I think critical programmers from every generation back to Ada Lovelace have made this observation. When the solution has started costing more than the problem...stop.