W3C home > Mailing lists > Public > www-style@w3.org > February 2006

Re: [CSS3] tabbed(stacked) ...

From: Matthew Raymond <mattraymond@earthlink.net>
Date: Tue, 07 Feb 2006 14:40:36 -0500
Message-ID: <43E8F7B4.4020309@earthlink.net>
To: Laurens Holst <lholst@students.cs.uu.nl>
CC: "www-style@w3.org" <www-style@w3.org>

Laurens Holst wrote:
> Matthew Raymond schreef:
>>>> 3) It pollutes the URL.
>>>>       
>>> No it does not. This preserves the ability to link to a certain state 
>>> within a document (for free, it is inherent to the design), which is 
>>> quite a valueable merit.
>>>     
>>    That's somewhat true, but we need to distinguish between a
>> presentational state and linking to a specific part of the document. If
>> you're using an ID in the URL solely to trigger the :target
>> pseudo-class, they you're talking about presentation.
> 
> I don’t think it is different than any other fragment identifier. You 
> are linking to a specific clearly identified section of the document 
> (e.g. interface settings, auto-reply settings, personal information, 
> etc). Whether it is displayed as a tab is a presentational detail.

   The trouble is that true tabs are not hyperlinks. Tabs are more
closely related to the scroll bars that are controlled by the "overflow"
property. Think of how problematic it would be to have to simulate
"overflow: scroll" using URL tricks. While I do feel that tabs are
largely presentational, I don't feel they represent a document state any
more than your scrollbar position. Linking to a tab should probably
display it, but hyperlinks shouldn't be REQUIRED to display a tab.

   Let me make my position clear. I don't think using :target and
hyperlinks are an abuse of CSS and HTML. I just don't think they should
be used as an excuse for not specifying and implementing a better solution.

>> If the user
>> specifically wants to link to a certain element within the document,
>> then that's a navigational state. What was proposed was using the
>> document URL for presentational purposes.
> 
> Tabs aren’t navigational elements?

   No more than scroll bars are. Imagine for a moment that a web author
created a simulation scroll bar, where clicking on the scroll bar acts
as a hyperlink to the next section. Thus, if you want to view a section,
you must navigate via URL to that document fragment.

   Not fun, is it? People don't want to go to a different URL just to
browse a page. They just want to see the rest of the page. Similarly,
tabs are just a more finite way of scrolling through sections. It's a
distinction of navigating /through/ the document as opposed to
navigating /to/ the document fragment. If you don't understand that
distinction, remember that with hyperlinks, every tab click is going
into your browser history. Imagine the browser history if every section
in a document you scrolled over were entered into your browser history
and you'll get the idea.

   The bottom line is that we either need a widget or the ability to
style some non-URL-related state (like :selected perhaps?), both of
which would probably require some kind of markup-level support. (Unless
you want to make something similar to the CSS table model, but that may
be to complicated for the average web author to bother with.)
Received on Tuesday, 7 February 2006 20:56:19 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:43 GMT