W3C home > Mailing lists > Public > public-respimg@w3.org > October 2014

New website, new name

From: Eric Portis <lists@ericportis.com>
Date: Fri, 31 Oct 2014 15:57:27 -0600
To: public-respimg@w3.org
Message-ID: <9D7177B8-A83D-4996-93F3-4F9866F4A2F3@ericportis.com>
## Scope creepy!

The RICG is about to tackle [Element 
Queries](http://responsiveimagescg.github.io/eq-usecases/). To prepare 
for this imminent broadening of our scope of work, we’re doing some 

- It’s official! We’re now the Responsive *Issues* Community Group
- We’ve got a new website, and it lives at [ricg.io](http://ricg.io). 
It’s a bit sparse at the moment but it’s a work in progress.

Element Queries are a GHOULd idea! But it’s still very, very early 
days; to draft a use cases document, SPOOKification, and ultimately 
bring them to browsers we’ll need all the help we can get. 

## image-spook()

A week ago Jason Grigsby 
“what ever happened to `image-set()`?” `image-set()` brings 
`srcset`’s functionality to CSS; Jason does a great job of explaining 
why we need it in a subsequent [blog 
`image-set()` was implemented with a prefix in WebKit [two years 
(!) and then we all kind of... forgot about it.

The discussions that came out of Jason’s post were productive. 
`image-set()` is happening; [it’ll be specced as a part of CSS Images 
[it’s going to have feature parity with HTML’s respimg 
and the RICG is going to push for it however we can. TERRORiffic!

## Boo!-link

Yoav’s respimg implementation in Blink is already shipping, but that 
doesn’t mean it’s done. Some changes made over the last couple of 

[Blink now uses gemetric means to pick `srcset` 
sources](https://codereview.chromium.org/667763004/). What in the HELL 
does that mean? Blink used to pick the smallest source that supplied *at 
least* as many pixels as it needed; now it picks the source whose 
dimensions are the *closest* to what it needs. For example: before, 
given an option between two resources, one with 0.9x the image’s 
device pixel width and the other with 3x, Blink would have picked the 3x 
source; now it’ll compromise a little bit on sharpness and save a lot 
of bytes by selecting the 0.9x resource.

Different browsers are going to do different things here. With any luck, 
they’ll continue to experiment with, refine, and ultimately improve 
their `srcset` picking logic over time. Developers can’t predict which 
source will be chosen out of a `srcset`, and that’s a good thing. 
[“`srcset` is about letting 

Blink also now [avoids downloading smaller images if bigger ones are 
already available in the 
cache](https://codereview.chromium.org/674923004/). So, now, shrinking 
your browser window (or changing the orientation of your phone from 
landscape to portrait) won’t trigger a new request for a new resource; 
Blink will happily continue to scale-down the old, larger one. A clear 
and obvious win!

Lastly, Blink now gives developers [console warnings about bad `srcset` 
descriptors](https://codereview.chromium.org/649183007/). Helping you 

## Yay! Link!

The [video from Wilto’s excellent talk at Refresh 
is up and it is chock full o’ Zelda references.

## Tricks and treats

There’s been no SCAREcity of respimg links over the last two weeks:

- [Picturefill 2.2 hit 
The new version brings better spec compliance, better performance, fewer 
bugs, and it plays nicely with Asynchronous Module Definitions.

- The NCC Group broke [Betteridgde’s Law of 
by asking, [“Is it time to start using `srcset` and the `picture` 

- Eiji Kitamura published a bit of Javascript that uses [Service 
(which were just [green-lighted in 
to manipulate image requests based on a global manifest. Yoav 
[asks](https://twitter.com/yoavweiss/status/525568123333017600), “who 
will write the tool that does the same thing server-side?”

See you in a couple of weeks!

Received on Friday, 31 October 2014 21:57:59 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:06:15 UTC