W3C home > Mailing lists > Public > www-style@w3.org > June 2015

Re: [mediaqueries] undefined behavior for the resolution media query

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Mon, 22 Jun 2015 14:11:58 -0700
Message-ID: <CAAWBYDC4YNieFZvyU-m6fHiCeXO7P=XU7KXRRjBU1zwd2x7w-A@mail.gmail.com>
To: Florian Rivoal <florian@rivoal.net>
Cc: www-style list <www-style@w3.org>
On Thu, Jun 18, 2015 at 4:33 AM, Florian Rivoal <florian@rivoal.net> wrote:
> After thinking again, now with the naming cleared up, I guess what was confusing me with your solution when trying to write sample media queries for various scenarios was not so much the way you split it between resolution and image-resolution, but rather the fact that you had a none value rather than an infinity value. infinity plays a lot nicer with inequalities:
> Given an vector output with no pixels, this:
>   @media (resolution >= 2ddpx) { }
> is a lot nicer than that:
>   @media (resolution >= 2ddpx) or (resolution: none) { }
> And also, the fact that this:
>   @media (resolution: none) { /* vector output */ }
> would be different from that:
>   @media (not (resolution)) { /* vector output or non graphical medium */ }
> is lousy.
> If you're ok with infinity rather than none, I'm ok with your "solution 4" (as long as we use appropriate naming given the semantics, like downsample-resolution or image-resolution, not rasterize).
> Deal?

I'd intended for "none" to mean infinite; I was using "none" to allow
for an easy way to write a query for vectors.  But after some thought,
your examples show that that's clumsy; an explicit "infinity" value
that's defined to be larger than any length (and is truthy) is better.

Received on Monday, 22 June 2015 21:12:47 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:55 UTC