W3C home > Mailing lists > Public > public-html@w3.org > July 2007

Re: ToolTips: bug or feature?

From: Sander Tekelenburg <st@isoc.nl>
Date: Mon, 30 Jul 2007 03:50:05 +0200
Message-Id: <p06240617c2d2f11f1fce@[192.168.0.101]>
To: public-html@w3.org

At 09:52 -0400 UTC, on 2007-07-29, Gregory J. Rosmaita wrote:

> lachlan wrote, quote:
>> The alt attribute is only rendered as a tooltip in IE, and
>> that's considered a bug for a variety of reasons. See this article.
>>
>> http://hixie.ch/advocacy/alt-tooltips
> unquote
>
> i find hixie's arguments, decidedly unconvincing.  there ARE valid
> reasons why @alt and @title should be exposed to the user

Agreed.

> onMouseOver

Disagreed :) How @alt is presented is up to the UA. A tooltip is nothing more
than one possible implementation[*]. If that implementation satisfies the
need for @alt as defined in the spec, that's fine. When it doesn't, the
implementation is broken. @alt marks not a tooltip, it marks a textual
equivalent. That's all.

[*] Lynx and iCab for Mac OS 9 for instance provide access to @alt, but not
through tooltips.

> -- as i indicated, what you perceive as a logical icon
> may not be understood by a sighted visitor from a vastly different
> culture, so a tooltip exposition of ALT text is actually beneficial
> to users of all stripes.

Providing easy access to the equivalent is beneficial. Whether that is done
through mechanism x or y is irrelevant, as long as it gets the job done.

> in the end, all hixie really addressed is
> author complaints

Well, I wouldn't dismiss the usability/accessibility problems he notes. But
yes, the article is about broken implementations (in UAs), not about the alt
attribute (in the HTML spec).


-- 
Sander Tekelenburg
The Web Repair Initiative: <http://webrepair.org/>
Received on Monday, 30 July 2007 02:00:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:02 GMT