Re: [css3-background] box-decoration-break and bidi reordering (ISSUE-182)

On Jan 25, 2012, at 10:22 AM, "L. David Baron" <dbaron@dbaron.org> wrote:

> I think box-decoration-break introduces a dependency (that didn't
> previously exist) on the exact definition of that splitting, which
> means we probably need to define it.

I think it may be more accurate to say that it highlights a problem (and possible implementation inconsistency). If browsers are splitting differently and drawing borders differently, then we should decide on a standardized behavior for that, regardless of whether or not box-decoration-break applies. 

Once we've done that, we will have consistent drawing behavior for borders and backgrounds of inline bidi content. That should be a primary goal, but it then also makes it easier to say that it is the default behavior for box-decoration-break in those situations, and then maybe box-decoration-break becomes a simpler problem to tackle there.

> In other words, I think box-decoration-break is a classic example of
> the problem I described in http://dbaron.org/log/20100531-specs .  I
> wonder if we should consider dropping it from css3-background.

You say there "These types of statements scare me [...], in many (but not all) cases, it means that the required complexity won't be implemented the same way across implementations." But it seems we have that situation even without box-decoration-break existing (or applying to inlines). Box-decoration-break brings it to light, but extinguishing the light doesn't make the problem go away.

Received on Thursday, 26 January 2012 18:22:21 UTC