RE: Accessible tables

Hello,

You are correct. I used the term compliant, when I meant in accordance with
accessibility guidelines.

It is a straightforward matter to identify if a particular HTML coding
standard is followed strictly or loosely. I have often instructed others
that 80 to 90% of the battle for accessibility is in just writing good code
compliant with W3C standards.

To the extent that accessibility guidelines are adopted as standards and
even laws, it is almost as straightforward to apply such techniques. I do
believe judgement calls come in to play in the sense that a part of
accessibility is general useability.

To the extent that W3C guidelines and standards are not fully implemented
within mainstream browsers, or are rendered differently within each of them,
I do think universal accessibility becomes less straightforward and more of
an ideal.

I get paid to produce results, pages that work, i.e. pages that display in
accordance with style standards we have established. How I do that is up to
me. As long as it looks good and is accessible and easy to use. I write HTML
4.0 loose code.

Ideally, I would incorporate elements and attributes within all my coding
that are suggested and even required by W3C standards and guidelines. But
some of them don't work in the browser that my end users have to use. Others
are basically redundant because my screen reader already provides me that
information or functionality. And I don't have the time for redundant
coding.

I am not arguing against universal accessibility. I think the attitudes
expressed on this list are wonderfully refreshing. I *think* I am trying to
say that even when time constraints, lack of resources, and other business
limitations do not allow for full application of all standards and
guidelines, a realistic approach can still result in pages that are quite
accessible.

Brian

-----Original Message-----
From: John Foliot - Another 4:00 AM Web Thing [mailto:foliot(at)fouram.com]
Sent: Tuesday, February 26, 2002 8:29 PM
To: brian walker
Cc: aware-techniques(at)hwg.org
Subject: RE: Accessible tables


Brian,

As an end user who deals with these issues on a daily basis, first thanks
for the honest input.  I would only add that compliant code is not the same
as totally accessible code... at least I don't think it is.  Compliant code
is code that passes an HTML validator (parser).  The validator parses the
code and determines whether it is accurate or not.  This is a black and
white process, and IMHO not an intellectual exercise, but rather an absolute
neccessity.  Without standards, we can never be sure that what we are
developing will work with what everybody else is using.  For a very
interesting article on the power of standards may I recommend the following
from Wired Magazine:
(http://www.wired.com/wired/archive/10.01/standards.html) It has very little
to do with web pages, but a whole bunch about the importance of standards.

When talking about creating accessible web pages (sites) I always stress
that the easiest thing any developer can do right off is ensure that the
page validates.  It's only one step of many, but because it is a simple B&W
process there is no judgement calls neccessary.  It's right or it's wrong -
Period.  And don't forget, accessibility is more than just making sites
available to the visually and mobility impaired... it means accessible to
all HTML rendering agents, be they web browsers, PDAs, or next years
toaster.  Once you have achieved a valid page, then you can proceed to
ensure that all of the extra attributes and features which make pages _more_
accessible are there.  Skip Navigation, LONGDESC, Summary, Title, even Label
are nice to have and certainly aid in the navigation of a site (and I'm not
advicating not to include them), but if the page is properly built in the
first place (ie: it validates) than even without the enhancements I'm sure
you could probably deal with the page/site in question (the one caveat is
with Client Side scripting, where valid but "insensitive" code can still
make a site totally inaccessible).

If however you were say compliant in referrence to 508 compliance, well,
that's not mental bubble gum either, it's the US law.  Here in Canada, the
bar is set even higher for Federal web sites, they MUST comply to WAI
Priority 1 AND 2 Guidelines.  And (some) developers are acheiving it,
although due to poor or total lack of proper training not without much teeth
gnashing and hair pulling <grin>.

As always, JMHO


JF






> -----Original Message-----
> From: owner-aware-techniques(at)hwg.org
> [mailto:owner-aware-techniques(at)hwg.org]On Behalf Of brian walker
> Sent: February 26, 2002 7:12 PM
> To: Gerhard Schoening
> Cc: aware-techniques(at)hwg.org
> Subject: RE: Accessible tables
>
>
> Hello,
>
> I think you should follow W3C guidelines, as close as possible. I
> have never
> tried using an empty summary attribute.
>
> I know how the screen reader I use responds, and since I write
> the help for
> it, I code for it. Since it is Microsoft HTML help, CHM format, I
> only have
> to take into account IE - which is also the recommended browser
> for use with
> my screen reader. So I am saying my perspective is rather
> restricted. I only
> have to code for one browser, and my primary focus is only one disability
> group.
>
> I am a pragmatist. Writing perfectly compliant code seems to me an
> intellectual excercise seldom applicable in the real world. In the real
> world, there are stiff deadlines, resource limitations, and statistical
> information on the audience who will use a site. I think accessible web
> design is essential, but is most effective when the emphasis it receives
> reflects the reality of the situation.
>
> I think no matter how much we wish to strive towards universal
> accessibility, these factors do come into play.
>
> If I were writing a commercially accessed web site, I would code
> for the big
> three browsers - market share wise - using JavaScript to sniff out the
> browser in use and apply the best external style sheet I could devise for
> it. Then I would make sure that with no JavaScript and no style sheet, the
> site was still useable.  And it probably still would not be totally
> accessible in Lynx, though I would try.
>
> Brian
>
> -----Original Message-----
> From: Gerhard Schoening [mailto:g.sch(at)onlinehome.de]
> Sent: Tuesday, February 26, 2002 6:36 PM
> To: brian walker
> Cc: aware-techniques(at)hwg.org
> Subject: Re: Accessible tables
>
>
> Hello Brian,
>
> On Tuesday, February 26, 2002 at 23:58 you wrote:
>
> > As a screen reader user, I personally do not want layout tables
> identified
> > as such. It is just extra verbiage.
>
> > The table specific commands available through my screen reader are
> > intentionally not functional in tables with only one row. This
> works very
> > well.
>
> > I do not use summary attributes in the tables I create unless they are:
>
> > 1. data tables
> > 2. complex enough that a summary truly adds value, i.e. makes the table
> > easier to understand.
>
> > I may be blind, but I do not think it is that difficult to figure out.
>
> > However, I recognize that mileage may vary for others.
>
> > Brian
>
>
> Does that mean I should use an EMPTY summary attribute for non-data
> tables, just like the empty alt attribute for non-relevant images
> (like a spacer.gif)?
>
>
> Gerhard
>
>
>
> http://www.cp-web.com
> gerhard(at)cp-web.com
>
> http://www.schoening-online.com
> gerhard(at)schoening-online.com
>
>

HWG: hwg-basics mailing list archives, maintained by Webmasters @ IWA