Quite true - in the script I use, I have this for the W3C validator: if (stristr($_SERVER["HTTP_USER_AGENT"],"W3C_Validator")) { $mime = "application/xhtml+xml"; }
As to why, my own personal reasons are three-fold: 1. The W3C is clear that XHTML 1.1 should not (different than must not, I'm aware) be sent as text/html - http://www.w3.org/TR/xhtml-media- types/ (I think that's the right link - I'm using a PDA and browsing is a pain). 2. Conforming UAs *should* refuse to render invalidly marked up pages if sent with the proper MIME type - that saves me a lot of time in development. 3. When dealing with MathML or XHTML Ruby at all - your documents have to be sent as XHTML (application/xhtml+xml, application/xml or text/xml) - not HTML, unless of course you want to stick it into an <object> tag. I do believe that's all for now - however I'm not at work and my mind is in a different place, so hopefully this came out coherently enough. :) Cheers, > On a related note, since the W3C's validator doesn't send an > "HTTP_ACCEPT" header, you should also look at the "HTTP_USER_AGENT" > header as well. While I normally would advise against browser sniffing, > I make exceptions for the W3C Validator, the W3C CSS Validator, and the > WDG Validator. > > > > > ****************************************************** > The discussion list for http://webstandardsgroup.org/ > > See http://webstandardsgroup.org/mail/guidelines.cfm > for some hints on posting to the list & getting help > ****************************************************** > > > -- ****************************************************** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help ******************************************************