[www-annotation] <none>

A397@inrialpes.fr>
Subject: Re: Orphaned annotations
Date: Tue, 19 Mar 2002 11:20:46 -0000
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4522.1200
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200

 "Jose Kahan" <jose.kahan@w3.org>
> Hello Matthew,
>
> On Mon, Mar 18, 2002 at 06:14:10PM +0000, Matthew Wilson wrote:
> > At 17:22 18/03/02 +0000, you wrote:
> >
> > ... which isn't really true - eg do HTML4 tables contain an implied
<tbody>
> > element if it isn't explicitly present in the document? From memory,
> > Annozilla says yes, and Amaya says no.
>
> That's correct. Amaya does add that element when it's missing. However,
> as we can't be sure that all browsers normalize HTML in the same way,
> we decided to not take into account the internal added elements when
producing
> an XPointer. On the other hand, if everyone normalized HTML in the same
way,
> then, for a valid HTML document, we could have an XPointer working.

Amaya is currently producing XPointers incompatible with Snufkin and
Annozilla which both do have implied elements (ie we're working on a
Normalised parse tree.) so TBODY insertion (a very common normalisation
on the web) is needed.

Is it possible for Annozilla and Amaya to automate the generation of
XPointers?  If we say generate XPointers for every element in 20 or 30
agreed webpages we can soon see how close the normalisations are and then
look if they are feasible.

As this is related to work we've already done in ER (comparing openSP's
and browsers normalisations for pointing to invalid elements in a page.)
I'd gladly co-ordinate any research into this, but don't know anything
about hacking Amaya (I could probably hack Annozilla myself.)

Jim.

Received on Tuesday, 19 March 2002 06:23:07 UTC