W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 1999

Re: AlertBox: The death of single-design pages?

From: Al Gilman <asgilman@iamdigex.net>
Date: Mon, 14 Jun 1999 13:24:31 -0400
Message-Id: <199906141719.NAA257439@relay.interim.iamworld.net>
To: w3c-wai-ig@w3.org
At 03:18 PM 6/14/99 +0100, Brian Kelly wrote:
>Further to my previous posting I'd like to give two examples of why the
>principles of universal design may not be applicable.
>Current guidelines state that web pages should be device independent, and
>that phrases such as "Click here" should be avoided (because some users may
>not be using a mouse-driven interface).  On the other hand, "click here" may
>be a good phrase for, say, someone with learning disabilities who does have
>a mouse, but has difficulties mastering a user interface.  This was brought
>to my attention recently when testing a web site designed for children - I
>found it difficult differentiating between images with just ALT text and
>images with ALT text which acted as links.

This is where the system of 'label' and 'description' proposed in the
Scalable Vector Graphics is easier to understand and use well than the
perplexing welter of constructs in HTML.

For others interested in continuing progress against the 'click here'
problem, review the 'fat links' notion we have talked about in the User
Agent working group.  Here the user agent would recognize that 'click here'
is not a discriminant for different link targets, and automatically back
off so that the minimal display to a screen reader or other peephole access
technology would include contextualizing material from the vicinity of the
link, such as the immediately enclosing list item or sentence.  If we put
in guardband rules for how the sensitive regions don't get too close
together, this is also beneficial for people with fine motor control problems.

PS: also fixes label-after-checkbox.  Make them a clickability atom.

But this doesn't mean the end of single-design pages.  What it means is
that the class of pages for which one can have a single design is not a
single layout but a single design of a polymorph, with an infrastructure of
semantic structure and a variety of interface incarnations.


>Similarly, "Return to home page" is a bad name for an anchor (the user may
>not have been to the home page).  However it might be a good name for an

>anchor generated on the fly, when the system knows the user have been to the

This link should identify whose home is the page linked to.  Don't assume
that the user has this notion in their recall memory.  Any page that can be
retrieved can be started at.  People don't think to make their pages robust
against the fact that anybody can start anywhere.  Given the popularity of
search engines as portals, many hits come from out of the blue, not from
the site's carefully laid out entry paths.

>"Click here" and "Return to home page" should be avoided in static HTML
>pages.  However they *may* be desirable in other circumstances.
>Brian Kelly, UK Web Focus
>UKOLN, University of Bath, BATH, England, BA2 7AY
>Email:  b.kelly@ukoln.ac.uk     URL:    http://www.ukoln.ac.uk/
>Homepage: http://www.ukoln.ac.uk/ukoln/staff/b.kelly.html
>Phone:  01225 323943            FAX:   01225 826838
>----- Original Message -----
>From: Brian Kelly <lisbk@ukoln.ac.uk>
>To: Kynn Bartlett <kynn-hwg@idyllmtn.com>; <w3c-wai-ig@w3.org>
>Sent: Monday, June 14, 1999 2:04 PM
>Subject: Re: AlertBox: The death of single-design pages?
>> >
>> > Jakob Nielsen's AlertBox column talks about web accessibility and the
>> > WCAG, and states the following:
>> >
>> > "But I am not sure that single-design pages will be able to deliver
>> >  optimal usability in the future. For example, screen sizes will soon
>> >  differ so drastically between high-end office workstations and small
>> >  mobile devices that the same pages will not satisfy both. And I also
>> >  think that one can make pages much more usable for blind users and
>> >  users with other disabilities by designing explicitly for these
>> >
>> > http://www.useit.com/alertbox/990613.html
>> >
>> > Agree or disagree?  To tie into a recent thread:  how does access
>> > for cognitively impaired users tie into this?  We've been saying that
>> > you don't need a separate web site for blind (et al) users, but --
>> > if content needs to be actually rewritten for a new audience in order
>> > to make it comprehensible -- is this the end of "writing once" and
>> > the beginning of specialized web sites for each type of disability?
>> >
>> > Your thoughts are welcome.
>> Hi Kynn
>>      I agree with Jakob's comments.
>>      I've become aware since attending the first day of the WAI meeting in
>> Toronto of two web accessibility communities.  One focuses on the needs to
>> make HTML accessible, the other is looking at developing a richer web in

>> future.  The former group gives priority to education, importance of ALT
>> tags, etc.  The latter regards HTML as on its way out, and priority should
>> be given to the new stuff (XML, XLink, RDF, etc.)
>>     To give an example.  The ALT tag provides a simple, one-dimensional
>> description of an image.  Multi-lingual support is missing (an obvious
>> barrier to understanding).  But ALT also  fails to provide information
>> may be needed to address cultural issues.
>>    I touched on this in my talk at the WAI meeting - see
>> 1.htm
>> As an example of the richer form of metadata I think is needed (e.g. for
>> images) see
>> 2.htm
>> I discussed these issues with Julie Howell of the RNIB (Royal National
>> Institute for the Blind) here in the UK who stated that:
>> "Rather than encouraging 'simplicity' in Web design ... we try to
>> 'flexibility', so that Web sites can be tailored to individual need
>> 'simply'. Flexibility affords the personalisation which people with sight
>> problems require."
>> Brian Kelly
Received on Monday, 14 June 1999 13:19:04 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:04 UTC