I think there are valid arguments for both sides of this. This is also where I agree with the approach of the Apache/Cocoon advocates in that you serve up the solution which be suits the user agent.
As standards developers we are working in an imperfect world and it's what frustrates us all. What I feel actually saves the HTML argument, in this case, is that browsers actually have very very intelligent parsers in them. It takes a lot of work to handle the thousands of cases that quirksmode can throw up. Thankfully that also works in our favour when we slip up on our own validation. If you took all the quirksmode logic out of the parser in the browser and it only did valid DTDs for HTML4 and XHTML, and spat the dummy like SGML and XML parsers when it found invalid markup, and where all documents were valid, I bet you that out of all the designs the XHTML would render more accurately. The reason being that if you are not closing all your tags it can become a guessing game for the parser where the CSS declaration may end in various parts of the document. It always strikes me that when using HTML4 you are at the mercy of the arbitoriness of the parser. I know there doesn't seem to be any obvious problems, but just the basics of parsing seem to leave you with the feeling it is inherently there. That to me is the main reason to use XHTML over HTML in this argument. ----- Geoff ****************************************************** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help ******************************************************
