W3C home > Mailing lists > Public > public-respimg@w3.org > February 2013

Re: CfC: to publish "The srcset attribute" specification as a First Public Working Draft (FPWD)

From: <lists@ericportis.com>
Date: Wed, 6 Feb 2013 14:58:54 -0700
To: Yoav Weiss <yoav@yoav.ws>, "Nathanael D. Jones" <nathanael.jones@gmail.com>, Mat Marquis <mat@matmarquis.com>, "public-respimg@w3.org" <public-respimg@w3.org>
Message-ID: <B393B43E9FC0453B98E84590705F620A@ericportis.com>
On Wednesday, February 6, 2013 at 10:36 AM, Yoav Weiss wrote:
> There's a big difference between adding display width/height and adding the resource's width/height to the HTML. If at some point you modify the image resource with a different one (while keeping the same proportions), you would have to modify your markup, where otherwise you wouldn't.  

This seems backwards to me: HTML is exactly where we should describing our resources, rather than defining how they should be displayed.

If we're concerned with markup durability, srcset="" & <picture> require you to do a very tricky thing: refigure your in-markup, viewport-based assertions every time you tweak your *layout*. I wrote about this here:

http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2012-November/037772.html
http://lists.w3.org/Archives/Public/public-respimg/2012Nov/0001.html

Nathaniel's proposal would still require authors to do viewport->element sizing math, but in CSS -- potentially DRYer, but not fundamentally simpler.

I make no claims regarding browser technicalities or standards-process expediencies, but philosophically, I'm with Fred. Context-based source-picking would allow for a full separation of presentation and content and is thus an important use-case that should be addressed, somehow, by somebody, eventually.


—eric
Received on Wednesday, 6 February 2013 21:59:25 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:12:39 UTC