On Thu, Nov 20, 2003 at 09:10:42AM +0100, Alfredo Braunstein wrote: > PS: to make *me* happy, you should look at the two chunks below and tell if > replacing the first one with the second one will miss something important.
Even if it did (I can't see anything obvious), I'd go for the second version. The first looks like a mixed structure-and-position-based solution (as mathed does), but I don't think this is necessary for the outer world as structures are more 'regular' (i.e. rectangular) there. The code as it stands is probably due to some evolutionary process rather than created from scratch with a certain plan in mind. [If we need this mixed approach in the end (which I don't believe) we could promote the math solution instead of trying to keep this mixed solution alive (and possibly try to unify two incompatible mixed approaches).] Andre'
