Earlier I asked about alternative media could be provided for a11y use. 
Brian Smith suggested 3 alternatives: 

1) The media could have alternatives embedded 
2) There could be n number of links in the feed to alternatives, each with 
a new rel value, e.g. rel="large-print", rel="blind", rel="deaf", 
rel="transcript", rel="captioned".
3) Put alternative text in the atom:summary field. 

Option 1, using something like SMIL (or Flash or Quicktime or some format 
that supports embedding), is a good solution, but totally separate forms 
of content might be preferable for some publishers. 
Option 2 might not be acceptable since it seems like this community feels 
it's best to not have too many possible values for rel.
Option 3 would be OK for a fairly short description but wouldn't be 
suitable for alternative media

Most a11y architectures provide for a short description like the title, a 
longer description like the summary, and then the content itself.  Often 
the title and summary will be good enough.  But often an alternative file 
is needed.

I can think of these ways to provide the alternatives:
1) Option 2 above, i.e. links with new rel values
2) Use n links with rel="related" and let the reader sort it out via the 
MIME type.
3) Handle the whole thing inside the content element with some form of 
foreign XML.

Does anyone have initial thoughts on those three options?

I'm looking into application to Atom of a system developed by the IMS 
Global Learning Consortium called Access For All.  It's standardized both 
by IMS/GLC and ISO/IEC.  It provides for users needs and preferences to be 
defined via one spec and media meta data to be defined in a companion 
spec.  Systems (non-Atom based) exist that use these two specs and match 
the user preferences to the meta data choosing the appropriate alternative 
resource based on the user capabilities. 

The handling of the user preferences and the matching would have to be in 
a feed reader or a mashup handler.  The media (or links to the media) and 
resource meta data would have to be passed through Atom to the feed reader 
or mashup client.

Based on that short description does anyone have a feeling how to best 
handle the transmittal of the metadata and media?

I will be meeting with someone that's implemented Access For All to 
understand how this is done in a system called TILE, "The Inclusive 
Learning Exchange".  See http://inclusivelearning.ca/  Hopefully this will 
help me understand better how A4A could be applied in an Atom environment.

Some links:

A4A specs from IMS/GLC:
The preferences spec is call AccLIP (Learner Info Package), the resource 
meta data spec is called AccMD
http://www.imsglobal.org/accessibility/index.html

ISO equivalents: 
The preferences spec is called PNP (Personal Needs and Preferences) an the 
meta data spec is called DRD (Digital Resource Description)
Overview:  http://jtc1sc36.org/doc/36N1139.pdf
PNP:  http://jtc1sc36.org/doc/36N1140.pdf 
DRD:  http://jtc1sc36.org/doc/36N1141.pdf

Pete Brunet
                                                                          
IBM Accessibility Architecture and Development
11501 Burnet Road, MS 9022E004, Austin, TX 78758
Voice: (512) 838-4594, Cell: (512) 689-4155
Ionosphere: WS4G

Reply via email to