Re: ToolTips: bug or feature?

On Jul 29, 2007, at 23:51, Gregory J. Rosmaita wrote:

> the main point is, there should be no restriction on the means of
> exposing ALT and/or TITLE, other than the implementor's creativity
> and imagination, provided that user agents are capable of providing
> multiple options for their exposition as well as for ignoring them
> altogether...

I think your position makes sense for features that an user can opt  
to turn on but that are off as factory defaults. However, I think the  
factory defaults that a browsers ships with need to be much more  
prudent and should even cater to control freak *authors* instead of  
directly catering to *users*.

The browser factory defaults are what influence the most how authors  
perceive a feature. If the value of the alt attribute is shown as a  
tooltip almost always (in the spectrum of UAs that the author has  
cared to examine), the author will treat the attribute as meaning  
"tooltip augmenting the image"--not as meaning "textual alternative"-- 
and will write alt text that is inappropriate for use as an  
alternative. Or if the author thinks the tooltip interferes with his  
graphic design, he might even try to defeat the tooltip altogether  
and deprive all users of any alt text.

A couple of other examples:

Focus rings are very important for users who use a visual rendering  
with keyboard navigation. Yet, I think the users who need them would  
be better off if browsers had focus rings disabled by default and  
inconvenienced the users by requiring them to flip a switch to turn  
them on as part of the post-install configuration of the browser.  
When focus rings are on by default, graphic designers who think the  
focus ring is interfering with their art try to fight the focus ring.  
The author might try to use CSS to make the focus ring invisible or  
the author might use JavaScript to move the focus in a surprising  
way. Either way, users who need the focus ring are worse off it the  
author fights the focus ring. This was a particular problem with Mac  
IE 5. It used a focus ring that was consistent with the platform  
focus convention and the convention was more pronounced than the  
Windows convention which meant that graphic designers were even more  
distressed with the focus ring interfering with their art. Safari,  
smartly, makes full focus traversal a feature the user has to turn on  
in the preferences. This way user who need it get it, but graphic  
designers don't feel a need to fight the more pronounced Mac focus ring.

Back in the days of Mosaic, images that were also links were rendered  
with a link-colored border. In theory, this is good for the user,  
because it helps the user figure out which images are links. The  
problem is that virtually all authors as well as many users think  
that the border is horribly ugly. Therefore, as soon as HTML got a  
feature for suppressing the border, authors made getting rid of the  
border part of their routine. The default border is an annoyance for  
authors. Everyone feels the need to take care of getting rid of it.  
On the other hand, a user who'd prefer explicit link indication over  
aesthetics is worse off that he'd be in a hypothetical situation  
where borders could simply be enabled as a preference or with simple  
user CSS (that wouldn't need to counter all the ways authors use to  
suppress the border).

-- 
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/

Received on Monday, 30 July 2007 08:35:44 UTC