Date: Mon, 24 Jul 2006 04:43:45 +0100 From: Geometry and Topology To: gt-announce@msp.warwick.ac.uk Subject: Site closure Colleagues: The purpose of this message is to inform you that in the near future Geometry & Topology, and Algebraic & Geometric Topology, will no longer be open access journals, free electronically to the world. We have kept these journals open access for 10 and 5 years, respectively, depending on volunteer labor and a minimal amount of subscriptions income. Although the journals are open access, we never granted libraries permission to connect to our sites without paying us a very small fee. However the income generated by this fee has never grown to cover our modest costs and we have been forced, reluctantly, to close the sites. Nevertheless, we intend that the journals shall stay freely accessible to individuals and, to this end, we shall continue to store the individual papers at the arXiv. There is an ongoing evolution in the math libraries to switch from paper to electronic journals. For example, UCBerkeley no longer takes paper copies of any Elsevier or Springer journal. Since our two journals were open access, and libraries were consequently unwilling to pay for access, our modest number of subscribers has been dwindling, just as our volunteer labor wishes to move on to other endeavors. Thus, within weeks, G&T and AGT will become accessible only to those people whose libraries subscribe and provide us with their IP addresses. Please use your negotiation skills to persuade your math librarian to take out an electronic subscription for GT and AGT (which automatically gives access to the monograph series as well). Our journals will continue to sell at 10cents/page for electronic, and 15cents/page for paper. This is below almost all journals, as can be seen by the tables appended. What follows is a discussion of quality and the economics of publishing math journals, which you can skip if you wish. This discussion will contain some arguments in favor of G&T and AGT which you may want to use with your math librarian. One way to measure quality is through citations and impact factors. For a journal J one counts all the citations in journals published in 2005 to papers that were published in J in 2003 and 2004, and then divides by the number of articles in those two years. The resulting number, roughly the average number of citations for a given paper in the year 2005, is called the impact factor. This is the method used by The Thomson Corporation's ISI Web of Knowledge, also known as Science Citation Index, which can be reached at this URL: http://portal.isiknowledge.com/portal.cgi?DestApp=JCR&Func=Frame They list only 181 math journals, the journals which they believe are established and worth of ranking (the AMS surveys 274 journals). The top 20, plus selected topology or geometry journals, are listed below, and you can see that Geometry & Topology is eleventh, or actually eighth among regular math journals. Rank Abbreviated Journal Title Impact Factor 1 J AM MATH SOC 2.323 2 ANN MATH 2.009 3 COMPUT COMPLEX (not math) 2.000 4 COMMUN PUR APPL MATH 1.841 5 B AM MATH SOC (not research articles) 1.800 6 ACTA MATH-DJURSHOLM 1.778 7 INVENT MATH 1.652 8 J EUR MATH SOC 1.414 9 MEM AM MATH SOC (monographs, not articles) 1.315 10 DUKE MATH J 1.304 11 GEOM TOPOL 1.275 12 J MATH PURE APPL 1.195 13 PUBL MATH-PARIS 1.182 14 DISCRET MATH THEOR C 1.061 15 DISCRETE CONT DYN S 1.025 16 INTERFACE FREE BOUND 1.024 16 RANDOM STRUCT ALGOR 1.024 18 ANN SCI ECOLE NORM S 1.000 19 ADV MATH 0.991 20 AM J MATH 0.978 37 TOPOLOGY 0.770 46 J DIFFER GEOM 0.676 84 K-THEORY 0.474 131 GEOMETRIAE DEDICATA 0.330 133 J KNOT THEOR RAMIF 0.323 149 TOPOL APPL 0.297 However G&T also has one of the the lowest subscription prices. If we divide the impact factor by the price per page for an electronic subscription, then we get a rough approximation of value per cost, of bang-for-the-buck. Some journals do not have a separate price for the electronic version, and some companies bundle their journals, so for the purposes of this calculation, we have used 90% of the price per page given at http://www.mathematik.uni-bielefeld.de/~rehmann/BIB/AMS/Price_per_Page.html For some of the more mainstream math journals listed above, here is an ordered list based on this calculation. Note again that this is a rather odd number, citations per article divided by price per page, but it gives some idea of relative worth. Annals Math 18.26 (= 2.009/.11) G&T 12.75 JAMS 10.56 (= 2.323/.22) (AGT) 6.76 American J. Math 5.43 Acta Math 4.20 Duke J. 3.10 J Eur. Math. Soc. 2.12 Pub IHES 1.85 Invent. Math. 1.72 Comm Pure App Math 1.61 Advances Math 1.48 J Diff. Geom 1.99 Topology .86 K-Theory .59 J. Knot Thy Ram .58 Geom. Ded. .38 Topology Applic. .28 Algebraic & Geometric Topology is not yet listed at Thomson ISI, but it is possible to get citations and impact factors from the AMS at mathscinet by entering journal names one by one. The AMS impact factors are computed by counting citations to articles in the year 2000 through 2004. See http://www.ams.org/mrcitations/search.html The citations usually do not vary too much from Thomson, and the ranking do not change much (unless a journal has changed quality in 2003-04 compared to the previous three years). The impact factor for AGT at the AMS site is .676 (on 19 July 2006; the AMS figures change as new information comes in, whereas the Thomson impact factors are fixed for 2006). This is a remarkably good number for such a young journal, and if you divide by its cost per page, the bang-for-the-buck figure is an even more remarkable 6.76. This places it fourth in the above table. Take these numbers with a large grain of salt. If the approximations in the above calculation all fall on one side, a journal can be quite too low or too high. Note also that some journals which are not in the top 20 in impact factors, are also very cheap, so their bang-for-the-buck number will place them high on the list. A good example is the Indiana Univ. Math. J. which has an impact factor of .769 and a cost of .15, giving 5.13, placing it 6th on the list. Other university press journals do well, e.g. the Pacific J. Math at 2.82. Rob Kirby Colin Rourke _______________________________________________ gt-announce mailing list gt-announce@msp.warwick.ac.uk http://www.msp.warwick.ac.uk/mailman/listinfo/gt-announce