Conversation

which solves the (*foo).bar() awkwardness, but still leaves (&foo).bar(), which in C++ is solved by having transparent references...
1
I always feel like just having postfix ref/deref operators would be much better but I haven't really thought it through
1
yeah... no reason foo->bar couldn't just be foo*.bar except that e.g. foo*(x) is ambiguous :/ for some reason postfix ref seemed less appealing but I don't remember why
1