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

Re: [css3-background] percentages in border-radius - really relative to height?

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 14 Oct 2009 15:21:40 -0500
Message-ID: <dd0fbad0910141321r3b54439dqdd12004141bb341b@mail.gmail.com>
To: "L. David Baron" <dbaron@dbaron.org>
Cc: Zack Weinberg <zweinberg@mozilla.com>, W3C Emailing list for WWW Style <www-style@w3.org>
On Wed, Oct 14, 2009 at 1:43 PM, L. David Baron <dbaron@dbaron.org> wrote:
> On Wednesday 2009-10-14 11:30 -0700, Zack Weinberg wrote:
>> If this particular argument was already considered and rejected, I
>> would appreciate a pointer to the discussion.
>
> Most recently at last week's teleconference:
> http://lists.w3.org/Archives/Public/www-style/2009Oct/0042.html
> but also a number of times before that.

The highlights of that discussion were that, when you want a
quarter-circle corner, it's nearly always sufficient to just specify
it with a length.  The major case where you really do want a %
quarter-circle corner is to make "lozenges" (typically used for
buttons, where the sides end in a half-circle), and for those you can
just set a very large length instead - the length scaling will reduce
it appropriately and keep things correctly proportioned.

On the other hand, the major use-case for per-side %s is to create
circles/ovals, and that *can't* be done at all if % always refers to
width.

That all said, Bert, Elika, and I discussed it a bit later that day in
the #css irc, and thought that perhaps a keyword would suffice to
switch between the two behaviors.

~TJ
Received on Wednesday, 14 October 2009 20:22:30 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:21 GMT