[whatwg] a few thoughts

[Andrew Fedoniouk]

----- Original Message ----- 
From: "Henri Sivonen" <hsivonen at iki.fi>
To: "Travis Miller" <greenie2600 at yahoo.com>
Cc: <whatwg at lists.whatwg.org>
Sent: Friday, July 06, 2007 11:37 AM
Subject: Re: [whatwg] a few thoughts


> On Jul 6, 2007, at 21:24, Travis Miller wrote:
>> New list member here. I'm excited to see the W3C moving toward 
>> standardized support for web apps.
>
> Welcome.
>
>> 1. Slider form controls:
>
> There's already <input type='range'>:
> http://www.whatwg.org/specs/web-forms/current-work/#range
>
>> 2. Expandable/collapsible hierarchical trees: Every modern  operating 
>> system of which I'm aware supports this UI element  natively, so it would 
>> be easy for user agents to implement, and  they are very well-suited for 
>> representing many kinds of data  structures.
>
> Does <datagrid> meet your needs?
> http://www.whatwg.org/specs/web-apps/current-work/#datagrid0

Why <datagrid>? I think <select> is just enough in this case as
<optgroup> can contain other <optgroup>s now.

Problem with <select>/<option> is that they do not allow
markup inside - only plain text. This limits its use significantly.

>
>> 3. Tabbed content:
>
> This was in the draft at some point but it was dropped. I'm not sure 
> about the details of the reasoning why.
>
> -- 
> Henri Sivonen
> hsivonen at iki.fi
> http://hsivonen.iki.fi/
>
>

About Tabbed content:

Tabs control is a binding of set of radio buttons (or labels that behave as 
radio buttons)
with the set of panels. Value of radio group is bound to the visibility of 
correspondent
panel. This schema works perfectly if stripe of labels and panels are 
independent DOM
elements. But when labels and panels are combined into single element like 
<tabs>
it is impossible to define reasonable layout using CSS in its current state.

This, btw, reminds me the question I wanted to ask while ago:

All elements introduced by html5 should appear in
"master" style sheet similar to this: http://www.w3.org/TR/CSS1#appendix-a
otherwise if they will use some non CSS definable layout we
 will end up with another <table>s situation.
There are few elements emerged already that are impossible to define in CSS 
terms.
<footer>  and <header> are examples that I can recall immediately.

So is the question: does this imply that HTML5 shall have synchronous CSS5 
thing
or not?

Andrew Fedoniouk.
http://terrainformatica.com