Re: Unitless dimensions

Philip TAYLOR:
> There was a very informative message on this (or a related) list
> not too long ago describing what the author referred to (I think)
> as a "little-known feature" of CSS whereby unitless dimensions
> (possibly in certain contexts) behaved in a useful but generally
> little-understood way.

If it has not been 'line-height', it perhaps was some hackery for quirks 
modes, where unitless numbers are often interpreted as pixel values. You 
could use these to feed different values to different browsers/rendering 
modes, when this bug coincides with others (like IE's infamous 
alternative box model).