W3C home > Mailing lists > Public > www-dom@w3.org > July to September 2006

Re: Special group of functions for NON-ELEMENT nodes in document

From: Boris Zbarsky <bzbarsky@mit.edu>
Date: Tue, 26 Sep 2006 10:00:50 -0500
Message-ID: <451940A2.9010204@mit.edu>
To: Matt Dockerty <matt@nistrum.co.uk>
CC: Master Br <master@sitesbr.net>, www-dom@w3.org

Matt Dockerty wrote:
> I agree that the specification leaves too much room for maneuverability.

It's not quite that simple.

I realize the discussion in https://bugzilla.mozilla.org/show_bug.cgi?id=26179 
is very long, so here is a summary:

1)  IE/Windows clearly doesn't have the DOM as the underlying data structure.
     This becomes obvious because the styling of a node changes what its
     firstChild node is, for example, which is very much not the way CSS and the
     DOM are supposed to interact.
2)  Independently of this the HTML spec specifies all sorts of behavior wrt
     whitespace that's only half-implemented by various UAs.  It's hard to tell
     exactly what IE implements here due to item #1.
3)  Some UAs (for example Opera) do something to have IE compat in some cases.
     I'm not sure what the details of that are.
4)  The DOM allows XML parsers to drop whitespace in certain cases, as I
     understand.  But there were so many stakeholders in the DOM spec that only
     someone who was there could say why those parts are there.

 > Why is whitespace an optional part of a DOM tree

In which cases?  HTML or XML?  And with what sort of parser and what sort of 
parser flags?  ;)

As for browser compat, there's no way for a browser that claims CSS compliance 
to have IE compat, due to item 1 above.  So the real question is what compat 
should be with.

-Boris
Received on Tuesday, 26 September 2006 15:02:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:58 GMT