Reforming the research evaluation system is essentially a struggle between the academic aristocracy, which professes expert approaches to evaluation, and the scientific bureaucracy, which seeks to ensure the dictate of numerical indicators. A representative example is the evolution of this confrontation in China, where a long-term commitment to international scientometric databases made it possible to achieve world leadership in terms of the number of publications. However, these achievements were accompanied by negative effects caused by deterioration in the personnel policy of universities, the reduction of academic freedoms, changes in the publication behavior of researchers, and violations of scientific ethics. This led to a reform of the national evaluation system, namely, the strengthening of the role of expertise, the priority dissemination of scientific information at the national level, and the restriction of open access publications. This review considers the history of the formation of and changes in the research evaluation system in China, its positive and negative effects, as well as new scientific practices, their likely consequences, and the possibility to apply them in other countries.

В сборнике биографий римских императоров Historia Augusta (HA), который пользуется сомнительной славой в научной среде, имеется несколько упоминаний о выдаче паролей дежурному преторианскому отряду, охранявшему особу правителя. В общей сложности в составе сборника содержится пять упоминаний о проведении этой процедуры, из которых два относятся к одному эпизоду. Проведенный анализ показал, что, несмотря на ряд сомнительных элементов в описаниях ее в HA, данная информация, скорее всего, является достоверной. In the collection of biographies of Roman emperors, Historia Augusta, which enjoys a dubious reputation in the scientific community, there are several references to issuing watchwords to the Praetorian Guard on duty, who protected the ruler’s person. In total, the collection contains five mentions of this procedure, of which two relate to one episode. The analysis showed that despite a number of questionable elements in the descriptions of this procedure in HA, this information is most likely reliable.

1. Rick Snodgrass, Summary of Conference Management Software, 1999. URL: http://www. acm.org/sigs/sgb/summary.html 2. Conry D., Koren Y., and Ramakrishnan N.. Recommender systems for the conference paper assignment problem // In Proceedings of the third ACM conference on Recommender systems, 2009. 3. Goldsmith J., Solan, R.H.: The AI Conference Paper Assignment Problem. URL: http://www. cs.uic.edu/~sloan/my-papers/GodlsmithSloanPaperAssignment.pdf 4. Afrapportering: Konferencesystemer – systemreview og pilotprojekter, 2009. URL http://www.deff.dk/content.aspx?itemguid={6557F973-A06F-47E0-B0AA-EBE7736D8988} 5. MoReq2 specification. CECA-CEE-CEEA, Bruxelles- Luxembourg, 2008.