Perhaps [~ann.campbell.2], we could raise an issue as soon as a method call beginning with a verb like add, remove, set, delete, ... is done inside the assert expression ?
\[~freddy.mallet] from an implementation standpoint, I think that would be a good idea. From a rule description standpoint, I hesitate to make that change. I believe this will be one of those rules that continually expands in scope. As the scope grows, I thing the best way to describe that scope will emerge.
\[~ann.campbell.2] [~freddy.mallet] IMO the rule should raise an issue whatever is after an assert. If the program is thought to not be working with special value then it might be a good idea to have something failing properly (guava preconditions, throwing an exception... ) instead of relying on an invariant that might not be triggered at runtime.
