Duplicate Content - Text Only or Text plus HTML?

Any evidence?

[rehabguy]

Does the Google duplicate content filter take HTML into account (IE - Same header/menu/footer structure on every page) or does it look at text only?

Any quesses or hard facts on this? Thanks!

[paybacksa]

personal experience - html won't matter, unless it makes up a large part of the total page content.

So if you have content, content matters most. If you ahve little block content, html can matter.

Example: page with small center content and large sidebar of html, sidebar same on all pages: trips dup tester. Same page with large center content passes dup filter test. Go figure.

I understand that G uses a sliding overlapping window, to generate regression factors (weights) which are compared across pages. So it makes sense that if content areas are defined by structural elements (to determine what content gets factored), this behavior would be noted.

If you have a page that gets snagged for toomuch html/sidebar and not enough central content, adding a paragraph of random content (a quotation, for example, or a text advertisement) can make all the difference.

[lizardx]

"page with small center content and large sidebar of html, sidebar same on all pages: trips dup tester. Same page with large center content passes dup filter test"

We've seen different, this might depend on how well coded the page is. Treat the page level HTML as seriously as you (hopefully) treat the backend programming and you might be surprised at the outcome.