Re: [css3-exclusions][css3-gcpm] Plan B exclusions

Also sprach Brad Kemper:

 > > I agree that spacing behavor must be described somehow, and in this
 > > case we don't have an alpha channel to use. This is noted in issue 13:
 > > "Some kind of spacing behavior must be defined." We could,
 > > perceivably, use some kinde of kerning parameter from the font in
 > > question. But this may be unavailable or hard to extract. So a
 > > property may be easier. We would only need one property, no? 
 > Maybe just 'margin' on the exclusion would be enough, to where it
 > was no longer a shorthand when applied to an exclusion, and only
 > applied to stuff that wraps it, and only collapsed with similarly
 > scoped exclusions? If that is too magic, then maybe call it
 > 'exclude-margin'.

I like the reuse of properties, but I think we need a separate one in
this case. (If we allow background images to define exclusions, we
can't really reuse the 'margin' property of the element which owns the
background.) So, 'exclude-margin' makes sense, I think. 

 > In example 70:
 >    • what is determining the stand-off so that the text doesn't touch the "y"? 

Nothing at this stage, this is why the note says: "Some kind of
spacing behavior must be defined". I've added 'exclude-margin' as a
possible solution.

 >    • Would there be a way to have text also go to the left of the
 >    descender of the "y", or for the other to choose the desired
 >    behavior there?

The way I think about the example is that other content would stay to
the right of a left-floating element. This way, content would not get
to the left of the 'y' descender. This definition is convenient to
achieve example 70, but bay be inconvenient in other examples.

 >    • Are you imaginging there would also be something to get the
 >    text to line up with the top of the lowercase letters?

Yes. A negative margin is probably the easiest solution:

  #dropcaps {
    font-size: 3em;
    float: left;
    exclude-level: 0.5;
    margin-top: -0.2em;

This solution is not ideal if the used font has a different x-height
from the specified font. Hmm.

              Håkon Wium Lie                          CTO °þe®ª        

Received on Monday, 6 February 2012 00:00:03 UTC