When a giant like Leslie Lamport says he knows next to something about a subject but then gives a view anyway, it's probably wise to consider that he might just

When a giant like Leslie Lamport says he knows next to something about a subject but then gives a view anyway, it's probably wise to consider that he might just be being diplomatic. His comments about ambiguities in UML and the difference between that and a rigorous mathematical specification are reasonable in my experience (as someone who has both done significant work with high reliability code and formal methods and also worked in enterprise world with UML and all that).