NOINDEX for PDF

[sfowler]

I include many of my pages in PDF form to download, but I am planning to make them available in American format (letter) and European format (A4) for convenience. Since I know that PDF documents are crawled and indexed and both will have identical content, it would be nice to mark one copy NOINDEX or equivalent. Is there any way of doing this with a PDF?

[HardCoded]

If you have consistent naming practices, like -a.pdf for American and -e.pdf for European, then you can use a robots.txt file:
Disallow *-e.pdf

[minstrel]

Alternatively, why not just put a prominent label for "American format" or "8.5 x 11" format" or something similar on one version and "Metric format" or "A4" on the other and let them both be indexed? The worst that can happen is that both will appear in the index and your site visitors will be instantly aware of a choice...

[bhartzer]

Another way to handle it is to make two different folders/directories for the two different versions. Then have the robots.txt file disallow the two directories.

[minstrel]

Originally Posted by bhartzer

Another way to handle it is to make two different folders/directories for the two different versions. Then have the robots.txt file disallow the two directories.

I'm not sure I understand the rationale of creating pages and hiding them from search engines... what would you gain from doing that?

[Dave Hawley]

I would also create a HTML format version, optimize it, then simply offer the 2 different PDF versions to download via a link.

[sfowler]

That is what I do. The main page is a HTML, and I have a downloadable version as PDF. Since a PDF and HTML are different enough, I have no worries about this, but I don't want two identical PDF files being indexed. I guess a sepaate directory is the best solution and just quarantine the American ones.