Surftoolbar...

Google's code change shifts hits content farms hard.

Kjell Gunnar Birkeland Bleivik March 9 2011. May be updated with additional information.

It is well known that since the turn of the year Google has announced that sites with shallow content, socalled content farms, will be hit in the future. Now the rumor has become a reality, and don't think that the search engine wars are over.
They have just begun

The battle isn't for the hearts and minds of the public; It is for the heart and soul of the internet - a battle to put "content farms" in their place and suffocate sites which, due to creative SEO, somehow manage to appear higher in search results than the original sites whose content they are re-blogging or flat-out stealing.
Source: Staving off the search war apocalypse

And these algorithmic twists or changes have had monetary implications. It is estimated that $1 billion will be redistributed across the online publishing industry. Some companies have had to lay off some of its workforce. As always, there will be loosers and winners. Here are some related sources:

Non-linear signals in search engine algorithms.

Kjell Gunnar Birkeland Bleivik June 7 2010. May be updated with additional information. Updated august 31 2010.

Based on these observations, I will conclude that there is a non-linear time signal in the algorithms of search engines that may dominate the other signals as long as the discussed topic is hot.

A scientific hypothesis testing may start with observations. The requirements of scientific testing are strict and we shall not report them here, but mention one important source:

This short article is based on my own observations and not tested scientifically. The heading of the article is non-linear signals in search engine algorithms. I will use the log normal distribution as an example of how I mean to have identified time dependent signals that are treated non-linearily in the search engine algorithms.

If you look at the graph of the log normal distribution, where x represent time, you will observe a fast increase and then a slower decrease like those observed in momentum curves. This illustrates the non-linear time effect I have personally observed on the SERPs. For example when a topic that is discussed on a forum is hot, this non-linear effect may be the dominant element in the SERP algorithm and the forum thread, blog post or discussion is brought to the first position on the SERPs. It stays in the top position as long as the topic is hot and then it gradually finds its natural position on the SERP pages where the time element is not longer important. Based on these observations, I will conclude that there is a non-linear time signal in the algorithms of search engines that may dominate the other signals as long as the discussed topic is hot.

Search evolution, some thoughts and trends.

Kjell Gunnar Birkeland Bleivik August 21 2009. May be updated with additional information.

Imagination is more important than knowledge. The future of search may be a combination of art and science.

Search engine bots and web crawlers crawl the world wide web 7 days a week and 24 hours a day for news, new subjects and information on new sites. So long key word search has been the dominating technology. Now there seem to be an ongoing competition between the highly respected company Wolfram Research and the most popular search engine Google. See the videos and articles below.

Wolfram will make knowledge computable and Google will extract and present data in a spread sheet (square matrix) format. Google has also announced a new version of Google Search that is in Beta stage when this article is written.

A web page or document is a tag soup that can be viewed in the ordinary way or as markup from the view + source menu in your browser. The W3C's Document Object Model (DOM), a language independent platform, is basically an object oriented representation of your document. Every element in your X(HT)ML document is represented by an object in that document's DOM tree. The tree structure is called a node-tree. This object oriented presentation of the document opens a whole new world of opportunities. You can set the objects as properties in new classes (objects). Then it is up to your imagination how you will "compute" and present information. Some years ago

I wrote a matrix class in C++ using templates. There are some good classic literature references there like Andrew Koenig's "Templates and generic algorithms" written in the june 1994 issue of the Journal of object oriented programming (JOOP). There are two related articles by the same author that is not mentioned there:
  • "Generic iteartors" JOOP September 1994.
  • "Function objects, templates and inheritance" JOOP September 1995
Generics is a very strong form of polymorphism since it can be used to parameterize code.

Combined with AJAV, the XMLHttpRequest Object and other web application technologies the web's search space should grow in dimension and depth and the way search results are presented is up to your fantasy. As Einstein once said: "Imagination is more important than knowledge". The future of search may be a combination of art and science. We list a simple example of what is possible by DOM manipulating this: simple static table

There is nothing different if the cells of this table are links or other X(HT)ML elements or text snippets produced e.g. by the DOM method:
var linksInDocument = documnet.getElementsByTagName('a');

Litterature:

Michael W. Berry and Murray Browne:(2005) Understanding Search Engines: Mathematical Modeling and Text Retrieval (Software, Environments, Tools), Second Edition

David A. Grossman:(2004): Information Retrieval: Algorithms and Heuristics (The Information Retrieval Series)(2nd Edition)

Amy N. Langville and Carl D Meyer: (2006) : Google's PageRank and Beyond: The Science of Search Engine Rankings.

Bing Liu:(2008) Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data (Data-Centric Systems and Applications)

Michael Schrenk:(2007) Webbots, Spiders, and Screen Scrapers: A Guide to Developing Internet Agents with PHP/CURL

A short note about SurfToolbar.com

As you will note, SurfToolbar.com is about search engines and search engine related topics, more precisely:

  • Search engines,
  • other search resources and
  • some important articles and papers

that we have found surfing the web since its start.

Below you find some inks of historical interest and some links to the latest search technologies.

Related video