W3C home > Mailing lists > Public > www-style@w3.org > January 2009

CSS Transforms : computed value

From: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
Date: Tue, 13 Jan 2009 17:53:11 +0100
Message-ID: <496CC6F7.4040707@disruptive-innovations.com>
To: "www-style@w3.org" <www-style@w3.org>
Cc: hyatt@apple.com, Dean Jackson <dino@apple.com>

Hi there,

We're currently implementing CSS Transforms [1] in BlueGriffon, our new
Gecko-based editor, and we have a hard life dealing with
the matrix used as computed value of the (-moz-/-webkit-)transform
property. We find it beautiful from an intellectual point of view
but horrible from a practical one. As usual, it's perfect for browsers
but awful for editors. We do understand the need/wish to
be coherent with the SVG 1.1 spec, but this clearly makes the life of
editing environments (very) painful, and the computed value of
'transform' almost useless from our point of view since we can't from
a single matrix get all the individual and ordered transformation
matrices that result in that one when multiplied together.

I strongly recommend keeping the computed value of 'transform' more
humanly readable, in the form of a series of individual transformations
rather than a single matrix. I do understand that means the layout
engine has to store more information than just a single transformation
matrix. But if you want real-life wysiwyg editors to implement this -
and it seems web designers want it too - then it's almost a necessary
cost, at least from out POV.

[1] http://webkit.org/specs/CSSVisualEffects/CSSTransforms.html


Received on Tuesday, 13 January 2009 16:53:49 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:23 UTC