If you want to make it a "guideline" I would be OK with that. If you want to make something stronger - I think that this would be a bad idea. The basic problem is that this couples a derived class to it's parent(s) and a library author doesn't always control it's parents: example: template<typename T> struct opaque_type : public T { ... const char *name = "..."; }; struct user_class : public opaque_type<T> { ... // user might really want to shadow const char *name = "asdfasdf"; }; This is a contrived example, so don't call me on this one. My real point is that it's hard to predict the cases where a "hard rule" would be problem. Robert Ramey -- View this message in context: http://boost.2283326.n4.nabble.com/Warning-policy-local-variable-hides-i-e-s... Sent from the Boost - Dev mailing list archive at Nabble.com.