Re: [css3-writing-modes] height: fill-available's behavior is suboptimal

On Fri, Jul 13, 2012 at 3:33 PM, Tab Atkins Jr. <jackalmage@gmail.com>wrote:

> On Fri, Jul 13, 2012 at 3:27 PM, Elliott Sprehn <esprehn@gmail.com> wrote:
> > On Fri, Jul 13, 2012 at 3:22 PM, Tab Atkins Jr. <jackalmage@gmail.com>
> > wrote:
> >>
> >> ...
> >> Neither, it should be 100px.
> >>
> >> The #outer div is width:min-content, so it asks all of its children
> >> for their min-content.
> >> > The #inner div is sized as 50%, but it's relative to an indefinite
> >> > width, so it's ignored and treated as 'auto'.
> >> > Under a min sizing constraint, 'auto' becomes 'min-content', so #inner
> >> > also asks its children for their min-content width.
> >> >> The inline-block is 100px wide.
> >> > The #inner div is thus 100px wide.
> >> The #outer div is thus 100px wide.
> >>
> >> Don't try to be too smart about percentages; that way lies madness
> >> (like tables).  Percentages are meaningful iff their containing
> >> block's size is a definite size.
> >
> >
> > This doesn't match anyone's current implementation. Both Firefox and
> Webkit
> > make it 50px.
>
> Ugh, that's dumb.  I can see why one would think it reasonable to
> consider 'min-content' as defining a definite width that you can
> resolve a percentage against, but it's obviously wrong in situations
> like this, when it produces both underflow and overflow.
>
> But yeah, you're right, that's everyone's behavior in floats. Makes me
> sad. ;_;
>

Maybe the % thing is a bad example because we can't change that. But for
fill-available, and if we added a new percent-like value that actually
worked, wouldn't your suggestion that started this thread lead to 500px
because we'd skip over the min-content div?