Google Relevancy Poor Based on Generalization

Apr 12, 2005 • 3:19 pm | comments (0) by twitter Google+ | Filed Under Google Search Engine Optimization
 

Everyman (Google Watch), has been after Google for as long as I can remember. I do not blame him, I actually think it is good for the industry. He recently posted a thread at Search Engine Watch forums comparing Google to Yahoo, MSN, Ask Jeeves and Gigablast. He named his thread Google has big problems, and showed how Google's results were less relevant then all the rest. The example given:

Exact search term: apache "2.0" namevirtualhost

On Yahoo you get the exact documentation from apache.org that you are looking for in positions one and two.

On MSN you get the exact documentation from apache.org that you are looking for in postions one, two, and three.

On Ask Jeeves you get the exact documentation from apache.org that you are looking for in positions one and two.

On Gigablast you get the exact documentation from apache.org that you are looking for in positions one and two.

On Google you find, if you haven't given up by then, the exact documentation from apache.org that you are looking for in positions 70 and 71.

This example brought me back to my college days when I took a Philosophy class that focused on Logic. So I went over to Google and did a search on "logic fallacy" and was presented with an excellent resource named Stephen's Guide to the Logical Fallacies. Then I clicked on the "Index of the Logical Fallacies" and found what I believe I was looking for. The Hasty Generalization of Inductive Fallacies. I believe, I am not an expert in this field, that Everyman's logic falls under what is defined as a Hasty Generalization. If not that, then an other logic fallacy. But my point is, I used Google to find the technical name of the fallacy I was looking for, to show how one specific example in a specific time during a day, is not a solid proof of reasoning.

Previous story: Become.com Out of Beta
 

Comments:

No comments.

blog comments powered by Disqus