Re: Alternative Interfaces for Accessibility

Hi Scott,

I agree that in general serious mark-up languages are more powerful if 
they are developed on a subject by subject basis. On the other hand the 
Web succeeded in part (I believe) because HTML, while not very good at 
most things, was good enough (or nearly so) for a lot of things.

As you point out, now that we ahve XML and some more semantically rich 
possiblities for dealing with content on the Web we should be able to 
make things better (a small step at a time, anyway <grin/>).

One of the things relevant in this context is the XML Accessiblity 
Guidelines. This is a W3C working draft - http://www.w3.org/TR/xag - 
which describes how to make your subject-specific language accessible 
(at least that's the idea...). Among the requirements are that there 
should be default presentations for different modalities - for example 
speech output, text output (which is not quite the same use case), 
graphic output, different kinds of input, etc.

Comments on techniques that we could demonstrate for doing this, or on 
the document in general, are very welcome and will help to make it a 
better document. Although the draft is quite old it is still being 
worked on - as the editor I only have limited time, but always 
appreciate reviews and constructive suggestions about it (e.g. 
explanations of how to do something, or of why some requirement doesn't 
make sense or how it could be better expressed). In an ideal world I 
would like to publish a new draft in the next month or so.

Cheers

Charles McathieNevile

phoenixl wrote:

>Hi, Nick
>
>General mark-up languages need to be blind to different subjects.  It
>wouldn't make sense that the mark-up language would include the tags for
>representing the Krebs cycle and also the process of gender issues
>encrouging on ecology.  The Krebs cycled might need to be transformed in
>one way and gender issues in a different way because of the subject
>matter.  Subject specific transformation is different than mark-up
>transformation.
>
>
>The simulation is actually an interesting example of something that
>can't transform very readily.  The simulation is to act like a GUI for a
>database.  This means that it must be dynamic in its interactions while
>being displayed.  Since it is designed to act like a GUI, it needs to be
>able to respond to arrow keys.  How many browsers support dynamic use of
>arrow keys under the control of the web page?
>
>How would a mark-up language support transforming this type of web page
>without compromising the essence of the web page which is to act like a
>non-web page GUI?
>
>
>This link will get you by the first page on the demo web site and to the
>help page where the punctuation problems occurred:
>
>  http://members.aol.com/criptrip/command_control/command_demo_help.htm
>
>If you try to use'the link to the demo, it probably won't do much.
>
>
>
>Actually, databases and XML help with semantic issues and the semantic issues
>can often be handled via dynamically generated web pages.
>
>Scott
>
>  
>

Received on Friday, 11 April 2003 12:12:48 UTC