Much of what we accept as moral norms is essentially the result of an encoding of generations of consequentialist observations surrounded by myth. Certainly many dietary rules, sexual norms, and other aspects of human behavior have been the subject of rules that serve some purpose, but may or may not still serve that purpose as optimally now due in part to the impact of technological advances. From time to time it is good to reexamine these norms and either rediscover their utility or alter them in some manner designed to produce a better system.
Old mores may be suboptimal. The mechanism that motivated their initial adoption may be better understood today and a modification of the rule may be in line. The rule may have been conditioned on some state that is no longer in existence. It may be that the rule is excessively restrictive or limited in application. New rules may contradict old rules forcing the revocation of the old to make way for the new. In all these cases, the decision for change is free from risk only if the universe of consequence is well understood and all ramifications of the revision can be calculated. This is never the case so instead we tinker -- making small revisions over time in the hopes that we are not rending some essential seam that binds our society. Progress is nearly never free of risk and so we forge ahead (or in cycles).
There is a greater and more profound limitation to our ability to usefully revise morality that has escaped many rationalists. This is the limitation imposed by the domain itself. The encoding of moral norms in a determinative system presupposes determinism. However, the essential aspect of human nature is that we are not determinative organisms -- we have free will. Thus, there is a realm of morality that transcends our ability for codification -- it can only be revealed.
The transcendental morals are distinct from mores and are immune to utilitarian calculations. In our rush to break free of obsolete mores we are at risk of confusing moral axioms with mores. There is a simple method of distinction: moral axioms refer to that which makes human life nondeterministic; the rest are mores.
Information theory only applies to mores.