambasting post, Arrington called Yahoo desperate and possibly neurotic:
Quite simply, it looks to me like Yahoo is effectively paying Google off to step in and (1) keep Jerry Yang, Sue Decker and the current board of directors in power, and (2) avoid a desperation deal with Microsoft for as long as possible, or longer. It’s not even clear to me that Google wants this deal, based on the terms. It almost looks like they’re just doing Yahoo a favor, and trying to keep them out of Microsoft’s hands.

At the other end of the spectrum, venture capitalist Fred Wilson thinks that Yahoo did the right thing by choosing Google over Microsoft as a partner.
Yahoo! finally woke up and did what they should have done years ago, cede search monetization to Google who simply does it better and will always do this era of search better than anyone else.
Now Yahoo! will do what it needs to do. Clean house, get lean, get out of businesses it shouldn’t be in. Focus on what it’s good at. And start making money and growing again.
They may need new leadership to do that. But selling this asset to Microsoft just because they had the wrong leadership and probably still have the wrong leadership is a mistake.

From my reading of the events over the last five months, Yang regrets that Microsoft walked away from the acquisition talks. “We all felt and understood a combination done right has a tremendous amount of power and leverage,” Yang said during an interview with Walt Mossberg at the D6 conference.

As a founder, Yang preferred that Yahoo stay independent and that he have the chance to turn the company around as CEO. Microsoft historically was not the kind of partner that Yang considered for a marriage. And his board of directors, led by non-executive Chairman Roy Bostock, seemed to go along with that line of thought.

But the entire affair turned out to be mostly about the money, as Decker admitted. “We never got through the price door,” she stated during the same D6 interview. Yahoo’s board believed that the company was worth more than $35 per share based on future promise, and Microsoft wasn’t on the same page. In effect, Microsoft called Yahoo’s bluff.

It also wasn’t helpful that Yahoo was negotiating the search deal with Google at the same time Microsoft was pursuing its hostile bid. After months of rejection, Microsoft basically became less enchanted with the potential marriage, and despite the pummeling from the shareholders, Carl Icahn’s camp, and the press, Yang and his advisors held out for more money.

Unable to come to terms with Microsoft on a generous deal just for the search business, Yahoo took the less complicated, non-exclusive Google deal that allowed the company to remain in the search game.

As I wrote in my post “The battle for Yahoo’s soul,” Jerry Yang and Sue Decker have a short runway–about six months–to prove that they can “redefine” the essence of Yahoo in a way that yields more revenue, profit, and positive buzz. With the continuing board room distractions, employee defections, and morale issues that go along with being under siege by various parties, the duo have their work cut out for them.

Google grows stronger in Microsoft-Yahoo fallout

Autor admin | 14.06.2008 | Category Google, Internet News, Microsoft, Yahoo

SAN FRANCISCO - Microsoft Corp.’s abandoned takeover bid for Yahoo Inc. appears to have culminated with a disheartening thud for those two companies but amounted to yet another coup for online search leader Google Inc.

What began in January as Microsoft’s most audacious attack yet on Google instead paved the way for the Internet’s most powerful company to gain even more clout through a deal that gives Google access to a large chunk of Yahoo’s advertising space.

By submitting to a partnership that endorses Google’s search advertising technology as a better choice than its own, Yahoo is giving online marketers even more incentive to spend most of their money with its biggest rival, according to industry analysts.

It looks like such a sweet deal for Google that the U.S. Justice Department and lawmakers are expected to take a hard look at the arrangement to make sure it doesn’t give Google too much control over the Internet’s search advertising market.

Google currently has about 75 percent of the U.S. search advertising market followed by Yahoo at 9 percent, according to the research firm eMarketer Inc.

Although they contend their alliance won’t lessen competition, Google and Yahoo have agreed to wait until late September to begin working together so the U.S. government has more time to assess the potential impact.

Even more importantly to Google, the Yahoo partnership keeps a potentially valuable weapon out of Microsoft’s control.

Without Yahoo’s renowned franchise, Microsoft once again is scrambling to find a way to fix its unprofitable online operations and narrow Google’s commanding lead in the Internet’s rapidly growing ad market.

Google shares gained $18.56 to close Friday at $571.51 while Microsoft shares added 83 cents to close at $29.07 — an indication that some investors were relieved the world’s largest software maker concluded it would be too expensive and troublesome to buy Yahoo.

On the other side of the fence, Yahoo shareholders had been clinging to the possibility that Microsoft would revive its last offer of $47.5 billion, or $33 per share, to buy the Internet pioneer. But those hopes evaporated late Thursday after Yahoo disclosed Microsoft had “unequivocally” rebuffed an attempt to renew the negotiations.

In a sign of investors’ frustration, Yahoo shares dropped as much as $1.77, or 7.5 percent, Friday before rallying late in the session to finish at $23.47, down five cents. The downturn marked Yahoo’s lowest stock price since it closed at $19.18 at the end of January, just before Microsoft launched its takeover attempt.

That leaves Yahoo’s market value 29 percent below Microsoft’s last offer, which was withdrawn May 3 after Yahoo asked for $37 per share. Yahoo’s stock hasn’t reached that price since January 2006.

At least Microsoft still has a strong, highly profitable backbone — a suite of software products that run most computers around the world.

Yahoo, though, may have made a Faustian bargain by hiring Google to show ad links next to a significant portion of the ad links appearing alongside search results on its Web site in the United States and Canada. The Sunnyvale-based company also will pluck Google ads to show on other Web sites in its marketing network.

Yahoo expects its annual revenue to get an $800 million lift from the arrangement with Google while still showing show the majority of its own ads alongside its own search results. But most analysts viewed it as an act of desperation, asserting it’s only a matter of time before advertisers shift all their business to Google because they know their messages will show up on Yahoo either way.

Deutsche Bank analyst Jeetil Patel described Yahoo’s decision to farm out advertising to Google as “one of the worst strategic maneuvers seen in the Internet industry.”

Google will get such great access to Yahoo’s highly trafficked Web site that it should be able to gather more insights about the correlation between search requests and advertising, ThinkPanmure analyst William Morrison wrote in a Friday research note titled “Giving Away The Store (To Google).”

And that additional data could help Google further improve its advertising formula to become an even more compelling marketing magnet.

The partnership also cast doubt on a turnaround plan Yahoo co-founder Jerry Yang began drawing up year ago after he replaced Terry Semel as the Sunnyvale-based company’s chief executive.

A big part of that strategy hinged on Yahoo becoming a “must-buy” for advertisers — a strategy that the Google deal appears to contradict.

“This raises very important questions about the long-term vision for (Yahoo) and its place in the industry,” said Cantor Fitzgerald analyst Derek Brown.

Yahoo shareholders will get a chance to vent their frustration at the company’s annual meeting Aug. 1 when activist investor Carl Icahn will seek to replace the board with nine alternate candidates.

Icahn was primarily interested in selling Yahoo to Microsoft, so his campaign to replace the board may be hurt if he can’t persuade shareholders he has other viable ideas on how to boost Yahoo’s stock price. He didn’t return a call seeking comment Friday.

Yang and his top lieutenant, Susan Decker, defended the Google deal as a profitable move that will better position the company to capitalize on the Internet advertising market’s growth from roughly $40 billion worldwide this year to a projected $75 billion to $80 billion market in 2011.

Microsoft contends it offered Yahoo a better alternative even after losing interest in buying the entire company.

When the latest talks broke off June 8, Microsoft was prepared to buy Yahoo’s search operations for $1 billion and pay $35 per share to accumulate $8 billion worth of Yahoo’s stock, according to an internal note sent Friday by Kevin Johnson, who oversees Microsoft’s online operations.

Microsoft also would have offered guarantees that could have boosted Yahoo’s operating cash flow by an estimated $1 billion annually, Johnson wrote.

Yahoo estimates the Google partnership will increase its operating cash flow by $250 million to $450 million annually.

“Regardless of Yahoo’s decision, we will continue to move forward on our strategy in online services and advertising,” Johnson assured Microsoft employees.

Microsoft left the door open to renewing talks about buying Yahoo’s search operations. Yahoo also gave itself some wiggle room by including a clause in the Google partnership that would end the alliance for a termination fee of up to $250 million.

Some analysts and investors still think Microsoft eventually might try to buy Yahoo in its entirety, although at a price well below $47.5 billion.

“Yahoo seems to have backed itself into a corner pretty effectively here so it would appear Microsoft has a lot of leverage,” said Dan Davidowitz, a portfolio manager for Polen Capital Management, which owns about 750,000 shares of Microsoft and 37,000 shares of Google.

Davidowitz said he isn’t interested in owning Yahoo’s stock.

Yahoo!

Google lets users measure the power of words

Autor admin | 11.06.2008 | Category Google, Internet News

SAN FRANCISCO (Reuters) - Number-crunchers can rejoice as Google Inc offers deeper access to the underlying figures for users’ Web searches, giving some insight into trends based on the relative popularity of various words.

The Internet search leader is expanding its existing Google Trends service to allow users to see underlying numerical data on the popularity of any particular search in Google’s vast database of search terms, relative to others.

Google Trends was begun two years ago as an entertaining but limited way to indicate what the world is thinking about over time, at least in terms of Web searches.

Now Google is giving users the ability to search across terms in its database, instantly chart how they compare to other search terms, then export the underlying numerical data into a common spreadsheet format to compare with other data.

Google Trends (http://trends.google.com/) lets users compare demand for various search terms and see how popularity differs across geographic regions, cities or languages.

A year ago, the company introduced Hot Trends, which gave users insight into fast-rising Web search trends with data refreshed several times daily. The tool’s power only grows as people conduct more and more of their everyday activities online, with Web search often their primary starting point.

The data in Google Trends stretches back to 2004. While the service is based on the many billions of individual searches performed each year, Google Trends only reveals data on the aggregate numbers of searches, not the searches themselves.

National differences in the endless human search for sex or love can vary widely, according to a Google Trends chart. http://tinyurl.com/5jt5ce/

Google Trends users can also chart the explosion of interest in the term “backdating” since 2006, reflecting the scandal over how hundreds of companies backdated options for executives. http://tinyurl.com/5l8osu/

Searches for the word “Microsoft” had a more than two-to-one-lead in searches over “Apple” three years ago, but Apple had virtually closed the gap by the end of 2007.

Then news reports of its takeover bid for Yahoo appears to have stoked a recovery in Microsoft this year. Searches for Microsoft have outnumbered those for Apple by about 7 to 5 in recent weeks, according to Google Trends data.

Users must be registered and signed into a Google account to use the service. One can then see the evolution of new terms or concepts through Google searches, including the rise of “Google Trends” itself. http://tinyurl.com/6zd6pg/.

(Editing by Braden Reddall)

Yahoo!

Google lets sites tailor what searches users see

Autor admin | 03.06.2008 | Category Google, Internet News

SAN FRANCISCO (Reuters) - Internet search leader Google Inc is expanding an existing service to let businesses customize the search results that visitors see when they search for information within their own sites.

Instead of offering visitors to a particular website the same slice of search results they find by searching Google.com, the new Google Site Search Service lets website owners show previously untracked pages deep inside their sites.

The new Google service is hosted on Google’s network of computers instead of requiring customers to install and maintain search equipment of their own. Google aims to improve the search quality of sites with rich content but cluttered ways of finding information lodged within the sites.

Site Search is the new name for Custom Search Business Edition, which Google introduced in late 2006.

It adds business integration features through a machine-readable computer programming interface, the option to turn off advertisements, a more tailored look-and-feel for searching the site and technical support via e-mail or phone.

“Google Site Search is targeted more for businesses and government sites that want search but don’t want to display ads,” Nitin Mangtani, a product manager in Google’s Enterprise division, said in a phone interview.

Hundreds of thousands of website publishers rely on Google’s AdSense program to offer classic Google Web search. They make money running advertisements from Google’s network of online advertisers.

Site Search is an alternative to the Google Search Appliance, which Google offers to sites wishing to maintain their own search services inside their own data centers. Google counts more than 10,000 active Search Appliance customers.

Google is the only mainstream Web search provider to offer businesses a hosted service. By contrast, rivals like Autonomy Corp Plc and FAST, which was recently acquired by Microsoft Corp, require customers to install and manage their search software in-house, Mangtani said.

Other features allow site administrators to add their own customized synonym dictionary, allowing website visitors to use site-specific terminology geared to a particular industry or culture. Google Sites also offers users a spell checker.

Webmasters have the option of fully customizing Google Site Search to the look and feel of the website, or they can rely on Google’s simplified Web page style and make clear the ties to Google’s services.

They can also weight search results to favor more recent documents or specific pages such as product catalogs. While the service encourages sites to customize site search results, the service does not affect the ranking of searches done via Google.com, which promises to keep search results impartial.

Yahoo!

Google brings Friend Connect to the masses

Autor admin | 12.05.2008 | Category Google, Internet News

As expected, Google has unveiled a preview of Friend Connect, a way to add social features to a Web site without programming.

David Glazer, director of engineering at Google, described Friend Connect, whose site is inaccessible Monday morning, as plumbing for the rest of the Web.

“The Web is getting better by getting more social. We’ve baked social features into the infrastructure of the Web, and it is not tied to any particular site,” Glazer said. “Users can interact with any of their friends anywhere they go on Web, and with any app.”

I asked Glazer if Friend Connect is a response to Facebook Connect and MySpace.com’s Data Availability. “People will speculate a lot in that direction. We didn’t create this code in the three days (since Facebook and MySpace made their announcements).”

Unlike Facebook and MySpace, Google lacks a dominant, centralized social-networking hub. Friend Connect works the edges of the Internet, applying an open and distributed approach, and bringing a social dimension to the 99-plus percent of sites that aren’t socially enabled.

“The distributed model has worked well for the Web. That is what the Web does–many points of light loosely coupled and massively distributed, allowing users to connect to pages of information,” Glazer told me. “Now it is working to connect people to other people.”

Friend Connect-compliant sites will be able to view, invite, and interact with newfound friends, or with existing friends, from established social-networking sites, including Facebook, Google Talk, Hi5, Orkut, and Plaxo via secure authorization application-programming interfaces.

Currently only a few sample sites, including Google’s Guacamole site, are available to end users. “We are looking to get feedback from Web site owners about what kinds of sites and apps they want,” Glazer said. Ingrid Michaelson, an independent musician, integrates iLike’s OpenSocial application with Friend Connect to connect friends without having to leave the site.

 John McCrea, vice president of marketing at Plaxo, said Google’s Friend Connect is “flipping the model” from walled gardens (such as Facebook) to a more open social Web:

Instead of widgetizing apps and bolting them on to some corporation’s proprietary social graph, why not widgetize the social graph and socially enable any Web site or Web page?

That’s a big, bold vision that Plaxo is 100 percent aligned with. As to Facebook and MySpace, it is certainly great to read the rhetoric they are now putting forth. The meme of data portability, open social Web, and bill of rights for users of the social Web has certainly caught on!

Alas, the devil is in the details, and we haven’t seen any details (yet) from Facebook–just a Friday blog post signaling intent. It might be great, and we hope it is, but it’s not clear what the actual substance will be.

With regard to MySpace, the rhetoric is over-the-top goodness, including a declaration of the end of the era of walled gardens. Alas, the details, as they currently exist, for their “Data Availability” effort fall far short of the vision many of us share for users having ownership of their data, control over who can see it, and freedom to take it with them, wherever they go across the social Web.

In the MySpace “Data Availability” model, the user can take their data for a walk anytime they want or to any place they want, but the data remains on a tether. There is no notion of copy, move, or sync. Participating sites must agree to have MySpace serve the data live in their page. That’s a half-step wrapped in a beautiful flag of openness.

“Friend Connect provides wizardlike pages. Webmasters just fill in the information, select social apps, copy code, paste, and save. No coding is required. It passes the ‘easy’ test, and it does something useful,” Glazer said. It provides features such as user registration, invitations, member galleries, message posting, and reviews, as well as OpenSocial applications.

At the core of Friend Connect are three emerging social standards–OpenID, oAuth, and OpenSocial.

“Today is the right time to connect all emerging standards to give users the ability to go anywhere on Web and interact with any set of friends on any application,” Glazer said.

Google’s Social Graph API is not part of the Friend Connect preview, Glazer said. “The Social Graph API is part of the same conversation, but we didn’t need to connect those two dots.”

Glazer emphasized that Google is focused on keeping users in control of their information. “The Webmaster has no business knowing who my friends are, but I can choose to link my login to my Facebook account and invite friends,” he said. “It’s up to each site to publish APIs, with appropriate terms of use,” Glazer told me. “I would expect as Friend Connect matures in the market, we will see more people connecting to it and more standard interfaces to turn on and register for it. It’s not fully standard now.

Friend Connect covers many of the use cases for the social Web, but a single, standard “friend” API is still lacking.

“There are a few good candidates, such as the OpenSocial RESTful APIs, which are at a rough consensus stage but not running code,” Glazer said. “We don’t know enough to call a winner, but there will be a standard.”

CNET news.com

Alarm at Google Yahoo partnering

Autor admin | 12.05.2008 | Category Google, Internet News, Yahoo

Regulators in the US are being urged to investigate any potential online advertising and search partnership between Google and Yahoo.

The call by a coalition of 16 American civil rights and rural advocacy bodies comes despite the fact no firm deal has actually been announced.

“We all suffer in such mega mergers,” Gary Flowers of the Black Leadership Forum told BBC News.

The justice department is examining a trial the companies did in April.

It has been widely reported that it is looking into the anti-trust implications of last month’s two-week test.

However, the department says it has no comment on the coalition’s demands because there is no definitive agreement between Yahoo and Google at the moment.

But reports say that the two companies are presently hammering out the intricacies of a future potential advertising and search agreement, and are sharing their plans with antitrust regulators.

At Google’s shareholder meeting on Thursday, Chairman Eric Schmidt said: “If there were a deal [with Yahoo], we would anticipate structuring the deal to address the anti-trust concerns that have been widely discussed.”

‘Never positive’

This assurance is not good enough for the coalition which is made up of the League of Rural Voters, the National Black Chamber of Commerce and the American Agriculture Movement.

It also includes the Black Leadership Forum, an umbrella group of 36 civil rights organisations including the NAACP and the National Urban League.

In a letter to Assistant Attorney General Thoma Barnett, head of the Justice Department’s anti-trust division, the coalition argues that such a deal would give Google almost 90% of the search advertising market and strengthen its influence over internet users’ access to information.

“We face a possible future in which no content could be seamlessly accessed without Google’s permission,” the letter states.

The effect Mr Flowers says of such large partnerships is never positive and would for the black community, as for other communities, “condense competition, increase prices and limit new business opportunity on the internet”.

‘Do no evil’

League of Rural Voters’ executive director Niel Ritchie claims that the do-no-evil mantra may no longer apply in today’s marketplace in which Google’s reach is apparently without bound, touching more and more aspects of our everyday lives.

“We believe the government should give this agreement very careful scrutiny,” he says.

Mr Flowers says:

“Google has already exhibited a pattern of violating privacy, engaging in anti-competitive conduct and using its monopoly power in the search market to drive internet users to its affiliated services and its viewpoints on policy matters.

“Any joint combination with Yahoo could dramatically worsen these problems.”

The Centre for Digital Democracy, a consumer advocacy group, is also willing to push regulators to block any deal and wants European consumer groups to raise concerns with European Union officials.

“You can’t allow Google to operate a portion of its leading competitor out of its back pocket,” Jeffrey Chester executive director of the CDD told the Associated Press.

There has been no comment from Yahoo or Google.

BBC

20 (Rare) Questions for Google Search Guru Udi Manber

Autor admin | 01.05.2008 | Category Google, Internet News

Since there has been such a thing as Web search, Udi Manber has been working on Web search. Previously a computer science professor at the University of Arizona, then a senior vice president at Amazon and Yahoo’s chief scientist, Manber is now vice president in charge of search quality for Google, where he makes sure results are engineered to the utmost (near) perfection. In one of the only public interviews he’s ever sat down for, Manber gives PM a glimpse into how Google’s dominant engine helps you find what you want, how you can help it find you and how search is constantly evolving with the pace of technology. —Glenn Derene

How have you seen search evolve in the time that you’ve been working in the field?
It’s been tremendous. I like to say that it’s almost science fiction every five years. When the first search engine appeared in ‘94, compared with when I came out of academia in ‘99, compared with the way it was in 2003, compared with the way it is today—every five years there have been just incredible advances. What we do now, we couldn’t have foreseen 10 years ago. Today we’re finding a lot more information, and the questions are getting a lot harder. People expect more from us.

Does the increase in information make the job of search exponentially harder because the volume of material is so high, or easier because of the increased linkage of information?
I think some of it is easier, but overall it’s harder, because people’s expectations are so high. Ten years ago, when you found anything, it was a miracle. Okay, I shouldn’t say a miracle, but a surprise: “Hey, look at this! I got what I wanted.” Today, if you don’t find everything you’re looking for, then something is wrong. And that’s the way we look at it, too. If your question is clear, you should get the answer. But now we’re getting into a situation where the questions are no longer always clear. People want more complicated information, and that’s much harder.

What do you think a person expects from a Web search?
They want to get an answer. Our goal is very simple: We want to return to the user the answer that they need.

I’ve noticed, anecdotally, when watching people search, that they will rephrase their query over and over again until they get a proper answer. To what extent can that be fixed on the search engine side?
Many ways. First, we take that into account. The results we show you are based not only on what we know of the Web, but also what other people have searched for. Second, we are developing more tools to allow you to refine your queries—at the bottom of many pages, you’ll see query refinements. These are suggestions from us about what your next query should be. And we put it at the bottom because that’s where you run into problems—you tried to read the page, you didn’t find what you want, you may need other suggestions. Plus, we’re working on many other ways to help you with this process. [Search] is clearly a process.

What frustrates you when you do a search?
Whenever I don’t find exactly what I need. It’s frustrating. We’re very concerned with that here. When someone finds an example of something that doesn’t work that should work, we think, “How can that happen? How is it that we missed something?” Sometimes it’s a weakness in the algorithm, very often it’s something missing from the Web, like a restaurant that does not put their location on their page.

Do you find that the content on the Web is evolving to be more search-engine friendly?
It’s hard to say. It’s definitely still lacking. I wish people would put more effort into thinking about how other people will find them and putting the right keywords onto their pages.

How does search change when the content provider is in a more crowded space? Say, for instance, there are 50 Chinese restaurants all in the same area …
It goes two ways: The content provider should think about how users will look for their content, and the user should think about what words people use to write about their content. Very often people make the mistake of using a search engine as if they are talking to another person. They use all sorts of words that a person will understand, but are not going to be in the content they are searching for. You should think about what you expect to see in the actual page and search for that. Having said that, we’re doing this, too. We will take your query and try to “understand” it and match is as best we can to the content we find on the Web.

What makes Google philosophically different from all the other search engines? What is Google searching for that others aren’t?
I don’t think it’s about philosophy. It’s about getting people what they need, and about getting the results to be as accurate and fast as possible. We’re innovating, and concentrating just on the relevancy of results. Last year we made over 450 improvements to the algorithm.

There have been a lot of fads in search of late, such as Human Assisted Search and contextual search. Do those get folded into search as a whole? What are real trends in search and what are fluff?
So let me first tell you about Google. At Google we do not manually change results. For example, if we find for a particular query that result No. 4 should be result No. 1, we do not have the capability to manually change it. We made that decision not to put that capability in the algorithm—we have to go and actually change the algorithm. That is, we have to find what weakness in the algorithm caused that result and find a general solution to that, evaluate whether a general solution really works and if it’s better, and then launch a general solution. That makes the process slower, but it puts a lot more discipline on us and makes it more unbiased.

Whether it’s at Google or not, is there a market for human-assisted search, or is that something different?
I think that the general issue is, how do you get more input from people? How do you get people to contribute more information, more content? Search is about getting lots of signals and putting them all together. The art of ranking is, how do you collect lots of signals then put them together? Signals from people are the best signals. We have several tools—and we’re going to launch many more—that will encourage people to contribute more. This does not necessarily mean one should then create the search results manually.

I’ll give you an example of something that came last week. We were evaluating a certain algorithm that adds diversity to the result. We did live experiments, which means we launched the algorithm to a very small percentage of users and then see how that compares to the result without the algorithm. One of the queries that made a difference: The query was, New York Times address. And you would think you’d understand the query, and the first result right there on the snippet gives you The New York Times. It turns out that’s not what the user was looking for. They were looking for an address given out by a New York Times reporter the day before. And because of this diversity and because of our emphasis on freshness and highlighting fresh results, that particular address appeared somewhere in the results, and that’s what the user wanted—that’s what they went to and got the result. That was something that surprised even us. You don’t think that when someone searches for New York Times address that they’re not looking for the address. Language is like that. Intention can be ambiguous.

Putting privacy aside, to what extent does finding a profile of somebody help search?
Currently, if you allow us to keep your Web history, we will improve your search. By the way, if you do this, you can always go back and remove what you want to remove or remove the whole thing or revoke that permission. But it improves search in two ways. One is, we will tune the result for you slightly. We’re not going to change the whole page—we might change position 5 to position 3 here and there, but we’ll use whatever we can from your previous searches to adapt the current search to you. The second is, we allow you to search within your Web history, which can also be very useful. You may remember something you did three months ago and you don’t remember exactly how you did it.

Could that theoretically extend back forever in time? Is there a limit to how far back something like that could extend arbitrarily, or is there a useful limit?
When we look at the personal search algorithm, obviously time gets into it. As far as you’re concerned, if you want us to keep this, it’s up to you.

Is there a literally a slider of some sort where you say, 1 month, 3 months, etc.?
I don’t believe we do that, but that’s something we can consider if that’s a big issue. I don’t think it’s a big issue. I think it’s better to keep because you might need something from two years ago.

People have two expectations: One is the expectation of privacy and to some extent anonymity, and the other expectation is effectiveness—they do want their search engine to know them. To what extent is there a balance to be struck?
I think there is a balance. And I think as far as I’m concerned, I’d like to keep it within the user and give users the choice.

While we’re talking on the subject of personalization, a colleague of mine said that search as you know it is falling to the wayside and changing dramatically as social networking comes into play—trending toward this MySpace-Facebook model where people look to their friends or their community as the take-off point. Do you see that as a bona fide trend? And, if so, does search become less important?
Search has always been about people. It’s not an abstract thing. It’s not a formula. It’s about getting people what they need. The art of ranking is one of taking lots of signals and putting them together. Signals from your friends are better signals, stronger signals. On the other hand, many searches are long-tail kinds of searches. If you’re looking for what movies to see tonight, your friend can probably give you the best information. If you’re looking for the address of the business, the Web as a whole can give you better information. If you’re looking for something obscure about anything, again the web can give you much better information. It depends on the type of search you do—and how to take all those signals and put them together.

Is it possible that a new type of search could emerge that’s based on social networking, or does that type of thing fold naturally into existing search?
I think it folds naturally. It’s just a matter of more signals.

So in other words, when I come to a Google in the future or maybe even now, the context of my social network could be folded into the search that I have there?
I can imagine if you give us permission to do that, and we find that that’s useful for some queries. The question is, what percentage of queries and what kind of queries? When should you use it and when should you not use it?

What do you think are the next coming challenges beyond what we’ve talked about?
I think one of them is a better understanding of queries and content, which is what we do all the time. If I were to point to one interesting feature that we launched recently—and unfortunately I can’t talk about coming features [since] we don’t pre-announce things—one feature we launched recently is something called CLIR. It stands for Cross Language Information Retrieval. Basically, what we do is take your query, translate it into another language, do the search in another language, translate the results back to your language so that gives you access to content in, I think, 12 different languages. So if you’re a user in Egypt, for example, and you only speak Arabic, you can write the query in Arabic, ask to translate it into English. When you then click on the results, it will translate the Web pages to Arabic for you—all of it done by Google translation. That opens the whole world to everybody. I see this as incredibly high potential.

To what extent does the change of mobile devices to more interactive devices and always-on devices as networks like LTE and WiMAX start making this seamless network possible?
So the main problem is not that they’re always on. The main problem is that the device itself is hard to read and hard to type or harder than a big screen. So you have to be more careful about extremely focused and getting the results very, very fast, which is something we know how to do. We’ve been working on specific techniques to make cell phone search better.

You have nothing to do with the advertising side, but is there a sort of “church and state” separation between the advertising side and what you do?
Yes, I told you we launched our 450 improvements. When we decide to launch something, we have a weekly meeting where all those things come together and we look at all the evaluations and we make decisions—revenues and any effects on ads do not come into those meetings. We don’t even know what the effects are. We make the decisions solely based on how good it is for search, how good it is for users. The ads are on a different part of the page, and the ad people, I assume, do the same kind of thing and try to improve the ads.

Popular Mechanics

YouTube’s filtering issues still not moot

Autor admin | 18.04.2008 | Category Media News, YouTube

LAS VEGAS–A year ago Wednesday, Google CEO Eric Schmidt delighted an audience of TV and radio broadcasters when he promised to roll out a system that would mean the end of piracy at YouTube.

"We are in the process of developing tools which are called ‘Claim Your Content,’" Schmidt said at the National Association of Broadcasters 2007 conference. "If people tell us this is a licensed copy, our computers will automatically detect that an illegal copy has been uploaded and then automatically delete it."

Schmidt went on to say YouTube was "close to turning this (system) on" and once that happened, copyright violation at the site "becomes a moot issue." But following through on that promise has proven a challenge.

Executives with two entertainment companies that provide YouTube with feedback on its Video Identification system said the company’s filtering technology has fared well at times but is nowhere near perfect and overall test results are "inconclusive." The sources, who requested anonymity because of the ongoing relationship with YouTube, added that managers at the video-sharing site continue to try and refine the system.

YouTube defended its efforts to protect copyright.

"Since launching in October, our Video Identification system has shown terrific results in its comprehensiveness, accuracy, and scalability," a YouTube spokeswoman said in an e-mail. "Over 100 partners from independent content creators to large media companies are currently using Video ID to easily manage their content. Many have found it to be a helpful tool in generating revenue and exposure for their content in the world’s largest online video community."

For a long time, numerous copyright owners accused YouTube and Google of profiting from piracy and deliberately dragging their feet in developing a way to cleanse the site. They argued that the availability of professionally created content–uploaded by users–is what draws people to YouTube and without that the site would lose much of its luster. YouTube has always denied the accusations.

Nonetheless, the controversy has damaged some of Google’s relationships in Hollywood.

Viacom, parent company of MTV and Paramount Pictures, filed a $1 billion lawsuit against Google last year. That case is expected to last years before being resolved and it could help decide what, if anything, a Web site’s responsibilities are when it comes to policing for copyright violations.

Copyright clips abound
Certainly at this point, it’s hard to see much change at YouTube since launching Video ID.

Available on the site are literally countless clips from feature films and TV shows produced by small production companies as well as the largest entertainment conglomerates–including Viacom.

Key into YouTube’s search field the names of the last five Academy Award winners in the best picture category and scenes from each will appear. Want to watch the first 10 minutes of the gangster flick, The Departed? They’re there. Someone else posted a series of 12 separate scenes from the film, presumably to get around YouTube’s 10-minute clip limit.

Fans of Clint Eastwood’s Million Dollar Baby can watch the "Mo cuishle" episode on YouTube as well as the coin toss scene from last year’s best picture winner No Country For Old Men.

Identifying video is not easy, YouTube execs have long said. About 10 hours of video is uploaded to the site every minute. In addition to policing an enormous volume of video, YouTube must first obtain high-quality copies so it can create a digital fingerprint of the film or show. Ideally, the automated system will recognize when someone uploads an unauthorized copy.

While copyright videos are still plentiful on YouTube, there are seemingly fewer complaints from Hollywood. The sources who are part of YouTube’s testing say the entertainment industry has shown a willingness to give YouTube time to improve filtering.

Some content owners may have also concluded that some degree of piracy is inevitable.

"We still see our content pop up on YouTube," CNN.com Executive Producer Sandy Malcolm told the Associated Press this week. "You deal with it. You try to work with them on rights and things, but I don’t think you can completely stop it. You just try to beat the tide and try to get your content out as fast as you can."

Google execs continue to say they respect copyright and are working to protect it. Schmidt said protecting copyright was in Google’s best interest.

"We are critically dependent upon the production of copyright content," Schmidt told the NAB audience a year ago. "Literally, people come to Google to get to somewhere where there is something of value. It’s very important that we not violate copyright."

News.com