search engine

Sep 2 2008   5:21PM GMT

Chrome: A shiny Web browser from Google may just be the next global platform for running Web applications

Posted by: Alexander Howard
Google, Microsoft, business, Mobile, applications, Web 2.0, operating systems, Apple, software, Technology, Web services, AJAX, video, YouTube, Internet, search engine, innovation, cool, social bookmarking, social publishing, reviews, Silicon Valley, downloads, collaboration, freeware, advertising, desktop, Office, geotargeting, social networking, blogging, buzz, communications, Web applications, buzzword, cloud computing, the cloud, Windows, SEM, demonstration, SEO

Techies and geeks returned from one last weekend of sun, sand and summer to find news of a disruptive change sweeping the online business world. Meet Chrome, Google’s new Web browser.

News of the announcement was leaked yesterday when Philipp Lenssen, an avid blogger of all-things-Google, received the comic book Google put together for the release and posted it, along with his first impressions. My director, ahead of the curve as usual , picked up on it right away and added it to WhatIs.com’s Buzzword Alert.

Google has since put up a high resolution of the Google Chrome comic book. I highly recommend going over and reading through the comic. Google put considerable time into clearly explaining the challenges faced by the designers of modern Web browsers with respect to memory bloat, rendering engines, Javascript threading errors and much more.

Since Lenssen broke the news, the tech blogosphere has of course been awash with reviews, opinions and speculation about what, exactly, Chrome will mean. Walt Mossberg posted a comprehensive review of Chrome in the Wall Street Journal, including speed and feature comparisons with Safari, IE 8 and Firefox. Rafe Needleman liveblogged the press conference introducing Chrome over at Webware. John Furrier colorfully blogged that the search wars just turned into the operating system wars. That’s true — except (as he notes) that Chrome goes far beyond search. SEO/SEM hounds and search engine watchers, however, will find Danny Sullivan’s thorough evaluation of Chrome’s search functionality quite useful.

Following below is own my two cents, both with respect to the browser itself and the significance of its introduction. First, however, I’ll let the video embedded below provide a quick introduction:

Obviously, Chrome has a lean, clean interface. This is Google, after all. Menus, dropdowns, extra bars and dialogs are largely stripped away. So what’s left?

The Web pages themselves. What a concept! I downloaded and installed the browser this afternoon without a hitch, imported my bookmarks and search history from Firefox and was off to the races. Chrome is quite speedy.

The address bar has been merged with the search field you’d see on the right in IE or FF. Firefox 3 includes a predictive search in this field already, so this isn’t ground breaking, but it is a clear recognitiion that search has become the default navigation method for most Web users. Enter your desired search terms and away you go.

Google is calling the new address field the “Omnibox,” a nod to its ability to incorporate “everything” you might need to explore. The Omnibox’s utility is another sample of Google’s secret sauce, in this case combining a record of your search and browsing history with Google’s own PageRank for given terms. The Omnibox is eerily good. With only a little use, it could predict precisely which page I was looking for after only a few characters were entered.

Chrome also features tabbed browsing, a key improvement introduced by iBrowse in ‘99 and then popularized by Opera in 2000. Once Mozilla included it in Firefox, the feature took off and is now a default feature in Internet Explorer and Safari. Chrome expands the tabbed interface in a number of innovative ways, including grouping related tabs and designing each tab so that it acts as an independent browser. Bookmarks, the Omnibox, menubar icons and menus are all inside of the browser, which again frees up more space for displaying rendering Web pages.

The pop-up blocker and phishing or malware alerts also included in Chrome may not be innovative at this point but they’re certainly effective and useful. The private browsing mode, aptly called “Incognito.” (This clever feature name was perhaps made in hopes that it will avoid the “Porn Mode” moniker that has dogged a similar feature of IE 8, InPrivate.)

There’s another key development: Chrome may not be the fastest Web browser currently available but Google hopes that it will be the most stable for pages loaded with Javascript. In a Web 2.0 world ruled by AJAX, that’s no small thing. And anyone that’s used one of Google’s many online applications knows that a stable, reliable environment for this kind of scripting is crucial.

This hints at perhaps the most important detail of all, and one that I tipped my hat to in the title of this post. Microsoft made an early bid for Internet dominance in the infamous browser wars of the 1990s by including Internet Explorer in each copy of Windows. Despite the Justice Department’s successful antitrust suit, IE continues to have upwards of 75% of the world’s browser share. Firefox has made inroads on this market share, to be sure, and the most recent version of Mozilla’s browser has been the best option around for speed, privacy, safety and usability since its introduction this summer, following close upon the success of Firefox 2.

Now it’s Google’s turn.

Google’s introduction of its own browser has the potential to upset the market in a way that no other company can, simply because of Google’s ability to promote the download and use through its various Web properties. As Google’s various Web applications and cloud computing architecture continue to mature, the Web itself can develop into an operating system. If this sounds familiar, that’s because Sun’s vision of network computing in the 90s using Java popularized such a concept long ago. Vastly improved broadband connectivity, viable Web-based apps and an Internet technology giant flush with revenue from the world’s best advertising platform change the dynamic a bit, of course. Google built its own Javascript engine to improve performance and, crucially, integrated Google Gears with Chrome to allow true offline access to its various Web applications. That adds up to something that distinctly resembles a fully-fledged desktop operating system and productivity suite.

While it’s true that consumer and enterprises haven’t been making a run on thin clients running on Linux quite yet, the potential to further erode Microsoft’s dominance of the operating and desktop productivity software markets is embedded within Chrome. I’m far from the only writer prognosticating on this count, of course. Michael Arrington thinks Chrome is Google’s Windows Killer. As Michael points out, this clears the way for “millions of web devices, even desktop web devices, in the coming years that completely strip out the Windows layer and use the browser as the only operating system the user needs.” Given that both the enterprise and consumer markets haven’t exactly been hot about Vista, I suspect Microsoft may be somewhat concerned about this development. Henry Blodgett over at the Silicon Valley Insider sees the development from precisely this angle, blogging that Google has launched a cloud operating system and called it a ‘browser.’

Who else should be concerned? Maybe Mozilla, though judging by this interview with its CEO, they’re putting a good face on the development for the moment. What’s next? Harry McCracken asked 10 questions about Google Chrome over at Technologizer that address Mozilla’s future relationship (and relevance). Jeremiah Owyang has added a few more questions in thinking about what Chrome could mean long term. Both ask for response and speculation in their comment sections, so have at ‘em.

Microsoft hasn’t been standing still, of course. They’ve been chasing search revenue for years, as evidenced by the failed Yahoo! acquisition. As the folks over at the Google Subnet blog at NetworkWorld point out, IE 8’s InPrivate mode thwarts Google’s targeted advertising. Unless the world upgrades to IE 8 and begins to browse InPrivate en masse, however, I’m guessing that GOOG’s 3+ billion of revenue per quarter is gonna be safe for the moment.

That’s especially true when you consider another critical element of Chrome: its future relevance to mobile search. Google’s Eric Schmidt has been quite bullish in this area, estimating that mobile search revenue will likely surpass desktop search in the not-so-distant future. The iPhone has shown what a data connection and full Web browser can do to mobile search (Try 50 times as many searches originating from iPhones vs. a normal cellphone). Here’s a prediction you can take to the bank: Just as the iPhone features a stripped down version of Safari, Google’s Android OS will have a similarly light version of Chrome optimized for a mobile device and poised to fully take advantage of the possibilities for geotargeted advertising based upon a user’s demographics, Web history and location.

Louis Gray is dead-on when he points out that Web browsers are now about the hooks. Apple’s Safari will be increasingly optimized for the iPhone and working with the private cloud that is MobileMe. Microsoft has built IE to be integrated with Windows and Office, though because of the bundling issues presented by antitrust has always had to walk a fine line. Flock, the social media-optimized version of Firefox, carves out a niche because of its tie-ins with the various networks and services. Chrome is no different, as I pointed out above. If you are already a power user of Gmail, gDocs, gTalk, gReader or g-Anything, Chrome may make more sense. Chrome is, I should note, only available for Windows Vista or XP at the moment. Guess they figure Safari will do the trick for a Webkit-based browser for Mac users and that the Linux crowd will be satisfied with Firefox and Opera for the moment.

To poorly paraphrase Lando Calrissian, Google’s Chrome is likely to allow all mobile users to truly surf with them amongst the clouds.

Aug 1 2008   10:29AM GMT

Video: Jimmy Wales on Google’s Knol

Posted by: Alexander Howard
Google, business, Web 2.0, video, YouTube, Internet, search engine, search, innovation, commentary, learning, free, academics, public domain, social publishing, Silicon Valley, collaboration, wiki, conversation, community, tool, blogging, buzz, communications, Web applications

Jimmy Wales, co-founder of Wikipedia, talked to WNYC’s Brian Lehrer about Google Knol, a new competitor to the world’s largest online encyclopedia.

Jun 13 2008   1:45PM GMT

What is Unity? Lockheed-Martin’s implementation of a social computing platform wows Enterprise 2.0 conferees.

Posted by: Alexander Howard
Google, Microsoft, Networking, business, applications, Web 2.0, enterprise, software, Technology, search engine, search, innovation, feeds, portal, social bookmarking, social publishing, interesting, invention, collaboration, wiki, community, tool, howto, information, trend, social networking, CMS, blogging, communications, Web applications, buzzword, software development, conference, enterprise 2.0

One of the unexpected hits of the Enterprise 2.0 Conference this past week was a presentation by Lockheed-Martin on Unity, its social computing platform. One of the world’s largest defense contractors would seem an unlikely candidate for early adoption of enterprise 2.0 technologies, or at least that was the impression when the session kicked off. By the end of the hour, audience members were asking “Where can I buy it?”

[Image credit: TechLuver.com]

Shawn Dahlen and Christopher Keohan talked at length about what they’d learned over the course of eighteen months developing the platform, kicking off their presentation by noting that there was a compelling need in government sector to collaborate through social media. Chris noted that embracing social computing at Lockheed Martin a major component of recruiting talented Generation Y IT workers, the so-called “millenials,” as showing the company’s prowess in the adoption of cutting edge tools was a key differentiator.

Before Unity was implemented, the state of collaboration at their enterprise should be quite familiar to most corporate workers : email, meetings and office docs like Powerpoint presentations emailed around as attachments. “Project Unity” was conceived as a way of applying Web2.0 technologies for “mission success.” To that end, the team resolved to provide a user experience employees would love, address “what was in it for them” and balance the need to share vs the need to know — crucial in a defense contractor. Unity’s designers wanted to foster a social computing ecosystem around a standardized platform, integrating blogs, wikis and other documents into their current platform. Over time, they added discussion forums, a social bookmarking tool called “uBookmark” and weekly activity reporting to capture usage and adoption patterns. They included a suggestion tool to solicit community insights on the project as it rolled out and created an internal homepage to aggregate popular content. Unity’s internal team of developers also made a priority of maintaining a cohesive user experience and to ensuring that all information could be both feed-enabled and integrated.

How did they pull it off? By integrating Google enterprise search appliance (GSA) , Microsoft’s Windows Sharepoint Services (WSS) and Newsgator’s Enterprise Server. Take a look at this demonstration of Social Sites 2.0 to get a feel for what this looks like. They Unity development team took a close look at how to use social computing tools in an everyday business context and took the time to understand how they would integrate and evolve from the existing email/Powerpoint/meeting model.

The crucial question, asked over and over again this week, was addressed head-on by Unity’s designers: “What is the value of social networking in the enterprise?”

Their answer was, in the end, simple: Being able to watch what other people are doing, easily, and then being able to search it and ask questions raises productivity and leads to improved collaboration and knowledge exchange. Instead of tracking what your friends are doing on, say, Facebook with a “friend feed,” an enterprise derives value from tracking an activity stream of interconnected colleagues. At any point, a worker can see what others are working on, access shared documents and ask questions on shared virtual workspaces or directly to the relevant decision maker or technologist.

Lockheed-Martin built the basic Unity platform in 07 and then ran a beta pilot of it over the course of the year with 40 engineers building, testing and experimenting with the release. After the initial release, it took just six months for a second iteration that addressed both information security and legal issues.

A crucial question that they were asked to account for again and again will be familiar to CIOs: How did they quantify the return on investment (ROI) for the dedication of internal resources and purchase of software? Each time, the traditional productivity savings of a user finding information was a factor. What really sold them, however, was the soft case of customers interested in their social computing initiative. Unity helped in Lockheed-Martin’s bidding process, especially proposals that involved knowledge managememt.

As the project rolled out, a crucial component was the in development and distribution of a “collaboration playbook.” New standards for playbook and best practices were laid out in its pages. For instance, as a team member, you should ask questions on a group page, not wander over to ask or send a broadcast email; this helps to capture questions and answers for everyone. Adding to documentation whenever possible was crucial, along with teaching people the power of linking and understanding which communication type made sense for different business cases: blog posts, wikis, email, virtual conferences or in-person meetings. In the end, the Unity team created the playbook as much for themselves as they worked as for the company as a whole, “eating their own dogfood.” They used a project management office (PMO) blog to keep colleagues up to date about what the dev team was doing.

One of their other key discoveries was that pervasive enterprise search is key to keeping documents both relevant and accessible.

What’s next for the team? Adding filters to content that depend upon the clearance of those accessing it. In highly classified work, user-assignable taxonomies are crucial for opening up content for collaboration while maintaining information security. Also in the works are adding recommended content, similar to the Digg-model of social news, employee profiles, export control filters and network-based search.

If you’re looking for a great case study for enterprise 2.0 adoption, look up Unity.

Apr 24 2008   8:33AM GMT

Video: Sergey Brin speaks about search, Google, and life at UC Berkeley

Posted by: Alexander Howard
Google, business, video, YouTube, Internet, search engine, search, learning, academics, entrepeneurship, forum, advertising, information, SEM

Google’s distributed search model is at the foundation of the Internet giant’s current dominance in search. In the video below, one of Google’s founders, Sergey Brin, speaks at length about his company. You’ll need to turn the volume up on this one.

It should come as no surprise, therefore, that Google is rejecting claims of patent infringement made in a lawsuit brought by the Jarg Corporation, a Massachusetts-based technology company.

Apr 16 2008   9:58AM GMT

Video: Matt Cutts debunks 5 SEO myths

Posted by: Alexander Howard
Google, business, IT marketing, video, YouTube, Internet, search engine, search, entrepeneurship, resource, Development, community, howto, blogging, fundamentals, SEM, SEO

Google’s Matt Cutts has long since become the blogosphere’s “go to guy” for information on SEO and webmaster guidelines. WhatIs.com’s Word of the Day today, phantom page, has a link to his commentary on detecting undetectable webspam, for instance. I like the term “webspam,” incidentally, as it neatly describes spamming the entire web, as opposed to individual inboxes or SMS gateways. The video below features Matt Cutts debunking five different SEO myths. The video is from 2006 but is still quite relevant.

While it’s true that there are many other SEO bloggers and a burgeoning industry in search engine marketing, none are quite so well placed within the search engine giants nor so willing to share best practices and commentary. Thanks for your contributions to the Web community, Matt.

Mar 25 2008   12:01PM GMT

Video: John Lisbin on Domain Parking at SES NY 2008

Posted by: Alexander Howard
business, Technology, video, YouTube, search engine, search, event, advertising, ICANN, SEM

John Lisbin, the chief strategist at Point It! discusses domain parking and domain tasting with Greg Jarboe of SEO-PR at SES 2008. In the video below, Lisbin addresses the issues advertisers have with domains in relation to searches performed using Web browser toolbars.

Aug 20 2007   6:06PM GMT

WikiScanner: How Googlebombing Virgil led to radical transparency for Wikipedia editing everywhere

Posted by: Alexander Howard
applications, programming, Internet, search engine, innovation, useful, cool, hacks, free, academics, interesting, Silicon Valley, collaboration, wiki, community, tracking, tool, humor, Web applications, controversy

Sometime this past summer, future CalTech grad student (and self-styled”disruptive technologist”) Virgil Griffith decided that he wanted to see if he could elevate his personal home page to the top of search results for “Virgil.”

While that dream may or may not come true, there’s no doubt that the “quotes” should now be removed from disruptive technologist, (along with my self-styling). Virgil hasn’t written an epic quite yet but he’s clearly provided a useful tool to journalists and inquisitive netizens alike. Meet WikiScanner.

In two weeks, Griffith created a Web site that matches the IP addresses attached to edits of Wikipedia pages with the IP addresses listed in the publicly accessible whois database for companies and media organizations worldwide.

Using a simple, minimally designed interface consisting of a form, text, dropdowns and hyperlinks, Griffith’s site allows users to first determine what the IP address of an organization is and then plug it into to discover what edits have been made from that IP range.

By providing the means to cross reference who is editing what, Virgil’s code nearly approaches poetry, at least judging by Jimmy Wales, who told the AP that “It is fabulous and I strongly support it.”

It’s not quite fair to say that the effort is all about the Virgil Google bomb, either. According to this Wired article, Griffiths was inspired by the discovery that many Congressional offices were editing their own entries. Wondering what corporations were doing the same thing, he created WikiScanner.

As Griffiths notes wryly in his WikiScanner FAQ, the reaction to the tool’s discoveries has been more or less as expected. The CIA and FBI, Disney, Diebold, Exxon and a rapidly expanding universe of other entitities, large and small, have been editing away. Take a look at Wired’s “list of salacious edits” to see how far the count of “minor public relations disasters” that WikiScanner has enabled — and be aware that any edits that you make to Wikipedia may not be as anonymous as you might think.

Jun 11 2007   12:38PM GMT

Google search field hacks

Posted by: Alexander Howard
Google, Internet, search engine, search, useful, cool, hacks, education, free, listings, tool, howto, ebooks, cheatsheet

I’ll admit it: I’m a frequent Googler. I Google from the office, when I need to research new terms for WhatIs. I Google from home, when I need information about events, people, tide charts or news. These days, I’m Googling from the car and train as well, enabled by the handheld attached to my belt on an o-so-slow GPRS connection, using my MDA as a sort of primitive Hitchhiker’s Guide to the Galaxy. That’s rather useful, of course, when I need to determine the number from which the meaning of life, the universe, and everything could be derived.

Some time ago, however, I started to be considerably smarter about my Googling, as I realized that with just a bit of syntax ahead of my search terms, I could make much better use of the search field. These additional words are called “operators,” for my semantically-obsessed fellow travelers. They make life easier. Following is my short list of favorite Google search field hacks. If you have some of your own that I missed, please let me know in the comments so that we all can become more efficient Googlers too.

Google Phonebook: I particularly love this one. I stopped using the white pages because of this very feature. Just type in “phonebook: ” and then a name, comma, zipcode. I’m a “victim” of fixed-mobile substititution, so there’s no danger of revealing my digits to the world here in an example. If I did have a landline, however, you’d find it that way. It’s also possible to reverse engineer the lookup by entering a phone number, revealing the attached adress. For those a bit freaked out by this ability, it’s worth noting that you can request that your name be removed ,

Google Weatherman: While I look to NOAA.gov for all-things-meteorogical, if I just want to know whether to grab a jacket, sweater or shorts, typing in weather: zipcode is perfectly effective. For instance, here’s the weather in lovely Needham Heights, Massachussetts today.

Google Movies: I adore this feature. Just enter “movie: zipcode” to get a list of theaters and showtimes near you, with links to showtimes with available tickets and reviews. This stripped down, entirely textual results page is especially useful and usable when I’m mobile.

Google Dictionary: This is spectacularly relevant to my work, given that I write definitions for WhatIs.com. If you’d like to see all of the entries for a term, simply type “define: term” and you’ll be presented with a list of hyperlinks and short summaries. Try define: blog for a comparison of takes on that hotly contested term, for instance.

Google Site Search: While searching the entire Web is undeniably useful, sometimes you just want to look through one Web site, like, say, WhatIs.com. Just type in site: domain name search term (like site: WhatIs.com geek) and you’re off and running.

Google University: Just as you can restrict search to a specific site, you can also focus on certain domain names, like .edu. If you’re a developer, for instance, you could enter [ruby tutorial site:.edu] Of course, these days you can also just use Google Scholar.

Google for Media: Looking for ebooks on Java? Paste the following syntax into your search field:

-inurl: (htm|html|php) intitle:”index of” +”last modified” +”parent directory” +description +size +(.pdf) “Java”

If you replace .pdf with other extensions and Java with a different keyword, you can also find all kinds of other media out there too, though it’s worth noting that relevant intellectual property laws still apply to your actions.

As I wrote initially, this is only a short list the tweaks that I actually use with any frequency. For more information, see Google Blogoscoped’s post about using special syntax or Google’s list of operators, including a printable search cheatsheet.

May 11 2007   1:42PM GMT

Mechanical Turk, Powerset and Enterprise Search

Posted by: Margaret Rouse
Web services, search engine

You have to hand it to Jeff Bezos. He’s probably one of the most enthusiastic people on the face of the planet. When he speaks at a conference, you can almost feel the energy in the air. Jeff Bezos is one of the good guys. He has his pulse on big business and the lowly end user. He’s worried about planet Earth and all those Amazon boxes sitting in your basement. He’s watching out for businesses of all sizes, making sure they have access to the storage and servers they need, even putting them in a cloud for easy access.

What else do we find in a cloud? Angels.

An angel is not only a benevolent spirtual being who watches out for people, an angel is financial lingo for someone incredibly rich who’s willing to invest significant dollars in a startup that might go nowhere. Jeff Bezos is a very sought-after angel. (Ironically, one of the ventures he’s gotten the most press about is a scheme for putting tourists in space.)

I saw Bezos give a keynote at MIT last Fall and he devoted a big part of his talk to Amazon Mechanical Turk. He sure was enthusiastic, but I left the conference thinking “What is he really up to?”

Bezos describes Amazon Mechanical Turk as an “artificial artificial intelligence program.” Mechanical Turk gets its name from an 18th century chess-playing marvel that toured Europe and played against luminaries of the day, including Napoleon and Ben Franklin. Before each performance, the audience was allowed to inspect Mechanical Turk to view and marvel at its many functioning gears and levers — but like all good magic, the Turk was an illusion. A real person manipulated the Turk. A real person played the chess game, hiding in a secret compartment behind the gears and levers.

The official pitch is that Amazon Mechanical Turk is a web service API that allowes developers to integrate human intelligence into remote procedure calls. The application sends a request, a network of humans getting paid by the task fill the request, the service returns a result. Pretty simple.

Here’s what Amazon has to say:

For software developers, the Amazon Mechanical Turk web service solves the problem of building applications that until now have not worked well because they lack human intelligence. Humans are much more effective than computers at solving some types of problems, like finding specific objects in pictures, evaluating beauty, or translating text. The Amazon Mechanical Turk web service gives developers a programmable interface to a network of humans to solve these kinds of problems and incorporate this human intelligence into their applications.

For businesses and entrepreneurs who want tasks completed, the Amazon Mechanical Turk web service solves the problem of getting work done in a cost-effective manner by people who have the skill to do the work. The service provides access to a vast network of human intelligence with the efficiencies and cost-effectiveness of computers. Oftentimes, the cost of establishing a network of skilled people to do the work outweighs the value of completing it. By turning the fixed costs into variable costs that scale with business needs, the Amazon Mechanical Turk web service eliminates this barrier and allows work to be completed that before was not economical.

For people who want to earn money in their spare time, the Amazon Mechanical Turk web site solves the problem of finding work that they can do wherever and whenever they want.

Forgive me for doubting Jeff Bezos’ altruism. Why was he building and promoting Mechanical Turk?

In his keynote, Bezos was vague. His best example of how Turk was being used was when he described a company that used Turk’s human resources for transcribing podcasts. Ehhhhhh….don’t we have software that can do that?

When I got home, I signed up to work for Mechanical Turk and sure enough — there was a task you could sign up for were you transcribed podcasts. There was also a task where a guy was looking for people to write trivia questions, and a guy who was looking for people to write blog posts. After six hours of work on a Saturday night, I made exactly $3.20.

Ok…WHAT IS THIS THING REALLY FOR?

I tried to get a Turk developer to agree to be interviewed on video about Mechanical Turk, but Amazon’s PR rep politely said “no.”

Then yesterday, I got my first e-mail from Mechanical Turk. There was a list of tasks (called HITS) and in that list was the answer to my question. I think.

“Powerset — Evaluate the relevance of search results for Powerset. To complete these HITs, indicate how relevant is a set of results for a given search. “

Did Bezos build Mechanical Turk for Powerset, the secretive search startup backed by other A-list angel investors? Powerset is supposed to use natural language, the holy grail for enterprise search. And Bezos is rumored to be backing Powerset.

All the puzzle pieces fit. In December, Amazon shelled out the entire second round of funding for Wikia (amount undisclosed). Same month, Wikipedia founder Jimmy Wales toured the media circuit talking up a new project for Wikia — to build a community-driven “Google-killer” search engine.

I just can’t see them wasting their time, energy and money on another consumer search engine. But I could understand all energy being devoted towards enteprise search. And where would the community that Wales has been talking about come from? Could a big part of it be Mechanical Turk?

Is the secret ingredient of Powerset — and the key to enterprise search — going to be artificial artificial intelligence?

---

Thanks to Sophie G. for sending me this article by Don Dodger. Also check out “Enterprise Search — Tough Stuff!” and Greg Linden’s blog.

Apr 26 2007   12:53PM GMT

Ms. Dewey: Personalized search makes Live sexier — but the results are the same.

Posted by: Alexander Howard
Technology, fun, video, search engine, cool, buzz

Have you been over to search with Ms. Dewey yet? Ms. Dewey is a viral marketing campaign that Microsoft launched in October, 2006. The campaign centers around msdewey.com, which features actress Janina Gavankar playing a character of the same name. 

The Web site features an experimental Flash-based interface that houses Windows Live Search, Microsoft’s search engine. Unlike the more minimalist design principles favored by Google or Ask’s search interface, Ms. Dewey uses a dark city skyline with Gavankar placed front and center. She prompts the visitor to search, commenting on keywords entered. When idle, she’ll find other things to do, often using props from behind the desk or making observations that can only be described as snarky.

The site was developed by graphic designers McCann-Erickson and developer EVB, using over 600 video clips. While the results may be no better than those generated on the standard Live Search page, there’s no question that Ms. Dewey is both easier on the eyes and more entertaining than Jeeves ever was.