A website I'm working on uses xml files for each page (rather than php or html files). The xml is converted to html (soon html5) via xslt on the client side. I intend to make these pages usefully semantic and I was interested in how search engines see the page.
Since a human on the site sees the URL as "www.example.com/webpage.xml", I assume that the page is being processed by search engines as xml too, rather than the result of the xslt conversion. Is this true?
If this is true, do I even care about html5 microdata? Would I instead be adding semantics to my xml document? What language(s) would be appropriate for this?
Finally, what about the meta tags in the head of html pages? Do I not care about those either as they'd result from the xslt and would therefore be ignored?
I hope that all seems logical.
Thanks,
Patrick.
Is the convertion done on server side? If so, the search engine must see the generated html with all meta-data. You should check the http response with Firebug tool for example.