[whatwg] data: scheme in form submission

On Thu, 9 Dec 2004, Olav Junker Kj?r wrote:
> Ian Hickson wrote:
> > There is no use case, the section is merely defining what it does if it
> > is used, instead of leaving it undefined. I tried to cover all the
> > protocols that UAs commonly support these days.
> 
> Wouldnt it be better to leave it undefined (or just illegal) until 
> someone comes up with a use case where it would be useful to handle it 
> in a certain way? By chosing an more or less arabitrary way of handling 
> it now just for completeness, we may prevent some more useful idea which 
> might be invented in the future.

Well, the definition in the spec does what any submission system needs to 
do -- it gives the data to the target. I don't really see what other 
useful solution there could be.

I'd rather not leave things undefined, because at the end of the day 
people will just come and ask me what they should implement anyway, and 
then I'll have to come up with something which will end up implemented but 
not specified anywhere.

The definition in the spec now is not really contentious -- if we ever 
change our mind, we can do so pretty easily without breaking anything. It 
basically just says what to do in one case that would otherwise be illegal 
("%%" in a URI) and what to do in the case of a normal URI ("nothing").

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

Received on Thursday, 9 December 2004 14:52:41 UTC