Le Sun, May 14, 2006 at 04:47:00PM +0200, Bill Allombert a écrit : > On Sun, May 14, 2006 at 07:36:49PM +0900, Charles Plessy wrote: > > The packages ncbi-tools-bin and ncbi-tools-x11 have many many menu > > entries. Do you think that it would make sense to create a "NCBI" > > subsection in the Science section? > > Note that this is beyond the scope of the Debian menu structure, since > the Debian menu sub-policy allows packages to use new third-level > sections for private purpose. > > The criterion to add a new (unofficial) section is whether: > 1. that will help users find the softwares. > 2. the new section will almost never be with exactly one or two entries > and often with at least five entries when it is not empty. > > Here criterium 2 is fulfilled easily: there will be 0 or >=8 entries, > > For criterium 1. I think "NCBI" is far too cryptic. What does that mean > anyway ?
Hi, Sorry for my ignorance, I overlooked the relevant part of the sub-policy. NCBI is the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). Think about the Google for biological research. My reasonning was that if people install the packages ncbi-tools-*, they necessarly know what the NCBI is. These packages are not part of complex dependancies networks: sorbet【debian】$ apt-cache rdepends ncbi-tools-bin ncbi-tools-bin Reverse Depends: perlprimer perlprimer ncbi-tools6 ncbi-tools-x11 ncbi-data libncbi6 blast2 sorbet【debian】$ apt-cache rdepends ncbi-tools-x11 ncbi-tools-x11 Reverse Depends: vibrant6 ncbi-tools-bin libncbi6 blast2 Being a molecular biologist, I can certify that a user of any of those packages would know what the NCBI is. The only case where I guess that people would be exposed to the name without knowing it is in the live distributions with pre-installed scientific programs, such as Bio-Knoppix or Quantian, or if they mass-install packages sharing some tags. Bonne journée, -- Charles