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

Re: talking about text alternatives Re: ALT as required attribute

From: Jim Ley <jim@jibbering.com>
Date: Fri, 7 Feb 2003 10:43:52 -0000
To: w3c-wai-ig@w3.org
Message-ID: <b202hc$pls$1@main.gmane.org>


"David Woolley" <david@djwhome.demon.co.uk> wrote in message
news:200302062042.h16Kgnx02727@djwhome.demon.co.uk...
>
> > Images as objects are badly specified though, since the UA cannot know
it's
> > an image until after it's performed an HTTP GET on the url, it can guess
it
>
> That's what the type parameter is for.

No, the type is indicitive, it's not definitive, the type parameter also
prevents content-negotiation on the best format of image to return, with
object a UA should surely send accept headers appropriate to all forms of
content it accepts, with image, it's reasonable to only send image/* types.

> > might be, but cannot know.
> >
> > Equally, what happens with an HTML document returned with a 404 status
code
>
> Because img isn't a first class link either, you lose the information
> with the 404.

Not necessarily, the 404 could have links to other locations for the
resource.

> > returned by the server, should the object fallback be used, or should
the
> > HTML document be used?  What about other status codes?  At least with
image,
>
> I would expect the fallback.

but that way you're throwing away the HTML document returned, and I don't
see how a UA can know what the author intended here.

Jim.
Received on Friday, 7 February 2003 05:44:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:08 GMT