On 5/2/11 7:26 PM, Ian Hickson wrote:
That makes sense, though I think it'd be better for that to be a style
scoped to the binding that defines the<select>, personally.
OK, but more on this below.
I would clearly prefer that the behavior be defined in terms of CSS; UAs
that under the hood want to ignore the styles and just do something
magic can still do that, of course.
The behaviour is defined in terms of CSS and a hypothetical binding
language similar to XBL; in theory that should be sufficient for your
needs, no?
I don't think so; we need to define at least some details of the
binding. That's what I meant by sites depending on the details. For
example, width calculations for <select> need to work in a particular
way (or rather small range of ways)....
If not, I guess we have to work out what we can get browser vendors to
converge on. I am concerned that this might not end up being exactly what
you need, though, which would be of no more help to you than the status
quo, but with more complicated rules.
That's entirely possible, yes. At the moment we're getting bug reports
because people write their HTML+CSS, test in only WebKit or only IE, and
then it breaks in Gecko. I would assume that there are others who only
test in Gecko and then it breaks in other browsers....
-Boris
