There are several challenges with RSS a11y. The most important is that
there is significant ambiguity in the RSS content model that makes it
difficult (at best) for software to determine automatically whether an
entry contains markup or plain text. Content is not explicitly labeled
in RSS like it is in Atom.
- James
Pete Brunet wrote:
What a11y features does Atom have that RSS doesn't?
*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
*James M Snell <[EMAIL PROTECTED]>*
Sent by: [EMAIL PROTECTED]
01/28/2008 11:26 PM
To
Pete Brunet/Austin/[EMAIL PROTECTED]
cc
[email protected]
Subject
Re: Atom and accessibility
Hey Pete,
Glad to see this :-). There are several features of Atom that have been
designed with accessibility in mind.
1. Explicitly typed text - title's, summary's, subtitle's, rights,
and content are all explicitly typed as either plain text, html,
xhtml, xml or some other media type
2. Required text content - title's are required for every entry and
textual content in the form of either an atom:summary or
atom:content element is required. It is possible for either of
these to be empty, but the elements themselves are required,
allowing applications to know explicitly whether or not text
content has been provided.
3. Language tags - The use of xml:lang attributes allow the language
for every piece of text to be explicitly declared.
4. Link titles - The atom:link element has an optional
language-sensitive title attribute that is a rough analog to the
html img tag's alt attribute.
5. Link href lang - The atom:link element specifies an hreflang
attribute that identifies the language of the referenced resource.
6. Link types - The atom:link element specifies a type attribute that
identifies the media type of the referenced resource
7. Accessible (x)html embedded in an atom text element or atom:content
should continue to remain accessible within the Atom feed.
There are a number of areas where accessibility is somewhat inadequate:
1. The atom:icon and atom:logo elements do not have title or type
attributes.
2. Despite requirements for the use of atom:title, atom:summary and
atom:content, it is possible for an entry to contain zero
human-readable text
It would be helpful if someone with a strong accessibility background
could run some tests on a corpus of atom feeds to see what accessibility
issues appear to be most common. As far as I am aware, no such analysis
has ever been done.
- James
Pete Brunet wrote:
>
> I am interesting in evaluating Atom to determine what the accessibility
> needs are. I'd limit this, at least for now, to enhancements to help
> blind screen reader users. I'd like to eventually develop a list of
> recommendations for improvement, e.g development of new technology or
> creating usage guidelines. My primary interest is in relation to the
> use of Atom in mashups, for example, a web page may start out being
> accessible by using W3C WCAG ( Web content Accessibility Guidelines, see
> http://www.w3.org/TR/WAI-WEBCONTENT/ ) but by incorporating content via
> Atom it may become inaccessible. If you've done any research in this
> area or if you have pointers to background material I'd like to hear
> from you. I am starting at ground zero with respect to Atom (but have
> been in the accessibility field for many years) so it's likely that no
> information will be too basic :-)
>
> Should further discussion related to accessibility take place on this
> list or atom-protocol?
>
> Thanks,
> *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
