[whatwg] Implementation complexity with elements vs an attribute (responsive images)

[Mathew Marquis]

On May 13, 2012, at 7:03 PM, Boris Zbarsky <bzbarsky at MIT.EDU> wrote:

> On 5/13/12 3:20 PM, Mathew Marquis wrote:
>> I doubt any UAs will be forced to misinterpret common media queries because they haven’t been accounted for.
> 
> Opera has already been forced to do this.  For example, in its projection mode it matches both the "projection" and "screen" media queries (technically a spec violation) because of all the sites that explicitly say "screen" when they really mean "not print" (or in the absence of that query in downrev UAs when they mean "screen, projection, tv").  Now of course part of the problem here is that "screen", "projection", and "tv" are mutually exclusive, which is in retrospect silly.
> 
> This is an excellent example of the fundamental divide about optimism vs pessimism here: the things that web authors doubt UA vendors will be forced to do because of web authors making bogus assumptions are things that UA vendors have already been forced to do for years because web authors ... make bogus assumptions.
> 
> -Boris

You make an excellent case for standards bodies working more closely with developers during the implementation of things like media queries. With a better sense of the ways developers might misuse, misinterpret, or fail to understand standards, issues like you describe could potentially be better avoided. But I suppose we’ve gone off-topic, to an extent.