On Mon, Jan 23, 2006 at 08:55:27PM +0100, Kevin Pfeiffer wrote: > Andre Poenitz writes: > > I'd guess it's almost impossible to get kerning right in all > > circumstances. > > That sounds like a NASA (space shuttle) answer. ;-)
Now, if they'd paid me like a NASA engineer ... > There can't be too many more than about 65K possible combinations I > would think -- apologies in advance if my math is wrong. Like UCS4 squared? Makes 2^64 at max. Ok. A few Klingon/Prehistoric Chinese combination could probably disregarded. > > Using a manual kern do not look overly ugly to me. > > It looks fine, but I have two volumes of 400 pages each and have no > tremendous desire to search (or even S/R) through them for bad > combinations. > > I'll concentrate my efforts on the first page of each chapter. ;-) Good choice ;-) Andre'
