W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > April to June 1998

Re: OBJECT Element

From: Al Gilman <asgilman@access.digex.net>
Date: Wed, 3 Jun 1998 11:03:39 -0400 (EDT)
Message-Id: <199806031503.LAA21914@access2.digex.net>
To: jongund@staff.uiuc.edu (Jon Gunderson)
Cc: w3c-wai-pf@w3.org, w3c-wai-ua@w3.org
to follow up on what Jon Gunderson said:

> In our telecon it was discussed that the OBJECT element needs a LONGDESC
> attribute, just like the IMG element.  The main reason is that the user may
> want both the OBJECT and the LONGDESC information.  In the current
> structure the user can only have one or the other using the OBJECT nesting
> structure.   

I am not sure that the browser can't give the user some method to
access the alternate (i.e. the OBJECT content) even when the
OBJECT data are displayed as the initial state of the page.  It
seems to me that if the content between the start and stop tags
of the OBJECT element is present, the browser has the ability to
display it.  The question becomes under what circumstances.  Is
it an interaction, just a static mode, an optional feature
controlled by a static mode, or what?  In particular, if the
logic of this function is to offer an action only when there is
non-blank content in the OBJECT element, then in most cases where
the option intrudes on the screen, it will be of sufficient
interest so that the reader and author accept it.

There is a philosophy, here.  We discussed at length with the
HTML language team that we might have to take liberties or
introduce new strictness in the browser handling of markup when
the browser guidelines were written.  So we have a relatively
free hand in saying what browsers should do with what is in the
file.  But what goes in the file should be viewed as frozen,
unless we come to a real show-stopper.  I don't yet see that this
one reaches that level, because I think the browser can do
something reasonable with the markup as presently defined.

What is the current thinking about the user interface
implementation of LONGDESC?  How and when is the user advised
that this information is there for the asking?  The same kind of
methods applied to the LONGDESC display could be used to offer
access to the OBJECT content in the case when the OBJECT data are

> Another reason is that the OBJECT nesting will be used to
> provide alternative representations of the function and most time this will
> probably be: "get a browser that can render this object".  This will also
> provide a more consistant implementation of images since both IMG and
> OBJECT will have LONGDESC atttributes.

I am afraid that renaming the description opportunity isn't going
to alter people's attitude that much.  If that is what they are
going to put in the OBJECT content, instead of doing something like
what I outlined in



I don't think they would do any better in terms of populating a

Received on Wednesday, 3 June 1998 11:03:26 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:38:19 UTC