Re: The phrase: Node created by Document needs clarification

Sounds like one vote for "no", which I partially support.

I suspect there are surely improvements that can be made
to reduce the number of behaviors requiring "backdoors"
in the implementations.


----- Original Message ----- 
From: "Joseph Kesselman" <keshlam@us.ibm.com>
To: "David Brownell" <david-b@pacbell.net>
Cc: <www-dom@w3.org>
Sent: Thursday, August 30, 2001 6:40 AM
Subject: Re: The phrase: Node created by Document needs clarification


> 
> >Would it be a Good Thing if the public interfaces were sufficient
> >for such purposes -- both implementation and testing?
> 
> Trying to do it in the general case would mean imposing unacceptable
> constraints on implementation.
> 
> The "Embedded DOM" chapter of DOM Level 3 was going to consider the
> question of combining nodes from different implementations into a single
> tree. This turns out to be a very ugly problem, and this feature has
> trimmed itself down and will only support nodes which are subclassed from
> the same DOM implementation. If there's a serious requirement that it do
> more, you may want to take that up with the EDOM editorial team and provide
> some help on brainstorming how this could be made to work properly. (First
> show that it's needed. Then show that it's possible...)
> 
> ______________________________________
> Joe Kesselman  / IBM Research
> 
>