W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > October to December 1999

Re: How Much Of A Problem Are Tables Used for Design?

From: T. V. Raman <tvraman@us.ibm.com>
Date: Tue, 16 Nov 1999 11:46:43 -0800 (PST)
Message-ID: <14385.46243.491582.172676@raman.almaden.ibm.com>
To: "Gregory J. Rosmaita" <unagi69@concentric.net>
Cc: WAI Interest Group Emailing List <w3c-wai-ig@w3.org>, User Agent Guidelines Emailing List <w3c-wai-ua@w3.org>
In addition to what Greg says:
It's not entirely clear whether decolumnizing complex tables
in some ad-hock manner 
as done by things like Lynx 
is indeed the road to making table-layout pages accessible.

Though decolumnizing does alleviate the pain of hearing
a significant part of the page's functionality is lost
because things that are meant to be in close physical
get spread out far apart.

Also, most layout-oriented sites only manage to turn into a
different form of gibberish when decolumnized in this way 
the issue gets especially bad with nested tables where the
decolumnizing gets even more ad-hock.

As an example, try 
the various yahoo sites e.g. broadcast, loans etc.

>>>>> "Gregory" == Gregory J Rosmaita <unagi69@concentric.net> writes:

    Gregory> aloha, kelly!  whilst i agree that the abuse of
    Gregory> tables for layout purposes has been mitigated
    Gregory> by advances in user agent and adaptive
    Gregory> technologies, i would caution against the quote
    Gregory> the tools at my disposal allow me to
    Gregory> decolumnize tables that are used for layout, so
    Gregory> how big an accessibility issue is the use of
    Gregory> tables for layout unquote train of thought...

    Gregory> why?

    Gregory> 1. expense: the tools you listed all represent
    Gregory> a significant investment on the part of
    Gregory> individuals who belong to a demographic
    Gregory> category (those with less than 20/200 vision)
    Gregory> which is plagued by an unemployment rate (in
    Gregory> the U.S.) of at least seventy percent (the 70%
    Gregory> figure i cited is derived from research
    Gregory> performed by the American Foundation for the
    Gregory> Blind)

    Gregory> adding this extra expense to what is already
    Gregory> (for most) the considerable investment that is
    Gregory> required to purchase a computer upon which the
    Gregory> software you listed will run efficiently and as
    Gregory> intended, constitutes -- at least in my mind --
    Gregory> an undue burden on the blind/VI user...

    Gregory> 2. internationalization: it often takes 6
    Gregory> months to well over a year for features that
    Gregory> are built into the English version of adaptive
    Gregory> software to be incorporated into non-English
    Gregory> versions of software...  therefore, just
    Gregory> because you and i (who happen, by accident of
    Gregory> birth, to reside in the U.S.A.) have access to
    Gregory> the cutting edge in both adaptive and
    Gregory> mainstream applications, does not mean that a
    Gregory> critical mass of other blind users have access
    Gregory> to the same advances...  just a single (and
    Gregory> well documented) case-in-point will suffice to
    Gregory> illustrate this point...  when Microsoft Active
    Gregory> Accessibility (MSAA) was first released,
    Gregory> applying it to a system which was running in
    Gregory> non-U.S. English mode caused the dialogs and
    Gregory> messages generated by the system to be output
    Gregory> in U.S.  English...

    Gregory> 3. mode of internet access: there are still a
    Gregory> lot of cybernauts who are using antiquated
    Gregory> technology with which to access web-based
    Gregory> content...  there are myriad reasons for
    Gregory> limitations upon an individuals ability (and
    Gregory> willingness) to switch his or her mode of
    Gregory> access, including:

    Gregory> A. steep learning curvesf

    Gregory> B. financial limitations

    Gregory> B1) shell access, where available, is often far
    Gregory> cheaper than a POP/PPP account, and is a more
    Gregory> cost efficient means of obtaining web-based
    Gregory> content when one is accessing the internet over
    Gregory> a phone line for which the user is being
    Gregory> charged per minute...  and, it is an
    Gregory> unfortunate fact of life that most of the users
    Gregory> who still use shell accounts are saddled with
    Gregory> an older version of Lynx or the W3 browser,
    Gregory> which may impose severe limitations upon the
    Gregory> content available to that user...

    Gregory> B2) many users with disabilities -- in
    Gregory> particular the blind -- may not have access to
    Gregory> computer systems that are capable of running
    Gregory> the type of software necessary to unravel
    Gregory> tables that have been used for layout purposes
    Gregory> -- a significant number of blind users are
    Gregory> using discarded or cast-off equipment that is
    Gregory> incapable of running much (if not all) of the
    Gregory> software you cited

    Gregory> 4) interoperability -- the use of tables for
    Gregory> layout is just plain wrong if you want your
    Gregory> pages to be as universally accessible as
    Gregory> possible...  tables are intended for the
    Gregory> display of tabular data, not for imposing the
    Gregory> desktop publishing paradigm upon the web, even
    Gregory> though that is what they are most commonly used
    Gregory> to do...  therefore, it is imperative that
    Gregory> authoring tools encourage the use of
    Gregory> stylesheets to control layout, and that
    Gregory> authoring tool developers and the users of
    Gregory> authoring tools place pressure upon UA
    Gregory> manufacturers to support stylesheets...

    Gregory> i could go on ad nauseam, but i think that the
    Gregory> reasons enumerated above suffice to illustrate
    Gregory> that, despite recent advances in technology,
    Gregory> the use of tables for layout is definitely
    Gregory> still a barrier to accessibility...

    Gregory> gregory.

    Gregory> At 10:11 AM 11/16/99 -0800, Kelly Ford wrote:
    >> There has been a great deal of discussion about the use of tables and the
    >> problems they cause for people who are blind when used for design purposes.
> I'm speaking strictly here of tables used for design, not to contain
>tabular data.
>I'm wondering how much of a problem such pages pose for folks who are
>blind?  The browsing combinations I use all decolumnize these types of
>pages in some fashion.  These include Lynx, Home Page Reader, Internet
>Explorer with either JFW 3.31 or Window-Eyes 3.1.

    Gregory> --------------------------------------------------------
    Gregory> He that lives on Hope, dies farting -- Benjamin
    Gregory> Franklin, Poor Richard's Almanack, 1763
    Gregory> --------------------------------------------------------
    Gregory> Gregory J. Rosmaita <unagi69@concentric.net>
    Gregory> WebMaster and Minister of Propaganda, VICUG NYC
    Gregory> <http://www.hicom.net/~oedipus/vicug/index.html>
    Gregory> --------------------------------------------------------

Best Regards,

IBM Research: Human Language Technologies
Phone:        1 (408) 927 2608
Email:        tvraman@us.ibm.com
WWW:          http://cs.cornell.edu/home/raman
PGP:          http://cs.cornell.edu/home/raman/raman.asc 
Snail:        IBM Almaden Research Center,
              650 Harry Road
              San Jose 95120
Received on Tuesday, 16 November 1999 14:42:03 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:38:24 UTC