W3C home > Mailing lists > Public > public-media-fragment@w3.org > June 2009

Re: Image sprites

From: Ryo Kawaguchi <rkawaguchi@mozilla.com>
Date: Tue, 16 Jun 2009 22:08:10 -0700 (PDT)
To: public-media-fragment@w3.org
Message-ID: <15688612.48751245215290728.JavaMail.root@cm-mail01.mozilla.org>
> There has recently been some discussion of the use of sprites on the
> CSS WG mailing list (see the threads [1][2][3][4][5][6]).
> 
> [...]
> 
> However, the way sprites are currently achieved is essentially just a
> hack, which has its limitations. Suggestions that have come up on the
> www-style mailing list to improve their support range from new
> properties that lift some of the current limitations to proposals for
> proper sprite support in CSS. Bert Bos, however, rightly pointed out
> [5] that sprites are not limited to CSS and referred me to the Media
> Fragments WG.

There has also been some related discussion on Bugzilla@Mozilla [1] regarding implementation of CSS sprites as a new proprietary CSS property (-moz-background-image-region) which displays a rectangular region of an image, and we have arrived at a similar thought that the proposal of Media Fragments WG [2] may be a better generalization of what we are trying to achieve.

We are particularly concerned with two points; (1) client-side implementation of media (image) fragments, (2) the impact of media fragmentation on the existing URIs that use the "#".

(1) We are interested in client-side media fragments in particular with static images. According to "Evaluation of fitness per media formats" [3], none of image codecs are listed as "fit", which I think implies fragmentation support for images has to be on client-side. But such attitude may go against what's written in "3.10 Focus for Changes" [4], so we'd like to hear opinions from Media Fragments WG about client-side implementation of media fragmentation.

(2) Having read "3.12 Fallback Action" [5] and Silvia Pfeiffer's message [6], we are concerned with the impact of media fragments on the existing URIs that already contain URI fragments and the "#". Is there any follow-up study, or is anyone aware of cases where existing web pages may be affected by spatial fragmentation and use of the "#" (e.g. #xywh=160,120,320,240)?

Thank you very much.

[1] https://bugzilla.mozilla.org/show_bug.cgi?id=113577
[2] http://www.w3.org/TR/2009/WD-media-frags-reqs-20090430/#naming-space
[3] http://www.w3.org/TR/2009/WD-media-frags-reqs-20090430/#fitness-table
[4] http://www.w3.org/TR/2009/WD-media-frags-reqs-20090430/#side10
[5] http://www.w3.org/TR/2009/WD-media-frags-reqs-20090430/#side12
[6] http://lists.w3.org/Archives/Public/public-media-fragment/2008Oct/0060.html
Received on Wednesday, 17 June 2009 08:35:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 21 September 2011 12:13:33 GMT