Re: Indexing

Hmmm,

I am notusre that metadata is yet at the accessibility requirement stage. So
I will waffle briefly on the topic. For those uninterested, stop here <grin/>

There are requirements to provide access to various parts of content. There
is not, as Kynn pointed out, an explicit requirement in WCAG to identify how
accessibility is achieved, which is one area where metadata woulod be an
extremely powerful technique. It is possible that the requirement should be
there, but in many cases it can also be done simply by using markup according
to specification (for example, the alt attribute on the img element is
explicitly described as an alternative, as is longdesc. the title attribute
on an abbreviation is a bit less clear from the HTML 4.0 spec...).

Providing access to content via a site map, structured content, etc, can all
be achieved by using metadata, and some of the really cool applications are
things like automatic site maps, and navigation that builds links by hunting
metadata from around the web. But this is a technique not a requirement.

I think what will finally make the difference is implementation (in this as
in so many facets of accessibility). If every browser for the last year had
come out with a metadata parser and renderer, then it might be the only
sensible way to do this. But in the current web, metadata-based solutions are
often not sufficiently compatible with the browsers available, as far as I
can tell. (Personally I am looking forward to the day when that is not
true...)

cheers

Charles


On Sat, 23 Dec 2000, William Loughborough wrote:

  I've been using the word "indexing" a lot and tying it in with RDF and
  trying to get something that furthers it into the guidelines.

  Whenever I get a (non-fiction) book about something I have any familiarity
  with I look first in the index to learn a lot almost instantly/painlessly -
  even before the Table of Contents. Occasionally I consult bibliography,
  seldom get deep into the notes (whether they're on the referring page or in
  a collection at the end), and almost always go really fast through the
  acknowledgements.

  All of these items (well, maybe not the footnotes) are important enough
  metadata (information about the content of the book) that it is extremely
  rare that any of them is missing. Edition #, copyright information, notes
  about the formatting/typography/+ are usually of more interest to
  publishing specialists. At any rate all of this stuff is taken for granted
  in books (and in fact in magazines and newspapers) so it should come as no
  surprise that although the Web is a completely different medium, because of
  the fact that it (in one of its major guises) presents not only information
  but information about the information, its parallels with certain forms of
  print media is valid.

  Without that metadata a book is very poor. Same for Web Content.

  Does this have to do with our scope/charter/accessibility/+?

  IMO it is absolutely central to accessibility for PWDs (Persons with
  disability) - as well as TABs (Temporarily able bodied). The ultimate
  "electronic curb-cut" might well be that Content on the Web, by becoming
  accessibile to our friends/clients, also becomes *much* more usable by
  everybody.

  What I am proposing is a means of
  enhancing/recommending/requiring/educating/+ the use of adequate indexing
  in all Web Content at a P1 level. Because Charles so timely stated
  "Objectivity is also a subjective concept" then the idea of a "barrier"
  making the Web beyond merely "difficult" to use - bordering on impossible -
  is clearly tenable from the POV of getting to the information strewn about
  on the shoulders of the infobahn.

  Before even attempting to use a Website, one must find out if it's in one's
  native language, whether it's accessible (and whether its claim of
  conformance is testable), and if there is any information about its content
  given,and if so, what is its nature.

  These sound like rather heavy demands at first until one examines parallel
  experiential histories undergone before entering into a book acquisition. A
  great many books are useless without indices and for PWDs with certain
  conditions they are useless without, e.g. claimed-to-be-useful
  illustrations, and other features that rise above merely useful to
  mandatory. The starting notion is that every element of a site ("element"
  in both the sense that term is used in markup languages and as used in
  everyday speech) must be revealed/referenceable/indexed.

  The techniques for this are very similar to the familiar ones from other
  media: indexing; tables of contents; "Dublin Core-like" stuff. The
  guideline is quite abstract/general though not-yet-written but why write it
  if we don't agree on its principle: access to Web Content requires
  information *about* the content to be readily available without the need to
  actually "buy the book".

  Anybody care to join me?



  --
  Love.
                   ACCESSIBILITY IS RIGHT - NOT PRIVILEGE


-- 
Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative                      http://www.w3.org/WAI
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia
until 6 January 2001 at:
W3C INRIA, 2004 Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France

Received on Saturday, 23 December 2000 14:15:37 UTC