[whatwg] Suggestion for a Specification: XUL Basic

[Matthew Raymond]

Ian Hickson wrote:
 >> You could make the argument against box, but hbox and vbox are two
 >> of the best supported XUL tags in existence.
 >
 > And <font> is one of the most supported HTML tags, but that doesn't
 > make it a good tag either.

    If you look at the second half of that paragraph (which you did not
quote), you will see that I stated we should probably drop those 
elements from "Keymaster" anyway. This is due to the fact that 
"Keymaster" was designed to augment XHTML, not duplicate its functionality.

 > I'm just saying that the markup XUL uses may not be the best, and
 > that anything that is put into the WHAT specs would be designed on
 > its own merits, with a look for compatibility with HTML. For example
 > the <menu> element already means something in HTML so couldn't be
 > used.

    The <menu> element is depreciated in HTML 4.01, and it doesn't exist 
at all in XHTML, which is what "Keymaster" was intended to be used with.
Furthermore, the child element for the HTML <menu> is <li>, while the 
child element for the XUL <menu> is <menupopup>, so in a well written 
web page the child elements are mutually exclusive. Just put in a check 
to see if there are any <li> tags for the sake of backward 
compatibility, and when the check turns back true, treat <menu> as a 
<ul> element and ignore all menu-related child elements inside it.

    Then again, if we actually integrate "Keymaster" into HTML, we can 
simply rename the tags to whatever's convenient.

 > The term "XUL motor" doesn't mean anything, maybe you mean "XUL
 > rendering engine" or "XUL host".

    Well, "XUL Motor" is a term used by Gerald Bauer, so I should have 
known better. ;)

 > In either case, the WHATWG work is not aimed at XUL
 > rendering engines but at HTML and XML+CSS rendering engines
 > (primarily IE, Mozilla, and Opera).

    Fair enough. I'll move the XUL Basic suggestion to a W3C list and we 
can continue to discuss "Keymaster" as a possible enhancement to HTML on 
this mailing list.