I think you are correct, its so hard to judge the quality of research that we have been using publication record/count as a proxy. Ultimately, having papers being easy to write is good, provided we find a better way of judging quality.