Le Tue, Mar 11, 2008 at 12:26:57AM +0100, Christoph Berg a écrit : > Package: autodocksuite > Version: 4.0.1-1 > Severity: serious > > Hi, > > it appears to me that autodock and autogrid's Suggests on > popularity-contest merely serve the purpose of pimping that package's > popcon scores. Please remove the Suggests.
Hi Christoph, hi Steffen, Indeed, the rationale is explained by Steffen in README.Debian: http://svn.debian.org/wsvn/debian-med/trunk/packages/autodocksuite/trunk/debian/README.Debian?op=file&rev=0&sc=0 Dr Garrett Morris of the upstream developers is very much looking forward to see AutoDock being distributed with Debian. The Scripps Institute asks for registrations prior to the download of the tarball or the binaries at this URL http://autodock.scripps.edu/downloads/autodock-registration which does not affect the distribution of this package under the GPL. Nevertheless, please respect upstream's genuine interest to know about the impact their work has on the field, which is key for writing grant applications. Also please consider installing the Debian package popularity-contest. This counts the installations and may thus further help upstream to further develop their software. In my opinion, the popularity-contest score would even more informative without the dependancy because there would be no bias compared to other packages. In none of the two scenario we can assume that a popcon score of one translates in one user anyway, and the popcon metric is too different from Upstream's metric of counting downloads. However, I hope that the ratio between "vote" counts for the programs we distribute and autodock is informative enough to be mentioned in Upstream's grant applications. Steffen, if you agree but are busy, I can prepare an update. Have a nice day, -- Charles
