W3C home > Mailing lists > Public > w3c-wai-hc@w3.org > July to September 1997

Re: Initial approach to the reading order issue (fwd)

From: Al Gilman <asgilman@access.digex.net>
Date: Tue, 30 Sep 1997 22:54:33 -2800 (EDT)
Message-Id: <199710010254.WAA19756@access5.digex.net>
To: w3c-wai-hc@w3.org (HC team)
Cc: w3c-wai-hc@w3.org
to follow up on what MegaZone said:
> This seems like a reasonable approach.  
> sequence {ID1 ID2 ID3}
> Contained in the style sheet you could easily have a different order 
> for differing media types. 
> The question I see is what to do with non-labelled elements?  Would you
> not present them?  I think that is extreme.  Present all of the sequenced
> elements in the given order, then the non-sequenced elements in the order
> in which they appear in the document?

I see this as an appropriate topic for browser creativity.

Given that we know the order may wish to vary based on the medium,
we should let those expert in the medium make the call.

An example of how one might want to handle this is speech is to
present things in HTML-text-appearance order until hitting the
element ID1 which starts the sequence constraint.  The the order
would be 

	element marked ID1
	  link to element ID4 textually between ID1 and ID2
	element marked ID2
	  link to element ID7 textually between ID2 and ID3
	element marked ID3
	element ID4

The elements subject to the continuous-text constraint would
be read in sequence once one of them was encountered, but the
places where other items had been deferred would be filled
with placeholders linking to where they appeared in the
presentation sequence.

That is just an example of a feasible policy with some thought to
minimizing disruption.

As I say, the HTML policy should be not to have a policy other
than that the browser should allow the user the choice to read
ID1, ID2, ID3 consecutively in the right order.

It's broken if you can't discover or follow the sensible order.
It's not necessary to fix it to make any global order a

-- Al
Received on Tuesday, 30 September 1997 22:54:51 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:35:00 UTC