Ilya Minkov wrote > interaction of the change with ALL possible features of the > language. What improves, what breaks? If for all those features the underlying mathematical models would be known, it would be much simpler to detect the interferences described. -manfred