Forum Moderators: open

Message Too Old, No Replies

Smiley on Google Tool Bar

What's it for?

         

cornwall

2:39 pm on Oct 26, 2002 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



I am sure the question has been asked before, but I cannot find it in the forum index.

What on earth do they use the results of the Smiley vote for. I put the smiley on the bar and the mouseover gives respectively "Vote for this page" and "Vote against this page"

Are Google expecting me to vote for my page and damn my competitors pages?

Do they do anything with the results?

Is the most "popular" site on the web the Homepage of some guys that has voted for themselves millions of times?

Are the results posted on Google?

David Steven

5:12 pm on Oct 26, 2002 (gmt 0)

10+ Year Member



Google says: "If you especially like or dislike a page, you can vote for or against the page by using these buttons. Just click the happy or unhappy faces to tell Google that you like or dislike a page as you surf. These buttons can also be used to report especially good or bad results after searching on Google. Indicate satisfaction or dissatisfaction with your results by clicking the appropriate button after performing a Google search. This feature is in testing; for now, you will not see any immediate effects by voting for or against a page."

David Steven

5:26 pm on Oct 26, 2002 (gmt 0)

10+ Year Member



And there's this from 2001:

"We do have a lot of safeguards in place to make sure someone can't hurt someone else unfairly," said Matt Cutts, the software engineer overseeing the new feature.

Google predictably didn't go into specifics about how this would be done, but the key reassurance to take away for the moment is that the voting data is not automatically being used to alter rankings. Instead, it is currently used as a flagging mechanism, to help Google understand which pages should be subjected to human review.

"It won't be used in the production system until a human has validated it or until we fully trust the methodology," Cutts said.

[searchenginewatch.com...]

Brett_Tabke

5:37 pm on Oct 26, 2002 (gmt 0)

WebmasterWorld Administrator 10+ Year Member Top Contributors Of The Month



Related background reading:

Rank Aggregation Methods for the Web [www10.org], by Cynthia Dwork (Compaq), Ravi Kumar (IBM), Moni Naor, D. Sivakumar (IBM). ("This work was done while the author was visiting the IBM Almaden Research Center and Stanford University")

We believe that reliance on evaluators in defining spam is unavoidable. (If the evaluators are human, the typical scenario during the design and training of search engines, then the eventual product will incorporate the biases of the training evaluators.) We model the evaluators by the search engine ranking functions. That is, we make the simplifying assumption that for any pair of pages, the relative ordering by the majority of the search engines comparing them is the same as the relative ordering by the majority of the evaluators. Our intuition is that if a page spams all or even most search engines for a particular query, then no combination of these search engines can defeat the spam. This is reasonable: Fix a query; if for some pair of pages a majority of the engines is spammed, then the aggregation function is working with overly bad data --- garbage in, garbage out.

tedster

6:26 pm on Oct 26, 2002 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Nice article, Brett. Thanks for the link.

Intuitively, a search engine has been spammed by a page in its index, on a given query, if it ranks the page "too highly" with respect to other pages in the index, in the view of a "typical" user...

This approach to defining spam:
(1) permits an author to raise the rank of her page by improving the content;
(2) puts ground truth about the relative value of pages into the purview of the users...
(3) does not assume unanimity of users' opinions or consistency among the opinions of a single user; and
(4) suggests some natural ways to automate training of engines to incorporate useful biases, such as geographic bias.

Definitely sounds like the Google Toolbar, doesn't it?

David Steven

7:03 pm on Oct 26, 2002 (gmt 0)

10+ Year Member



It does sound like the toolbar - but doesn't the toolbar face a number of problems if it is to fulfil this function:

1. People are presumably not voting in large numbers while there is no clear feedback on what their votes do - making it very hard for Google to test how it will work when an automated part of their algo.

2. There is no strong compulsion to vote - unlike with say, SpamNet, an email spam system where you block mail that has passed through the auto filter in the hope it will be blacklisted in the future; and where blocking a mail is as easy as hitting delete.

Just can't ever see it working really... but what do I know? I know that I used it a couple of times and then had stopped even seeing the buttons on the toolbar until this thread.

cornwall

11:01 pm on Oct 26, 2002 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Thanks for the feedback to my original question.

I guess the thought that I take away is

. People are presumably not voting in large numbers while there is no clear feedback on what their votes do - making it very hard for Google to test how it will work when an automated part of their algo.

Personally I see no incentive to vote, as I have no idea how it is really being used.

Yet Google have left it there for some time, if it was a lulu, then they would have removed it by now (presumably)

Just very odd :o