On Sat, May 31, 2014 at 12:11 AM, Anne van Kesteren <ann...@annevk.nl> wrote:
> On Sat, May 31, 2014 at 6:40 AM, Rik Cabanier <caban...@gmail.com> wrote: > > We already went over this at length about a year ago. > > It does seem useful to note some of the lessons you learned in the > specification. E.g. that skew() is largely misplaced, but that it's > there for compatibility. And what the plan is for evolving the API > going forward as it seems that might affect implementation strategy > reading Benoit's comments. I'm unaware of planned extensions to the API. Following Benoit's comments and some experiments we might change the internal implementation to be either use a 2d or 3d matrix and this will change the spec a bit. Benoit, please let us know what we can test to reproduce your concerns. You should already be able to reproduce it with WebKitCSSMatrix. The API is planned as a replacement for SVGMatrix and will be used in Canvas 2D and CSSOM. It also takes arrayBuffers in the constructor and in toFloatxxArray to simplify interaction with WebGL. It's not meant as a high performance replacement for a WebGL matrix. _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
