europeforvisitors - 4:32 pm on Dec 23, 2004 (gmt 0)

Actually I'm trying to show that the distinction isn't quite as clear as what some might make it out to be.

True, and that's why algorithms were invented. :-)

Google is certainly capable of developing an algorithm that takes many different factors into account before determining whether a page is an affiliate page, an e-commerce page, or none of the above. At the most basic level, such factors would obviously include the presence (or lack) of an affiliate or shopping-cart link, but other factors could come into play, too: e.g., certain layout characteristics, the presence of certain words or phrases, and the overall theme or characteristics of the site.

And remember, Google doesn't have to determine with absolute certainty whether a page is an "affiliate," "e-commerce," or "information" page. If, for example, Google Search offered buttons that allowed users to choose between "information" and "commercial" results, the algorithm wouldn't need to act like a binary filter--it could simply weight the results more heavily in favor of the page type that the user had selected.

A "no directories" option could work in the same way: not by filtering out directory pages entirely, but by giving them less weight in the search results.

Obviously, Google doesn't want to make the search process any more complicated than necessary, but it wouldn't have to make users click radio buttons or whatever on every search: Users could continue to accept the current potluck search results by not clicking the radio buttons, or--better yet--they could use "options" or "my preferences" settings to establish their personal defaults.