W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2000

Re: Accessibility barrier?

From: Al Gilman <asgilman@iamdigex.net>
Date: Thu, 13 Jul 2000 13:58:39 -0400
Message-Id: <Version.32.20000713125255.04175b90@pop.iamdigex.net>
To: Melinda Morris-Black <melinda@ink.org>, Accessibility Listserve <w3c-wai-ig@w3.org>
** summary

One general strategy for situations where you have a policy mandate to
support browsers "below level 4" is to make the style-sheet-free HTML a
reasonable transliteration of the standard legal styling, and have the
stylesheets override the formatting properties set in the HTML in browsers
that implement the stylesheet, to produce as nearly as possible the exact
repertory of effects that is traditional in the legal-document
typographical practices.

** details follow 

At 09:01 AM 2000-07-13 -0500, Melinda Morris-Black wrote:
>We are currently assembling technical requirements for our website
>redesign. One of my goals to move to a XHTML 1.0 and stylesheet solution
>for greater accessibility. I don't anticipate any difficulty for the
>majority of our portal, with one exception: We display bills for our
>state legislature which require strike through text. I'm required that
>these display properly, due to the legal nature of the content.

You have to do what your boss says.  But we also need to push back on this
point.  We need to get Cynthia Waddell on this issue.  The legal community
needs to understand the alternative of having a semantic class for
deletions (there is HTML for this) and that overstrike is only one of the
ways that it can be distinguished.  The requirement to distinguish
inserted, deleted, etc. text in markups of bills is valid and needs to be
supported.  But we need to get the legal brains to understand that
distinguishing these things in ACSS is as valid as with traditional print
text effects.  And that they need to define and strive to control what it
is that the reader needs to understand about this text, not what they need
to see.

In particular you can plead that you have conflicting instructions in the
area of "follow the markup presentation rules" and "serve the rural Kansans
with old technology."  You didn't create this problem, you are just seeking
guidance on how to resolve it (from the policy authorities who stuck you
with this dilemma).

>This will be no problem for newer browsers that recognize stylesheets.
>However, strike through will not degrade gracefully.  I still have a
>population using older browsers and don't anticipate any significant
>upgrades in the near future. Many are dialing in from rural Kansas.
>Any suggestions for this conundrum? We may have to continue marking up
>those documents for years to come otherwise.

First off, you have to audit the print style instructions and come up with
good names for what the distinctions are that are being made.  Hopefully
these will not be too many.  The spectrum of semantic categories determines
how you approach the next step.

Then you need to decide if your floor of HTML usage is 2.0, 3.2, or what.

Mark up the different semantic categories with text effect including bold,
italic, font changes, etc. using the deprecated HTML attributes until all
the distinctions are distinguished (and not by color alone).  Also use
class attributes in this markup to name the semantic category that is the
reason for the text effects.  

Use HTML 4.0 Transitional as your DOCTYPE.

The stylesheets keying of the CLASS attributes should be able to override
the formatting properties set in the HTML elements.

This will probably take some tweaking to get to something that will
actually work given the random things that don't work in this or that
browser.  But if you have a State policy mandate to support browsers prior
to Level 4, you should lean in this direction.

This approach is speculation on my part; I haven't done it or have a
reference site done this way to point you to.  But in theory it should
work.  If it doesn't work, I need to know so I stop telling people to do this.

Note also the artful use of browser sniffing and stylesheet adaptation at
the Trace Site.  You may wish to use some such technique if you want to use
styles for things outside the consensus core of effects that are actually
implemented the same in more than one browser.

** another approach

..you may want to look into is PDF for the traditional-look version.  In
this case you have to be careful to test out the full production path to
both the look-right and access-right equivalents.


>Information Architect
>Information Networks of Kansas
>FON: (785) 296-5143
>PCS: (785) 550-7345
>FAX: (785) 296-5563
Received on Thursday, 13 July 2000 13:56:59 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:35:57 UTC