[whatwg] Re: several messages

[Ian Hickson]

On Mon, 31 Jan 2005, James Graham wrote:
> > 
> > | <label for="d1">First Date:</label>
> > | <dateinput id="d1" name="d1" value="2005-01-30">
> > |  <select name="d1_month"><!-- Options --></select>
> > |  <select name="d1_day"><!-- Options --></select>
> > |  <select name="d1_year"><!-- Options --></select>
> > | </dateinput>  
> 
> I haven't been following all the discusion about date formats but, 
> subject to that proviso, this construct alone (without any inheritance 
> of attributes) seems to address the major concern that has been raised 
> about the datetime types (lack of a decent way of providing fallback). A 
> WF2 UA would simply display:none all children of the dateinput element. 

It has problems, as mentioned elsewhere in the thread:

 * It is easy for authors to not include any fallback, which makes it 
   worse than the <input> equivalent.

 * The fallback and non-fallback controls have different names.

 * The datetime types don't really need comprehenive fallback, given
   that the three cases that they could replace are:
     1. Text inputs, which would be improved, not hurt, by the new types,
     2. <select> controls, which do not need to be replaced at all, and
     3. Complex JS widgets, for which declarative fallback is not needed.

...not to mention the extra complexity and the implementation difficulty 
compared to just using a new "type".

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'