Forum Moderators: open
Example:
A page has 1000 pages, and 3 templates are used between these pages. Parts of each pages are the same, i.e. the template. Dynamically generated pages usually have more template than unique content, and that's usually a good reason to dynamically generate them.
Are we at a point where engines will "strip out" or de-weight unnecessary elements/words of the page in their weighting?
Maybe hyperlinks too, what "value" does a link to the home page have when its placed in the same position across all the pages, maybe its no real indication of relevancy to them?
When they create their indexes, I'm sure there's a mass of redundancy across webpages within the same domain, perhaps even across multiple domains......
The thing is, do they do anything about it and does it affect ranking in today's SERP's? :)
>>>>de-weight templates, what's everyone's opinion on this?
Sites that I have seen in the results or have built myself that use templates with little variance besides product information do not appear to be de-weighted now. Actually these sites seem to receive the benefit of achieving a clustered set instead of sites that do not follow a template pattern. Even if the search phrase is focused on words that appear on all pages, a clustered pair is given in the results. That being said, those similarities are not being de-weighted.
>>>Maybe hyperlinks too, what "value" does a link to the home page have when its placed in the same position across all the pages, maybe its no real indication of relevancy to them?
I believe they are treated with great importance if they contain at least a few words relevant to the particular terms. I have seen numerous cases where contact pages, privacy policies, home pages, etc. are listed as #2 in a clustered set on Google's results when a sub page is listed first. Therefore, I believe they are weighted more since these pages appear more often in a clustered set, instead of other pages that are not linked to on every page yet contain some of the same keywords.
Makes sense, 'specially with all the RSS/XML type information that's spread across the web. There must be millions of docs that are identical.
>>>little variance besides product information do not appear to be de-weighted now. Actually these sites seem to receive the benefit of achieving a clustered set instead of sites that do not follow a template pattern.
Are the templates fatter than the unique content? Or vice versa maybe. I see what you mean by "clustering" as it gives a bot a fair idea of what the content is about if there's steady set of terms that are being used.
I wonder which Webmasterworld page matches the search [google.com] "control panel, site search, glossary, PubConference, library, help, recent posts"
I was thinking it would the home page, having the highest PR and the home page of the template. But it's not. Although the home page and the page in SERP's have the same words (though the densities are different) the home page isnt listed at all.
Searching with the omitted results [google.com]- all the threads that mention the words outwith the template are getting the extra attention. The home page of WW is nowhere to be seen. I was thinking the PR of the home page would put it up there for those matching words in the template.
The only threads where the template matches is in member profiles, where the template is larger than the member profile itself.
Makes me wonder if the template words are being "decounted" or the threads that show for the search are the ones that "count only". Where content is thin, like member profiles, it seems to favour the templates.
