Re: algorithmic normative conformance requirements, design principles, etc.

Larry Masinter wrote:
> I was talking specifically about the question of whether
> image.width and image.height are simultaneously available,
> and the assertion that because some authors might depend on
> it

Not might, do.  If it were a might, the situation might have been 
somewhat different.

> the specification needed to require implementations
> to make them available

Right.  Implementations that want to correctly render existing webpages 
need to do this.

> and the inconsistency of that requirement
> with asynchronous processing.

Yep.  Sad, but true.  There are all sorts of requirements imposed on UAs 
and hence on the spec by legacy content; this is one of them.

-Boris