Re: IMG ALT attribute in HTML 4.0

Koga Youichirou (y-koga@ccs.mt.nec.co.jp)
Thu, 18 Sep 1997 13:26:10 +0900


To: www-html@w3.org
From: Koga Youichirou <y-koga@ccs.mt.nec.co.jp>
In-Reply-To: Your message of "Thu, 18 Sep 1997 12:28:23 +0900"
 <19970918122823D.y-koga@ccs.mt.nec.co.jp>
Message-Id: <19970918132610L.y-koga@ccs.mt.nec.co.jp>
Date: Thu, 18 Sep 1997 13:26:10 +0900
Subject: Re: IMG ALT attribute in HTML 4.0

My point is <IMG> which author doesn't specify alternate text has no
alternate text and user agents need not (probably `should not') show
any other alternate text other than there is image data.

> 	  3. Otherwise, if the included object contains text fields
...
> This means all user agents must GET image datas when they fail 1 &
> 2. And this requires user agents to analyze image datas too. Really? I
> think it's stupid.

Sorry, my using `must' and `requires' are my misunderstanding. But I
don't still think it's a good idea. I don't think many authors expect
that text in image data is treated as alternate text for image. And
there are some tools which create image data and set their own text as
a comment automatically, and then user agent will show nonsense
text. So I don't think user agent should extract text from image data
as its alternate text.

-- Koga Youichirou <y-koga@ccs.mt.nec.co.jp>