W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2002

RE: Support for LONGDESC (hide or minimize by default)

From: Al Gilman <asgilman@iamdigex.net>
Date: Tue, 06 Aug 2002 09:41:55 -0400
Message-Id: <>
To: w3c-wai-ig@w3.org

At 10:08 AM 2002-08-06, Brian Kelly wrote:
>It's interesting in your guidelines that you say "Avoid use of popups".
>I did wonder whether browser vendors would implement longdesc support
>using some type of pop-up, along the lines of support for ALT and TITLE
>attributes.  Is it mandated in the user-agent guidelines how support
>should be implemented?

No, just that support should be implemented, and that giving the user this
sort of 'pull' or 'inspect' support qualifies as implementing access to this
content.  See particularly Checkpoint 2.3 in


The fact that the default is to hide or minimize the LONGDESC and only 
render it on further action by the user belongs in the format specification,
not the User Agent Guidelines.

The idea that this rule governing UA behavior is, however, just a default, 
that may be overridden in a given delivery context by policies controlled by 
the user, also belongs not only in the specification of this format but in 
an architectural framework for device independence that this and other formats 
conform to.  See the WAI comments on Device Independence at



>> -----Original Message-----
>> From: John Foliot - bytown internet 
>> [mailto:foliot@bytowninternet.com] 
>> Sent: 06 August 2002 12:34
>> To: Brian Kelly; w3c-wai-ig@w3.org
>> Subject: RE: Support for LONGDESC
>> Netscape 6.2, Mozilla 1.0 have a form of support for 
>> LONGDESC.  Cursor over the image, right click and select 
>> properties.  From the Properties dialogue box, choose 
>> "Description" if it is present.  See 
>> http://www.bytowninternet.com/examples/longdesc.html for an example.
>> The latest version of JAWS will also support LONGDESC by 
>> announcing the link to the LONGDESC file.
>> As for guidelines, I've not come across any specificly 
>> authored guidelines, but I recommend that developers keep the 
>> following in mind:
>>       - The majority of users accessing the LONGDESC file 
>> will not be interested in graphic content on the descriptor 
>> page; keep it simple and clean - avoid excessive navigational 
>> elements.
>>       - Avoid the use of Popups (this may seem self evident, 
>> but I've seen it happen via JavaScript and the "d" LINK)
>>       - Provide a link back to the referring image/page
>>       - When authoring the descriptor page, have an associate 
>> describe in detail what the image shows, without you looking 
>> at it. Record this.  That's your description! <grin>
>> Good Luck
>> JF 
>Thanks for this info, John.  I've used Mozilla 1.0 to look at your page,
>and it gives me a feel for how browser support for longdesc can be
>It's interesting in your guidelines that you say "Avoid use of popups".
>I did wonder whether browser vendors would implement longdesc support
>using some type of pop-up, along the lines of support for ALT and TITLE
>attributes.  Is it mandated in the user-agent guidelines how support
>should be implemented?
>It's not clear from your example, whether the "Return to referring page"
>link is generated by the browser or is in the longdesc file.  Looking at
>your longdesc file -
>I think the link is hardcoded - which will not be what's needed if you
>want a single description linked to from many images.  You could use
>some JavaScript to generate the link to the referrer page, or this could
>be provided by the browser's implementation of longdesc support - but
>you wouldn't want to do both.
>Also, as the longdesc file is handled as a standard HTML file using HTTP
>by the browser, it is liable to be indexed by indexing spiders, so
>potentially could be found as a result of a search query, without
>appropriate context, navigational bars, etc.  (I appreciate that it can
>be useful to find such information - I'm trying to think through the
>Thanks for the suggestion of getting a colleague to describe an image.
>Something I've been wondering about is how to provide multiple
>descriptions of an image for multiple purposes, using a unidirectional
>technique such as longdesc.  i.e. how can multiple language descriptions
>be provided, descriptions for different groups of users, etc.
>Thanks again for your comments.
>Brian Kelly
>UK Web Focus
>University of Bath 
>BA2 7AY
>Email: B.Kelly@ukoln.ac.uk
>Web: http://www.ukoln.ac.uk/
>Phone: 01225 38 3943
>> > -----Original Message-----
>> > From: w3c-wai-ig-request@w3.org [mailto:w3c-wai-ig-request@w3.org]On
>> > Behalf Of Brian Kelly
>> > Sent: August 6, 2002 6:47 AM
>> > To: w3c-wai-ig@w3.org
>> > Cc: b.kelly@ukoln.ac.uk
>> > Subject: Support for LONGDESC
>> >
>> >
>> >
>> > I'm giving a talk on web accessibility shortly.  I'd like 
>> to know the 
>> > status of mainstream browser support for LONGDESC.  I've 
>> searched the 
>> > WAI-IG list archives and used Google to search Web pages.  From my 
>> > searching my understanding is that LONGDESC is not supported by 
>> > mainstream browsers (cab on the Mac is an exception) - 
>> although if you 
>> > use XSLT you could reformulate LONGDESC to supported HTML element 
>> > (thanks to Kynn for this suggestion).
>> >
>> > Is this currently correct?
>> >
>> > Also what formats can the longdesc file be in?  The HTML 4 spec [1] 
>> > states that:
>> >
>> > longdesc = uri [CT]
>> > This attribute specifies a link to a long description of the image. 
>> > This description should supplement the short description provided 
>> > using the alt attribute.
>> >
>> > But doesn't say whether the file can be HTML, XHTML, plain 
>> text, etc. 
>> > (or that it shouldn't be PDF, GIF, etc.!), can contains JavaScript, 
>> > etc. Although the example give a file with a .html extension.
>> >
>> > If use of longdesc is needed to future-proof Web resources for when 
>> > browser support is available, are their any guidelines on using 
>> > longdesc?
>> >
>> >
>> > References
>> >
>> > 1 http://www.w3.org/TR/html4/struct/objects.html
>> >
>> >
>> > ---------------------------------------
>> > Brian Kelly
>> > UK Web Focus
>> > UKOLN
>> > University of Bath
>> > BATH
>> > BA2 7AY
>> > Email: B.Kelly@ukoln.ac.uk
>> > Web: http://www.ukoln.ac.uk/
>> > Phone: 01225 38 3943
>> >
Received on Tuesday, 6 August 2002 09:43:53 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:10 UTC