Re: Comments on PAGEAUTH 2.4: Lists

Proposed technique example to clear this up. (I had to think for a while
before I remembered why it was there too...)

When using lists, if the list is ordered and 
the size stated explicitly it will provide extra orientation to many
users, in particular people relying on screen readers, single line
braille displays, and very large magnification screens, even when there
is no inherent ordering in the content. For example:

We use staiontery of 3 types, each of which is divided into several
subcatergories:

1. Paper - 4 type
   1. Envelopes
   2. Notepaper
   3. Letterhead
   4. Poster paper
2. Pens - 2 types
   1. Blue writing pens
   2. whiteboard pens
3. fasteners - 3 types
   1. paper clips
   2. staples
   3. Big lengths of rope.

Provides more clues to navigation and to the implicit structure of the
list than if unordered lists were used.

charles

On Sun, 7 Mar 1999, Al Gilman wrote:

  At 09:11 AM 3/7/99 -0500, Ian Jacobs wrote:
  >Gerald Oskoboiny wrote:
  >
  >> 
  >> But this doesn't seem like a good example of an ordered list at
  >> all -- the items in these lists don't have a particular order to
  >> them (green, purple, and mauve aren't in that order for any
  >> particular reason.)
  >> 
  >> I suggest that this example be changed to use list items that
  >> really ought to be ordered. (like giving instructions on how
  >> to do something.)
  >
  >Yep, I agree.
  >
  
  There is more to this guideline than this exchange reveals.  I was shocked
  when I first encountered this guideline, because the fact that lists give
  the appearance of order to things that are semantically sets is one of my
  hot buttons.  But there is a reason for it.  And it goes beyond just those
  cases that are natural OL applications.
  
  Numbering of list items is a navigation aid for someone using a screen
  reader, which really treats the document as a sequence of lines.  Braille
  formatting uses hierarchical numbering of sections and not indenting to
  show structure for a similar reason.  So the guidelines need to expose this
  issue.  There is benefit to ordinal marking of lists for navigation for
  this group.  This is not a pure win-win deal; so the practice has to be
  applied with a grain of salt.  But there is some motivation to use OL over
  UL even when the underlying semantics indicate UL.
  
  On the other hand, this particular example is so dysfunctional that we had
  better come up with a better one if we want to pass the laugh test.  A long
  single-level list would be a better idea.
  
  Al
  

--Charles McCathieNevile            mailto:charles@w3.org
phone: +1 617 258 0992   http://www.w3.org/People/Charles
W3C Web Accessibility Initiative    http://www.w3.org/WAI
MIT/LCS  -  545 Technology sq., Cambridge MA, 02139,  USA

Received on Sunday, 7 March 1999 18:42:18 UTC