Fib 2 = oe STANFORD UNIVERSITY MEDICAL CENTER / PALO ALTO, CALIFORNIA f f f DEPARTMENT of GENETICS School of Medicine \/ February 9, 1962 Dr. Eugene Garfield Institute for Scientific Information 33 South 17th Street Philadelphia 3, Pennsylvania Dear Gene: I am trying to catch up on my homework before the Scientific Advisory Committee meeting tomorrow and only have time for this brief note. 2. Would you also send 20 copies to.Dr. David Robinson, ~ Secretary, Science Information Panel, Office of the Science Advisor to the President, The White House, Washington, D.C. He would like them to distribute to the members of the panel, although many of them probably have such material, 3. I should have briefed you about the Lowry paper. It probably is the most frequently quoted Single article in biochemistry Since it refers to a now standard method for the analysis of pro- teins. Actually it is a little silly for it to be so constantly quoted since "the Lowry method" would be quite sufficiently indica- tive even without the reference. It is an interesting point whether references to methods of this kind would be valuable in a citation indexing system. One can certainly visualize circun- stances where they would help to define the content of a paper in a multiple intersection search System. However, probably one could afford to leave out such presentations--whenever the number of citations exceeded some reasonable level--putting in the print out only a notation that there had been so many such citation but one would certainly want to leave this in the memory. Another obvious method of abbreviation would be to ask the computer to look for redundancies of referants on any single page of print out. These will, of course, arise when several papers by a given author are Eugene Garfield 2 February 9, 1962 included in the bibliography of any single paper. Since the main function of the SCI is to recall literature rather than create objective resortings of bibliographies, this could be a worthwhile economy for the print outs. However, once again one can visualize analytical situations which would justify keeping such data in the memory. There is also the problem of self-referencing since it again does seem absurd to make an explicit list of references by an author to his own papers. I think the way to get around this is to have the citation index print out list every paper whose existence is indicated in the input, but to delete all self-refer- ences where the senior author of reference and referant match. In that case, and if there were no other referants, it would just appear as a bare reference. Junior authors would likewise be listed with a’see’ or ‘see also’ cross reference to the senior author. This should be an adequate answer to the multiple author problem in an economical print out, but the memory ought still to be keyed to allow for an output of complete bibliographies which would locate all the papers on which a given author appeared whether as senior or junior. I'll have quite a few more comments to follow soon. Yours sincerely, Igéhua Lederberg rofessor of Genetics JL: pg PIS. -#A casper J SCI des turn, int J BH dy uch ? Ditka TM pre cml Ave bree aaree sinall compan, mA recsan hy fre fod, Wal cor Mh nr Prahres an Aya Abra) — Irrupbe Ore eterf abn Aye proce Moe