Google Insight gains ground...

 

 The new Google Insight tools are gaining a good deal of poularity;
(Via ZDNet--)
"Google Insights for Search takes daily search data from the countries they operate in, and analyzes it to graph “interest over time” by seeing how many searches have been done for your query compared to the number of all searches done on Google over time. The result page shows all the top searches for what you are looking at — and it also gives you a nice map of the world showing you how “hot” your query is across the world."

 

 It's basically Google Trends on steroids, and shows you a ton of data on any search you try. An SEO wizard's dream, basically. It's described as:

With Google Insights for Search, you can compare search volume patterns across specific regions, categories, and time frames.

 Basically - you put in keywords and it renders attractive charting..

 One useful query to try is to search for your favorite website - like "gaiaonline.com" and specifically target it towards the US. It shows you a neat state-by-state breakdown of who is doing those searches.

 Although unscientific, it tells you a bit about the location of the people who use the website, since logically the folks in states where the product is popular would tend to search for it quite a bit. Interestingly enough, Nevada, Hawaii, Oregon, and Washington are some of the top searchers for Gaia Online. Fascinating.

 A great application of this is compare which websites are being used mostly by early adopters against now-mainstream websites. If so inclined, a user could just go and search a whole bunch of domains and see what kinds of graphs are produced.

 Here's a link to the actual Insight's page.

 

Microsoft proposes to outdo Google's "PageRank"

 

 Microsoft researchers and academic collaborators, this week, detailed an idea called BrowseRank that seeks to bring more of a "human factor" in comparison to the PageRank assessment process presently being used by Google.

 Essentially, the researchers tested out a system that replaces PageRanks' link graph - a mathematical model of the hyperlinked connections of the Internet - with what they call a user browsing graph that ranks Web pages by people's behavior.

 "The more visits of the page made by the users and the longer time periods spent by the users on the page, the more likely the page is important. We can leverage hundreds of millions of users' implicit voting on page importance.." said Microsoft researchers.

 The Microsoft researchers contend that PageRank has a myriad of problems.

For one thing, people can 'game' the PageRank system using Web sites called link farms, that inflate a website's "Google Importance Score" (which includes the number of backlinks).

Another PageRank issue is that the indexing process doesn't take into account the time a user spends on a particular site

 User behavior, monitored in anonymous form by Web servers and Web browser plug-ins, can be better, the authors argue. "Experimental results show that BrowseRank can achieve better performance than existing methods, including PageRank...in important page finding, spam page fighting, and relevance ranking".

 Microsoft researchers gathered their data from "an extremely large group of users under legal agreements".

  An excellent white paper detailing "BrowseRank" can be read here.

 

Aha! Alexa's not to be trusted, after all...

 

 In a very trustable blog today, Search Engine Journal, an article appeared titled "Alexa Bias Exposed by Top Google Engineers".

 

 Via Google Engineer Peter Norvig:
  "The data would be good if it truly represented a random sample of internet users, but in fact it only represents those who have installed the Alexa toolbar, and that sample is not random."

  Hmmm.. So the rumors and claims posted all over the Web for several
  years have proven reliable!

 

Google "opens the hood" on it's ranking technology

 

 Amit Singhal, Google Fellow in charge of the ranking team at Google has posted an very revealing blog article about how Google's ranking technology.

 

 As you may know, Google filed a series of "Phrase-based Indexing & Ranking" patents that were finally fully implemented earlier this year.

  Here are the three most relevant:

Phrases are identified:
Phrase identification in an information retrieval system
Application filed: July, 2004 / Published: January, 2006>/li>

Documents are indexed according to their included phrases:
Phrase Based Indexing in an Information Retrieval System
Application filed: July, 2004 / Published: January, 2006

Users search to find sites relevant to what they're looking for:
Phrase-based searching in an information retrieval system
Application filed: July, 2004 / Published: February, 2006

 Via the blog article:
   "Search in the last decade has moved from
   give me what I said to give me what I want.."
 and

Understanding queries: It is critical that we understand what our users are looking for (beyond just the few words in their query). We have made several notable advances in this area including a best-in-class spelling suggestion system, an advanced synonyms system, and a very strong concept analysis system.

 If you want a practically "once in a lifetime" glance into SEO for Google;
  The article is well worth checking out!

 

Cool new Geo-Coded news, from Google!

 

 Google and the New York Times have teamed up to offer an extremely cool new service. It's called "Google LatLong", and it allows users to access breaking news from specific Lattitude_&_Longitude locations - using Google Earth.

 

 Via the LatLong site, by Wei Luo, Tech Lead Manager, Google Earth:
"In collaboration with The New York Times, we've come up with a solution: The New York Times offers geo-coded news, and Google Earth offers the platform for reading that news in a 3D browser. This is the first time we've endeavored to show news updated in real time, and we're very excited to work with this first-class publication to bring you the latest and greatest news."
and
"To experience this new way of getting your daily dose of news, launch the latest version of Google Earth and make sure the "Geographic Web" folder is turned on. Click on a New York Times placemark and you will see the latest news and features pertaining to that geographic region. Want to see more than just headlines? Click on the "Show this layer" button at the top of the preview bubble and you'll get a list of news articles dating back one month"

 This is definitely worth a look.
  To read the post on the LatLong Blog, click here

Google's new home page Shopping Tab

 

 Google recently added a shopping tab to its homepage, which directly links to its on-line product catalog called Product Search.

 

 Through Google Shopping, shoppers get multiple listings for any product search, and can then purchase directly on Google, or go to seller's e-Commerce store. Submitting to Google Product Search is free.

 In addition, FastCommerce, a small business enterprise e-commerce provider, recently announced the successful launch of its automated submission service to Google Shopping. By grabbing the free version & creating a catalog, an eCommerce site can start selling instantly (a terrific help to small e-commerce start-ups). FastCommerce also submits the on-line catalog to Google Product Search.

 

Google's in the spook business!

 

 According to SFGate, the San Francisco Chronicle's website:
"Spy agencies are using Google equipment as the backbone of Intellipedia, a network aimed at helping agents share intelligence. Rather than hoarding information, spies and analysts are being encouraged to post what they learn on a secure online forum where colleagues can read it and add comments..".
and
"Agents can log in, depending on their clearance, to Intellipedia's three tiers of service: top secret, secret and sensitive but unclassified. So far, 37,000 users have established accounts on the network, which contain 35,000 articles encompassing 200,000 pages"

 ("In addition to the intelligence agencies, Google's government customers include the National Oceanographic and Atmospheric Administration, the U.S. Coast Guard, the National Highway Traffic Safety Administration, the state of Alabama and Washington, D.C.")

 "We are a very small group, and even a lot of people in the federal government don't know that we exist," said Mike Bradshaw, who leads Google's federal government sales team and its 18 employees".

 "Google is one of many technology vendors vying for government contracts -
A single deal can be sizable, such as the one Google made with the National Security Agency, which paid more than $2 million for four search appliances plus a support agreement, according to a contract obtained through a Freedom of Information Act request".

  Cool - Google's sort of in the "spyware" biz .. !!

 

Pele-Phone adds Google to it's 3G Handsets

 

 Bezeq Telecommunication Co. Ltd., a wholly-owned subsidiary of Israel based Pele-Phone Communications Ltd., has installed the Google search engine on its mobile portal. Pele-Phone 3G subscribers will be able to access the Google search engine when entering the web portal from their cell phones.

 Recently, most of the major mobile device manufacturers and cellular operators have a Google interface as built-in handset software. This increases mobile Internet use, which is a profitable feature for wireless carriers.

 

Google's in the city guide search business!

 

 Google's in the city search business! The new site, http://www.google.com/m/lcb, was discovered by ZDNet's Garett Rogers and then covered on Adam Broitman's blog.


 It's a lot like Yahoo Mobile Search, with a link-populated directory about certain topics, & is able to function as a "Your Location Specific" search in the mobile device's local areas. Even without mobile Internet access, you can try out Google LCB on the Web at the above address.

 The idea behind this cool new Google mobile search is to make it as easy as possible to drill down by browsing without fumbling on a keypad. The "LCB" site delivers the results page without the query vis-a-vis "seamless searching".

 Every link in LCB jumps to a new results page where users can drill down until they find a listing, and a Google map to get there.. of course.

Now..
How cool is that!

 If the connection speed is fast enough (hmmm) browsing LCB can be much faster than typing.
 So! Google is now also a city guide, and will of course immediately dominate (and probably buy) Zagat, Citysearch, and others.

 

Multilingual video search engine from Japan: Fooooo

 

 A Japanese-based video search website titled Fooooo is out and about, and it's pretty cool. Mr. Daisuke Tanaka, director at Bank of Innovation (the chief investor backing Fooooo), gave the Asiajin blog some exclusive insight recently.



Via the Asiajin blog;
 "At the moment, Fooooo is able to search a whopping 130 million videos from nearly 100 sources from the most different backgrounds. User can browse the site through accessing 18 different categories, they can view the most popular videos of the moment as suggested by Fooooo and -of course- can search via keyword input."

The entire exclusive article can be read here.

 

Quaero gets new cash, But it "isn't a search engine"

 

 In an article appearing in International Herald Tribune today, the alleged "Google rival" French-based Quaero is receiving "a total of €298 million over the five years of the project".

 But - apparently Quaero is changing it's mind:
   Via the article:
"But Quaero participants say that their project has been misunderstood by the public, that it is not the recipient of unfair subsidies, and that it is not even a search engine. Instead, the 23 companies and organizations involved will collaborate on researching and creating prototypes for multimedia and multilingual search tools. Whatever results would be licensed to any company interested."
and
"Exalead, which specializes in business search engine technology, is one of the few French-born technology companies to compete with Google on any level, and is one of the Quaero project leaders"

 But it's not really a search engine..
                                                                     

 

PageRank; Really the "be all, end all" factor?

 

 A great deal of importance is placed upon PageRank value these days, and a great deal of this emphasis is slightly off-kilter.

 Take a look at some searches:

A search for, say "wireless speech recognition headsets" shows us the Number 1 site, competing with 915,000 other sites 'about' this phrase has a Page Rank of 2/10; yet the Number 2 site has a Page Rank of 4/10.

Another search for the phrase "email address for Alltel cell phones" shows us the Number 1 site, competing against 410,000 other sites has a Page Rank of 4/10, while the Number 2 ranked website has a Page Rank of 5/10.

 What's missing from the "PageRank is most important" philosophy is really kind of selling Google's incomparable indexing more than little short.

 Here's yet another example:
A search for "script for scheduling defrag task" is very interesting example.
The Number 2 website indexed for this phrase is completely unranked;
and the Number 4 website is ranked at 4/10.

  What's the moral of the story?
Content, relevant content remains almost equally important in an awfult lot of how Google indexes the billions of documents it stores in it's document barrels.

 Google isn't fooled in many instances by huge amounts of inbound links that have little relative value.. links that are/were created by virtue of their immense popularity (like news websites), to Web documents that just touch on a subject in passing - do not make those documents more important than pages in a website that is comprehensively about that same subject.

 A good deal of this seeming anomaly hinges on Google's "word stemming"  indexing_&_ranking algorithm;
 But that topic is worth another article in and of itself!
     

 

Yahoo! moves toward Semantic Web searching

 

 Yahoo has announced it is adopting several of the key standards of the "semantic web'.

 The Semantic Web is a mesh of information linked up in such a way as to be easily processable by machines, on a global scale. It takes the metadata concept further; it involves publishing the data in a new language, Resource Description Framework (RDF), which can render an almost-human understanding and accordant database sorting of individual document components and also their individual "themes" or aggregated information.

 Says Amit Kumar, Engineering Manager for Yahoo! Search:
"By supporting semantic web standards, Yahoo! Search and site owners can bring a far richer and more useful search experience to consumers..
We believe that our open approach will let each of these formats evolve within their own passionate communities, while providing the necessary incentive to site owners (increased traffic from search) for more widespread adoption."



 In our HO - Google will quickly outdo any RDF indexing capabilities Yahoo!might incorporate and offer Semantic Web search engine return page (SERP) results in true Google fashion.. propelling RDF indexing to new heights in short order.

 The Semantic Web was conceived by Tim Berners-Lee, inventor of the WWW, URIs, HTTP, and HTML. The World Wide Web consortium (W3C) is working to improve, extend and standardize the system, and many languages, publications, tools and so on have already been developed.

 

Competing with Google..? Not very likely..

 

 There are murmerings, as of late, on the web about "Google competitors".
  Hmmm.. lets take a look.

 The European Union says it will help France with $152 million to fund a new Google competitor named Quaero (latin for "I Search").

 The Quaero project came to light in late 2004-early 2005, when France and Germany proposed a Google-killer search engine; the proponents put forth Google was developing a de-facto monopoly on Internet searching.
The European Commission is quoted as stating the Quaero project "brings positive externalities for the community as a whole."

 Another new arrivial is Wikkia Search; the alleged "human-powered" search engine - or theoretically might be, if humans get the chance to consummate the concept. The Wikipedia founder Jimmy Wales' brainchild whose alpha launched in Januatry 2008, has little substance to date.

 Then there's Blekko, conceived by Rich Skrenta, author of the world's first microcomputer virus. Skrenta's previous success NewHoo, the late 1990s web directory bought by Netscape and then renamed as the Open Directory Project presumedly gives investors reason to believe he might conquer Google.

 We're huge fans of Google, of course - some of our team were original beta testers. We've seen Google evolve into the super-performer it is now, and all of us (although we don't wish ill on anyone's project) can only say..
   Take on Google .. ? Best of luck.

 A brief real-world glimpse at what beating Google at their own game would entail..

  Terrabytes upon terrabytes of storage

  A global network like Akamai (Google's bandwidth partner)

  An infrastructure of super-brilliant Web search engineers

  All but undefeatable spam-proofing (and the time to perfect it)

  20 years experience with intricate webs of evolving algorithms

 Need we go further?

 

Content freshness and PageRank.. Hand in hand?

 

 In a patent granted December 2007, and just recently fully implemented, Google explains how they could place urls into "crawl layers" to determine how often the page needed to be crawled.

 For example breaking news type websites or heavily used popular forums would fall in a “real time” crawl layer to be crawled almost continually; And an average blog homepage, for instance, might fall in the “daily” crawl layer.

 The crawl layers could be altered daily with an algorithm that computes a score based on the document's PageRank and frequency of change.
* Pages with an abnormally high or low daily score might then be moved up or down a crawl layer or two, accordingly.

 With respect to various page content formats;

 Many news sites, blogs and CMS systems are given maximum PageRank right when their content is published; posts & articles (typically) move further and further down over time until they are archived.

** Other websites such as RSS syndicators and forums that display the latest posts glean an initial burst of PR followed by a natural reduction, as they increase in age.

 The web is laden with websites that duplicate content from directories, etc. and do so just for the sake of “freshness” - and it's now obvious that although Google is crawling these sites more often, Gbot is not rewarding them with permanent PageRank based primarily on their updating habits!

 

Milestone talks to Matt Cutts about On-line success

 

 Milestone, the "anonymous" blogger, has posted an article noting an interview with Google's Matt Cutts.

 The topic of the interview was the top ingredients for any small business that wants to be successful on the net.
The interview can be viewed here.

 

Neat-O things you can do with Google!

 

 PC World released a very interesting article today, that contains stuff one can do with Google..

 It contains stuff even we hadn't run across yet!
Here's the link.

 Enjoy..

Google + DoubleClick in Europe - together at last

 

Per CNet's News Blog, today:

 "European antitrust regulators on Tuesday approved Google's $3.1 billion merger with DoubleClick, paving the way for a blockbuster deal in Internet search and publisher-based advertising tools.

 Approval by the European Commission, which came without conditions, had largely been expected to occur this week. The Commission's announcement comes three weeks before its April 2 deadline, in which it had to determine whether to nix the deal.

 With the Commission's decision in place, Google announced on Tuesday that it has formally closed its merger with DoubleClick.

 The organization also cited concerns that the merger would aid Microsoft in its goal to acquire Yahoo. That deal is largely being driven by Microsoft's desire to bolster its online-advertising capabilities.

 'Instead of ensuring competition, (the Commission) and the FTC have literally paved the way for the emergence of a global digital duopoly over online advertising,' noted the Center for Digital Democracy".

 

New from Google's backer - A visual search engine..

 

 Today, Sequoia Capital, a major investor in Google and Yahoo, has unveiled its latest investment - an advertising-based visual search engine named Searchme.

 Via Kara Swisher (technology columnist for the Wall Street Journal), on her section in the the All Things Digital website:
"And, indeed, Searchme does look good, resembling a mash-up of Google with Apple’s popular Cover Flow three-dimensional graphical user interface used on its iTunes service, with a little of Ask.com’s categories thrown in.."


Kara notes that Searchme has been working with Adobe to crate a "page rifling" effect, where "The top screen, after a search term has been entered, then creates a stack of pages to represent the page links on a list below."




 Here’s a video interview with Kara and Searchme's CEO Randy Adams:

 

Another Google Dance, or just a small shakeup?

 

 In the Google Search News section of Webmaster World, a 22-page thread has started about major swings in Google search results for one and two word search terms.

 Some venerable experts are pointing to the new "categorized" search results appearing in the Google Experimental Search Tool results and many of the parallel Google datacenters; and the fact these results are now appearing in regular searches as evidence this is a notorious Google "Dance", not just the rolling algorithm updating Google's crawler, Gbot regularly undegoes.

 Google recently recognized and responded to a blatant need for fltered searches; it was painfully obvious that with the billions of pages stored in Google's "Document Barrels" across the world, there was a definitive categorization evolving of URI and URL's. Millions of web resources were of one kind or another, uniquely different and web surfers were specifically seeking these different types of various unique content.

 Regardless, it's obvious the new SERPs with segregated results for images, videos etc. that relate to the searched-for words & phrases are here to stay.
  In our Humble Opinion.. this is a nice add-on from Google.