W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2003

Re: Longdesc attribute for images

From: <tina@greytower.net>
Date: Wed, 18 Jun 2003 15:58:02 +0200 (CEST)
Message-Id: <200306181358.h5IDw2f12761@localhost.localdomain>
To: w3c-wai-ig@w3.org

On 18 Jun, Lauke   PH wrote:

> I think you may have slightly misunderstood the original email
> from this thread. The problem is that, in theory, browsers should
> give users a simple, easy, automatic way of accessing the page

  I think we are rather talking about two diffent things: structure and
  browser design.

  The LONGDESC attribute is a standardized method for supplying further
  information regarding a graphical object. This link is a complement to
  the ALT-attribute, and goes to a description of the object in
  question.

  This description is, lets hope, in an accessible format. This
  information, provided by LONGDESC, is all to the good. It allows you,
  for instance, to write a detailed description of a graph, something
  which shouldn't go an in ALT-text. Whether a more accessible version
  of the data in the graph should be included is really another
  discussion.

  Now we come to the user-agent. Firstly, it is clear that the agent
  need present the URI in *some* form - preferably something distinct
  from the normal activation mechanism used for links (ref. HTML 4.01,
  chapter 13.2). A context-based menu like Mozilla's right-click one
  isn't such a bad idea. Lynx has one which I classificy as contextual
  through the use of the 'l' command. No, it doesn't list a link to the
  LONGDESC - but there is no reason at all why it couldn't.

  So: yes, user-agents should have some mechanism of offering the user a
  method of navigating to the content of the longdesc. Personally I find
  Mozilla's quite acceptable *for me*.

  But there are over 1,800 distinct user-agents out there. I have no
  *idea* which does what and how - and I refuse to speculate on it.

  However: this does not in any way reduce the value of the
  *information* contained in the LONGDESC attribute - it isn't too far
  out to postulate a speech browser which, upon finding a LONGDESC,
  actually loads the relevant information and reads that *instead* of
  the image.

  Just because only a few actually *uses* a certain type of structure or
  information is no reason to stamp it as 'useless'. The LINK mechanism
  is a case in point - even if it IS given a highly subjective 'better'
  way of activation in agents which support it.

-- 
 -    Tina Holmboe                    Greytower Technologies
   tina@greytower.net                http://www.greytower.net/
   [+46] 0708 557 905
Received on Wednesday, 18 June 2003 09:58:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:10 GMT