W3C home > Mailing lists > Public > www-style@w3.org > July 2013

RE: [selector-profiles] confusion

From: François REMY <francois.remy.dev@outlook.com>
Date: Thu, 11 Jul 2013 22:55:22 +0200
Message-ID: <DUB120-W2382934091EC3449022E9CA57B0@phx.gbl>
To: Sylvain Galineau <galineau@adobe.com>
CC: "www-style@w3.org" <www-style@w3.org>
>>The one I could think of however is:
>> !label /for/ input:focus {
>> ... focus style for the labels of the focused input ...
>> }
>>This is only working on the complete profile right now for multiple
>>reasons but authors can accept that the styling of the label may be a bit
>>delayed after the focus change, that's not a big deal.
> If you'll pardon my French - so to speak - use-cases are the shit. In this
> case, whether authors 'can' accept a delay seems rather hugely dependent
> on what 'a bit' is. To go to an extreme for argument sake, if the delay
> nearly a second and makes transitions or other animations look bad I'm
> doubtful. More fleshed out ideas or mock-ups would definitely help in
> figuring out whether and how any specific proposal might help.

See my previous mail:

   @defer up-to 250ms { ... }

Actually, the author is in full control on how often he [want] the selector to be reevaluateed. Beyond that critical time, the browser is still allowed to reevaluate more often if he believes he can do it safely without degrading the user experience (or less often if it cannot even render at the expected frame rate).

> More seriously, your argument implies the author can
> easily figure out which rules are not worth keeping up to date. Given the
> wide range of devices and contexts the content may be run in, this sounds
> very dicey. 

This is basically an opt-in. Except if they want to use "static" selectors which will not work outside the async bubble (@delayed/@defer up-to, or whatever) they are completely free to ignore this optimization, like all kinds of optimizations. 

As a rule of thumb, things that do not affect layout (colors, shadows, subtle anims) can accept being delayed for slightly less than half a second without the user giving a shit. If you go below 250ms, users often don't even see the delay, especially if you transition things.

> People have a hard enough time building responsive designs;
> expecting them to also, somehow, factor in which selectors will perform
> well where ahead of time sounds like a stretch.

Sure, life is sooooooooo hard for web developers. Poor us :-D 		 	   		  
Received on Thursday, 11 July 2013 20:55:49 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:32 UTC