On 3/9/2012 10:48 PM, Nick Sabalausky wrote: > So making improvements that involve trivially-handled breaking changes is > good for C++ but bad for D? It's always a judgment call.