At best, someone has an answer, but I imagine that a discussion on the topic could be helpful also.
Basically (for those not familiar with XSL), when a current browser loads an XML file (www.mysite.com/xmldata.xml), and the XML file has a link to an XSL file, the browser will process the XSL file, but the source will still show the original XML. Based on a blog post I found, someone used this to load Flash on a presentation layer via XSL, but have XML in the source. It seems until this is abused, and the XML is formatted similarly to HTML, this is a viable way for RIA developers to have their content indexed by search engines.
This is great and all for an all Flash site (because there is a relatively small amount of HTML required to display Flash), but I am trying to find a solution for hybrid sites or “Progressive Enhancement” sites. Technically, I could start building all my hybrid sites in XSL files instead of HTML, but I don’t really want to do that.
I have tried a few different solutions to no avail. I have tried using php’s libxslt, but the translation is done on the server, and therefore the source shows the translated HTML, not the XML I am looking for (I imagine this would be the case for any server-side processing). I also tried JavaScript and AJAX, looking for a client-side approach, but that shows the HTML instead of the XML also. I have also tried outputting XML with the XSL link as a string, but it seems browsers will only process the style sheet if the file extension is .XML. The only solution I have found is to use an iframe to handle the XSLT processed XML and have HTML in the containing page. I have read that search engine bots will index the iframe source if not blocked, but the problem is that it is indexed as a separate page (duh!), not part of containing file (what I want).
Does anyone have a solution, or even a suggestion? I hope this wasnt too drawn out or confusing.
