Further down in the article, the answer is given. They are recommending that journals, funding bodies, etc, start to require scientists to define their chosen categories and justify the usage.