Forum Moderators: Robert Charlton & goodroi

Message Too Old, No Replies

English vs. All Web discrepancies

language filter is more than just that

         

Neticman

12:39 am on Dec 3, 2006 (gmt 0)

10+ Year Member



While changing the Search Preferences, I find some hard-to-explain discrepancies.

Searching for English Pages vs. All the Web usually provides very different results, even in English-only websites.

It seems that the English filter is also a quality-page filter.

For instance:

site:webmasterworld.com offers 113.000 English pages, while 183.000 all-languages pages.

Are there 70,000 pages in this site, that are neither English, French, Spanish, German, Portuguese or Italian? What language are they?

This patterns is repeated in other sites and other languages.

Any explanation?

Thanks.

tedster

1:57 am on Dec 3, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Interesting observation, Neticman.

I just did the same two searches for webmasterworld.com, but I get MORE pages with "English only" than I do with "any language" -- 398,000 for English only, and 296 for any language.

That pretty much tells me what I first thought. That these numbers are first of all VERY rough estimates and not something we can depend on for analysis unless we can pin down what factors generate the variation.

Neticman

6:51 pm on Dec 3, 2006 (gmt 0)

10+ Year Member



I repeated the same observation in 14 domains. In all cases, a significant part of the pages are associated with an undeclared language. This is, outside the 38 listed as options in the Preferences.

Back to an old argument: It is necessary to declare the language in the header?

Does google associate low-quality pages with an undeclared language?

MThiessen

10:44 pm on Dec 3, 2006 (gmt 0)

10+ Year Member



I declared the language in the header on a page that was translated into spanish. Adsense ads immediately started showing in spanish.

The language header may not be so important for english, but I believe it helps with other languages.