Re: outline-radius

On Mar 12, 2012, at 1:02 PM, Brian Manthos <brianman@microsoft.com> wrote:

> Paul Bakaus wrote:
>>> You mean if single box?  That's fine, but my point was that the spec 
>>> needs to describe behavior in the multiple-box case too.
> 
> Kang-Hao wrote:
>> Details are fun, but I feel like every time when someone talks in this
>> way, it *feels like* someone is discouraging people from sending
>> proposals. (Though I do have great sympathy for the whole
>> implementation industry.)
> 
> I'd prefer border-radius rounding of spans that split across multiple lines (and thus have all kinds of 1-to-many element-box relationships) was addressed directly and clearly in the specification (with rendering examples!)

Do you mean such as, should 'box-decoration' affect whether or not the online box is rounded at the break as with blocks, and if not (or by default), should it be rounded there, by border-top-right-radius and border-bottom-right-radius in ltr text?

I say yes and yes. But I think we resolved to leave it undefined for bidi. 

> before the WG even consider expanding the complexity to outline.
> 
> It might feel like discouraging, and perhaps in part that's true.  I strongly discourage introducing shiny-27 until dependency shiny-15 is specified fully.

I disagree. Having something that looks like a round peg in a square hole means having something that looks broken. Fixing it is more important than just layering on some new shiny.