Current desktop detection / app access

[Dave Cridland]

On Thu Aug 12 14:58:10 2004, Jody Goldberg wrote:
> There are some hidden gottchas to be wary of.  Different mailers
> support different pieces of the spec.
> 
> eg
>     mutt will not accept a mailto without a destination address.

Yes, I'm aware that not everyone follows standards. Sadly, if it 
comes to a point where I'm forced to choose between supporting broken 
implementations or supporting a standard, I'm forced to support the 
standard.

The original mailto spec had no parameters, and just a simple email 
address, as far as I recall.

So it'd be highly recommended to avoid the use of the more esoteric 
forms unless you really need to use them. I'd therefore suggest that 
you'd put a single recipient as:

mailto:foo at example.com

but multiple ones as parameters, to force non-compliant MUAs to error 
if they don't understand the parameters. Hopefully, anyway.

My point was that defining an entirely new standard would be a waste 
of Michael's time, when he could define a simple addition to the 
existing one, which would hopefully reduce non-interoperability.

Dave.