In these areas, it seems likely that if language engineering evaluation can make any headway, it will be contributing as much to the state of the art in requirements engineering as vice versa. What follows does not address all these issues adequately by any means, but they should be borne in mind for further work.