Re: Orphaned annotations

On Mon, 18 Mar 2002, Marja-Riitta Koivunen wrote:

> I agree that we cannot make all the content of the Web suddenly use ids,
> however, for a group creating a document for review purposes it does not
> sound too difficult. Amaya even has a command for adding ids automatically
> to a page.

No, no, NO!  That's only half the story!

For IDs to work properly, you need to *guarantee* that they will be
anchored to elements across edits, regardless of who edits a document
or what tools they use.  This is the province of a document management
system, not a mere editor.

> And I'm really happy that you are helping us to make the pointers more robust.

Robustness in the presence of consistently-used IDs is easy.
I want to see it everywhere!

> >I think it's clear fuzzy pointers without a mechanism to know how
> >reliable the fuzzy pointer is can't realistically be used.
>
> Here I disagree somewhat. It depends so much of the usage. We use the
> pointers in annotations all the time for reviewing purposes. Naturally you
> need to be aware of the problems but why should we stop using them as it
> helps so much.

A review is a situation that can be well-handled by annotation-aware
tools.

> I do agree with you that  there is lot to do here to make it better and
> more robust and also add supporting features that help to archive the
> annotations in their places before major document updates.

That one's easy: a snapshot represantation comprising both the document
and the annotations will suffice, and can be processed (eg with XSLT)
to whatever format you please.  It's exactly what Valet does to show
errors and warning in-situ in a document, for instance.

-- 
Nick Kew

Site Valet - the mark of Quality on the Web.
<URL:http://valet.webthing.com/>

Received on Monday, 18 March 2002 15:35:17 UTC