From owner-chemistry@ccl.net Wed Sep 14 23:14:26 2005 From: "CCL" To: CCL Subject: CCL: W:Disclose your data, or not publish ! Message-Id: <-29167-050914231333-32392-ETc9kVj+VqHwexQEb/3djg|"|server.ccl.net> X-Original-From: "Chemical Bond" Sent to CCL by: "Chemical Bond" [chemicalbond001|"|yahoo.com] Everyday, tons of new publications come out. Although there are some nice papers with novel ideas or new results, many times we see people are almost trying to repeat others' work, pouring the same water into the old bottle. The only goal for that seems to be just publishing, for tenure in academia or for promotion in industry, or for boasting some software! One obvious case is in the community of protein-ligand docking & scoring. Are not we tired of those studies? Yes, we are! We always see extremely excellent results published in papers or ACS talks by different kinds of people, especially those who made software and make money! The reality is that none of them really works! Not at all! Ask anyone in a big pharma doing docking & scoring business, see what you would get the performance for a scoring function applied in a real drug design case? That's all about business, to keep software company making money and keep computational chemists' in jobs. Fine, but in order to promote the advancement of science, we had better do something novel than that. I hope people in this list could make more suggestions, and here I like to propose a little to the community: To disclose your data, or not publish. (1)Disclose all numerical data files, input, output files & parameters in the publication (paper or talk), in the sense that at least it could be reproduced (statistically meaningful) if anyone else is trying to do it using the same software or method. So no one can be lying any more. (2)This could save much time for many other researchers on preparing testing data, so that new ideas could get tested & improved very quickly; (3)This could help build a very rich resource for research, things like docking decoys for many many protein targets and ligands; the data itself would be worth a million dollars ?! ... ... All in all, theose really good methods or software will survive without questions, and new technologies would evolve in the speed-of-light, we would see fewer but more meaningful new papers every day, what a relif from tons of new electronic papers... ... so that in the end science would get advanced by taking advantage of everyone's strength, and life could be improved in a different way...... Thanks for your attention! -Bond