W3C home > Mailing lists > Public > public-html@w3.org > April 2008

Re: [whatwg] several messages about the HTML syntax

From: Leif Halvard Silli <lhs@malform.no>
Date: Mon, 28 Apr 2008 20:46:18 +0200
Message-ID: <48161B7A.9010805@malform.no>
To: "Olivier G." <olivier.gendrin@gmail.com>
CC: Maurice <maurice@thymeonline.com>, HTML WG <public-html@w3.org>

Olivier G. 2008-04-28 20.31:
> Maurice a écrit :
>> On Apr 28, 2008, at 3:46 AM, Leif Halvard Silli wrote:
>>> Maurice 2008-04-28 04.10:
>>>> I think longdesc should be expanded to all attributes.
>>>
>>> Did you mean to say that @longdesc should be expanded to all 
>>> elements? Or to all <a href> elements? (In the code examples you 
>>> only added it to the <a href> element.)
>>
>> I think it  would be useful on all _elements_.  (if implemented 
>> visually the way I want )
[...]
>
> I don't think it would be an efficient way to handle tool tips, 
> because longdesc is just a URI pointing to a whole new webpage 
> (excepted for data URI), so it would request from the browser to GET 
> it, parse it analyse charset, maybe CSS and javascript of that new 
> page, then display it in a sort of absolutly positionned iframe... 
> Seems overly complicated to me.

What you say may hold true when the @longdesc do point to another page. 
But as Maurice himself mentioned, the @longdesc do absolutely not need 
to point to another page.

And as long as the @longdesc points to a resource on the same page, you 
also do not need to hack around with inconvenient iframes.

All you need is to hide the element which the @longdesc points to, and 
then to make it visible - for instance via the :target selector.

I have made a demonstration of this, showing how to do it for 
traditional @longdescs on IMG elements. [1]

[1] http://www.malform.no/acidlongdesctest
-- 
leif halvard silli
Received on Monday, 28 April 2008 18:47:09 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:54 UTC