W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2005

RE 3.1 proposal - first half

From: Gregg Vanderheiden <gv@trace.wisc.edu>
Date: Thu, 26 May 2005 00:31:12 -0500
To: <w3c-wai-gl@w3.org>
Message-Id: <20050526053109.809EA60C14C@m18.spamarrest.com>

Hi John


Thanks for the very thoughtful summary of the 3.1 responses.  My comments to
your proposals are as follows


1)      I still believe that we should not have reporting requirements in
the guidelines again especially above level 3.  Having level 3 I think we
might consider, but above that it requires that in order to conform people
must provide conformance claims.  We do not anywhere require, for example,
that in order to conform you must post a conformance claim.  This is
something that many companies can not legally do.  For the same reason, we
can not require that they make this statement about education level, which
is in effect a mini conformance claim, which would be required at level one.
The problem is as follows.  If the company is required by law to conform to
the guidelines it would be instantly required to make this conformance claim
of education level.  This triggers ISO 9000 and other warranty legal issues
which a company can not engage in.  Because they do not post the educational
claim they are defacto instantly in nonconformance so no further testing
needs to be done to prove that they conform.  Since we require level one the
trap is complete.  The only place a "you must tell us" might be allowable w
ould be at level three because it is unlikely that level 3 items would be
required however, even here I would argue that it does not effect the
accessibility of the page only the ability to locate accessible material,
thus it is not an accessibility provision but a notification issue.


The one type of notification I think is appropriate is if there are multiple
entries to a website and only one of them is accessible.  In this case it
would be appropriate that it be required that it be possible for the person
visiting the site to determine which is the accessible entrance.  Thus, if
the website or a delivery unit only met the guidelines if approached in a
particular way then it must be possible to determine what that way is in an
accessible fashion.  Other than that we should not have any requirements in
the guidelines themselves for reporting for the reasons stated above.


Note: actually there is a way that companies could conform.  That would be
that they would just figure out what the highest possible reading levels of
their website would be and declare that for their whole site.  E.g.
"Specialized technical knowledge and advanced reading skills of a college
level or higher are required for some portions of this site.  This would be
equivalent to a such and such reading level of xxx."  Such a statement would
be safe could just be universally applied.  It would allow one to conform
but it would yield no practical benefit.  One of the tests we should do for
each one of our success criteria is to figure out whether or not the success
criteria are worded in such a fashion that they can be conformed to without
creating any useful accessibility benefit.




 "RE Education level"


I think education level may be a reasonable measure for difficulty of
reading. Like contrast and almost all of the other guidelines whatever we
come up with will not be a perfect measure for accessibility, readability,
visibility, etc.  However, it is the best measure we have that will
generally cause things to be more accessible along this dimension. 


Having a requirement at a lower level (for example level 3) that a version
of the information is available at a particular reading level I think may be
appropriate.  It could not be at level 1 and I do not think it should be at
level 2, at least not at this time in history.  Putting it at level 3 would
provide us an ability to put it in the guidelines as an optional measure
which could be practiced by site and advocate and others who are very
interested in reaching this audience.  The experience of this group could
then be used to develop the idea further and evaluate practicality as well
as implementations.  The uptake and conformance of these success criteria by
organizations that advocate or promote or serve this audience would also be
a useful model and indication.




RE Cascading Dictionary


It is true that standards very often follow commercial practice; however
there are some areas that until there is a standard there is not commercial
practice. Thus if we only implemented new HTML technologies and standards
that were already in practice then we would have standards that bifurcated
and trifurcated rapidly and did not innovate.  On the other hand, proposing
something in a standard that has not been indicated is different from
casting something in a standard which is not implemented.  Thus, we should
not put the cascading dictionary into the standard if there are no
implementations of it.  Toward that end I suggest that those interested in
exploring this approach form a subgroup and actually come up with three
implementations at three different websites to demonstrate practicality.  If
it is to be a level 1 guideline then it would have to be done on a major
site or a mechanism suitable for deployment at a major site would need to be
done.  In fact this would probably be a good test for any level 1 guideline.
If among our crew and others advocating web accessibility we can not come up
with three sites of any size which include one site of significance size,
and then we should seriously question its conclusion. By "significant size"
I do not mean mega site, it would not necessarily have to be a corporate
site.  Since this is already known in advance that accessibility is not a
corporate priority and proving that new accessibility provision is possible
which might later be a conflict of interest.  On the other hand, creating
requirements which even the large advocacy sites are not large web
accessibility proponents are not willing to meet should not exist at level




RE Wording of "Mechanism for locating definition"


There is a concern in many countries that it may not be easy or possible to
create an online dictionary because the countries borrow so heavily from
every other language, and do so in an informal fashion.  Other countries can
concatenate words routinely with new words without formally acknowledging
these new words. In both cases there may not be an online dictionary for the
language that is in common usage in the country.  Finally, there are
countries that are small enough that an online dictionary just does not
exist for their language.  Therefore suggest this guideline be re-written as
for those languages where a free and stable online dictionary of the
language exists. 


" Information is explicitly associated with the content sufficient to allow
the definition of all words used in the content to be found."


Received on Thursday, 26 May 2005 05:31:16 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 21:07:39 UTC