Norman Walsh wrote about a simplified version of XML. The draft is called: Extensible Markup Language (XML) 1.1 Kernel. DOCTYPEs are irrelevant and would now cause a non well-formed error to be raised. Bye bye entities not part of the five known ones. Bye bye validating XHTML pages. I personally think this is good move forward to get rid of all that legacy stuff. No browser parses a DOCTYPE when it encounters one in an XML document. MSHTML MSXML used to do it, but stopped doing it by default in MSXML6 according to Sjoerd. As browsers have no quirks versus standards mode in XML they are not needed for that purpose and are better dropped. Now the only thing that is left is to merge that with the namespaces specification and get it all out as a single draft. Yay for when that happens.
The CSS WG released four working drafts, one of them is Last Call. If you have enough spare time before Januari 14 to read a draft and comment on it I suggest you do Selectors. Input on everything is always appreciated, but do Last Call first. Missing that can give problems. The other drafts are CSS3 module: Cascading and inheritance, CSS3 module: Multi-column layout and CSS3 Advanced Layout Module. This last one may at some point bring grid layouts to the web. (For real.) Now that would be cool. At the moment it is more a bunch of thoughts though, as it seems.
Comments
The only good thing the doctype declaration brings is entities in the internal subset. Entities are useful for, well, entities.
Posted by zcorpan at
In retrospect, I think it was a mistake that the set of predefined entities in XML was not specified to cover all the HTML4 entities. Well, too late now. Just use UTF-8.