Re: Aural extensions

aloha, al et. al.!

how sadly ironic that the MS proposal should degrade so ungracefully when
rendered by lynx or any other text-based browser...  i can get realaudio
to stream through my desktop when browsing with lynx32, yet i can't get
consistently comprehensible aural output from a document automatically
converted into hypertext by Word97...

if MS is to convince me that they are serious about accessibility and
standards, then they will either bundle an HTML parser/validator with
their HTML generators, or at the very least, take the time to _validate_
the output of their automatic conversion utilities before posting,
mounting, or circulating any hypertext documents generated by such
utilities... checked against

for compliance with HTML 3.2 the MS proposal contained 199 errors...

and, while--for the most part--the actual body of the document was
comprehensible, the front-matter most decidedly is not--save for those
fortunate few who have access to a cell-by-cell capable browser...  in
particular, anyone listening to the proposal as rendered by lynx 2.5
and greater, would have heard the table-ized content as:

Version Author Date Change 0.9 Or Ben-Natan 6/6/97 Initial version

of course, this is the root of the problem which we are attempting to
solve through our participation in the WAI and its working groups...
but what of the user who will not reap the benefits of our work?  those
who, for whatever reason, physical and/or financial, have no choice but to 
use antiquated equipment and/or access the web via a shell account that
features an ancient version of lynx which doesn't even support lynx's
de-table-ization kludge?  while it is the purpose of this list to look 
forward, we must not forget that true accessibility looks as far 
backwards as it does forward...

what am i talking about?  perhaps a listen to/look at/feel of

will make my point more forcefully...  and, should Microsoft (and
its rivals in the GUI-based browser market) consider implementing an Aural
Accessibility Protocol/Patch, such as that outlined at:

which would (amongst other things) allow the user to de-table-ize
table-ized information, backwards-compatible access is extremely realizable...

by way of conclusion, i should stress that it is not my intention to
scapegoat Microsoft on this issue--invalid HTML is endemic to the
output of HTML authoring/conversion programs/utilities...  and, while
SoftQuad has taken steps to remedy this with the release of HoTMetaL 4.0,
HTML validation within authoring/conversion programs is still in its
infancy...  this, coupled with the ever-increasing popularity and
ubiquity of such authoring/conversion applications and utilities, presents
one of the most serious threats to an accessible internet...


Received on Wednesday, 22 October 1997 21:31:11 UTC