From: Rafal Smigrodzki (rms2g@virginia.edu)
Date: Thu Jul 25 2002 - 10:48:02 MDT
Anders suggested:
If you count the number of papers with citation levels a
few quartiles above median, then you might get a kind of breakthrough
measure. Still hardly perfect, but calculable.
### You would have to include a way for removing technical citations -
widely used but relatively trivial improvements in lab methods, included in
the references for the purpose of continuity and definition of specific
methodology. There are some papers on sequencing methods which get cited
over and over again, although the breakthroughs that made them possible were
published in other works.
IMO a good way of measuring the quality of scientific output is to have a
group of smart people (both scientists and laymen) look at the publications
not more recent than 25 years old, and assess their value with the benefit
of hindsight, in terms of the impact of these publications on the
intellectual climate and daily life 25 or 100 years after their
dissemination.
Rafal
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:39 MST