[whatwg] Suggested changes to Web Forms 2.0, 2004-07-01 working

Ian Hickson ian at hixie.ch
Thu Aug 19 06:07:17 PDT 2004

On Thu, 22 Jul 2004, Jim Ley wrote:
>> My problem with this that now we have to overload the object element 
>> for every extension we want to make to HTML.  Ultimately everything 
>> becomes an object, which isn't very "semantic".
> Rather than the current situation where everything is an INPUT I don't 
> see the difference there.

"input" conveys that the data is a single field of data input.

"object" conveys that the data is from an external resource.

What we want here is a single field of data input, not an external 

> Also, degrading to IE6 is not a required long term strategy surely? So 
> at some point we can stop worrying about it and create a sensible new 
> model.

The "sensible new model" is presumably XForms.

If it wasn't for IE, we wouldn't have to even _consider_ abusing <object>, 
so I don't really see your point here anyway.

>> But I do agree that the more traditional HTML-like approach would have 
>> been not to try to force the existing input element to do everything by 
>> means of the type attribute.
> I much prefer it to overloading INPUT.

One possibility that has been risen (and which will be considered to WF3 
once we have more experience with the datetime controls in WF2) is to use 
a new element for the date/time controls. That would be preferable to the 
<object> solution.

But we shouldn't try to solve problems that aren't obviously problems. It 
isn't clear to me that the current date/time fields have inadequate 
fallback; I've found plenty of occurances of exactly the same UI in 
current forms, where it isn't fallback but intentional.

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

More information about the whatwg mailing list