On Wed, 7 May 1997, Charles Wicksteed wrote:
> 1. In some cases the metadata relates to the body of the same HTML
> file, and in other cases the metadata is in an HTML file but refers
> to a separate PDF-format file. How should we indicate the difference
> between these to the search engine? The metadata for the PDF file
I'd proposed using the HTML LINK element to the robots list with
limited success.
The idea was to use
<LINK REV=META HREF="http://a.b.c/doc.pdf">
in the header, causing robots to associate any metadata or fulltext search
in the HTML shadow document with the PDF resource. There's a similar
method in HTTP which could work with non-HTML metadata or in the
forward direction (from the PDF file to the metadata).
One could also use <A REV=META HREF="http://a.b.c/doc.pdf">Here!</A>
as a visible link.
The MathN broker in Germany uses the DC.IDENTIFIER (scheme=URL) approach,
with a custom search engine. There's also an explicit anchor.
Andrew Daviel
TRIUMF & Vancouver Webpages
|