W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2004

Re: Skip Sections

From: David Woolley <david@djwhome.demon.co.uk>
Date: Tue, 30 Mar 2004 07:00:51 +0100 (BST)
Message-Id: <200403300600.i2U60p902569@djwhome.demon.co.uk>
To: w3c-wai-ig@w3.org

> Thinking back to my design for print days, there was no need to insert a
> sub-story into the main story flow in Quark or PageMaker in order to create
> the text-wrapped display that float replicates. The two stories remained
> whole and separate regardless of how they were streamed onto the page.

PDF, since long before tagged PDF, has had the ability to represent this
sort of structure, even if I've never been aware of it being used in
anger.  Each contigous chunk is called a bead and and the whole 
article a thread.  I don't know how well the reader handles this as
I've never found such a document.

> The solution would need to be implementing the same sort of sophisticated
> box and story handling models in CSS that print designers enjoy in DTP

Whilst there are moves towards intrinsic multi-column layout, there are
a couple of problems that CSS has that DTP doesn't have:

- it has incremental rendering with minimal backtracking as a desirable
  design aim (even if most web pages use auto table layout, that 
  frustrates this);

- it doesn't have the human layout designer there to tweak things to 
  work well, which is necessary to avoid completely empty columns, etc.
  if a radically different font size is used.

I more and more think that the tagged PDF model is a better model for 
what most designers want to do and they way they seem to think; you 
create a hard page description in presentational form, then annotate
it with the structure, rather than creating the structure and annotating
it with soft layout rules.  Of course, with tagged PDF, if you are
not in the primary audience, that can appreciate the graphic design,
you are at the mercy of the tagger.
Received on Tuesday, 30 March 2004 02:36:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:27 UTC