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

Re: Minutes, 22 July 2009 CSS WG telcon

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 22 Jul 2009 16:06:48 -0500
Message-ID: <dd0fbad0907221406x4d9d7dbap4cac79c08db253bb@mail.gmail.com>
To: Chris Lilley <chris@w3.org>
Cc: www-style@w3.org
On Wed, Jul 22, 2009 at 12:16 PM, Chris Lilley<chris@w3.org> wrote:
> BB: Any other case where you want percentages, apart from elliptical
>   boxes?

Yes.  In general, if I have a box that can be drastically different
sizes (perhaps it's resizable itself) I may want the corner to
grow/shrink with it, so the corner doesn't get relatively tiny on a
large box, or relatively huge on a small box.  This allows me to
maintain the same *relative* shape/size of a corner no matter the

As I said elsethread, the current firefox behavior of basing
border-radius % off of the box width at all times does not address
this.  If my box ends up very tall and skinny the corners shrink,
perhaps too much, and if my box ends up very short and fat the corners
get very large, possibly to the point of triggering auto-resize and
making the sides of box completely round.

The only thing that Firefox behavior is better for is when I want to
keep the same absolute shape for my corners while allowing their size
to scale with the box.  However, when I do this I'm generally wanting
to also keep the box itself the same absolute shape (or else you run
into the problems I described above).  Currently this can't be done in
pure CSS without hacks, but in time when it can be done (perhaps with
the introduce of the W unit, as discussed in the minutes?) then the
per-side % behavior I favor will work correctly in this case as well.

Received on Wednesday, 22 July 2009 21:07:50 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:27 UTC