- From: Sylvain Galineau <sylvaing@microsoft.com>
- Date: Sat, 17 Dec 2011 00:08:56 +0000
- To: Edward O'Connor <eoconnor@apple.com>, "www-style@w3.org" <www-style@w3.org>
> > +www-style,-public-fx > > Simon wrote: > >> CSS Filters [1] is using a comma-free syntax for arguments to the > >> filter functions. This leads to things like: > >> > >> filter: gamma(0.5 0.2 0.2); > >> > >> I think it's wrong for this to be different to transforms[2], which > >> currently uses commas. We have to do one or the other, not mix and > >> match. > >> > >> Simon > >> > >> [1] > >> https://dvcs.w3.org/hg/FXTF/raw-file/tip/filters/index.html#FilterFun > >> ction [2] http://dev.w3.org/csswg/css3-2d-transforms/ > > Tab replied: > > This change was made at my request to match other functions which we > > are attempting to develop in a comma-less manner when possible. > > It seems to me that CSS functions fall into roughly two buckets, which for > lack of better terms I'll call "mathy" and "wacky". :) > > The plan to only use commas when separating parallel constructs makes > sense for lots of (wacky) CSS functions, and really improves their > readability. > > But for Sufficiently Mathy™ functions, dropping commas looks weird and > makes them harder to read. I think we should keep commas in such cases. > > We discussed this on the telcon in the context of attr() in css3-values. It's reasonable but I don't think we've agreed to do this yet so we shouldn't request changes in CSS Filters or other modules. I support improving our future extensibility but I'd rather do this once based on a good view of the full problem vs. making changes here and there.
Received on Saturday, 17 December 2011 00:09:37 UTC