open-source software for bioinformatics (was Re: Unix vs Linux - the movie.)

Sean Eddy eddy at wol.wustl.edu
Mon Jul 31 09:21:39 EST 2000


In article <8lvh5d$m8h$1 at jetsam.uits.indiana.edu> gilbertd at bio.indiana.edu (Don Gilbert) writes:
  >The point I want to make is that it is not reasonable to hit
  >bioinformatics software developers over the head with blunt
  >analogies such as 'source code [an individual's creation] is like
  >the human genome code [nature's creation]'...

Don, if I understand correctly, you're arguing that the more creative
you think your work is, the more right you have to keep it secret from
other scientists?

The purpose in making DNA sequences public is that they are a key
resource for moving science forward. Without open sequence release, a
paper about a DNA sequence is little more than an advertisement.  For
any bioinformatics software that cannot be fully described in a paper
(say, as a mathematical algorithm or as pseudocode), the source code
is an analogous resource.  I don't think the amount of "creativity" in
the work is relevant (and in any case, I wouldn't accept your argument
that a genome sequence is not an intellectual work).

I'm not saying we should force developers to release source. I'm
saying that if they publish a paper on a software package, *then* they
should be expected to release source. In other words, if people want
scientific credit, they should play by all the rules, not just a
subset that pleases them. 

-- 
- Sean Eddy
- HHMI/Dept. of Genetics, Washington University in St. Louis
- http://www.genetics.wustl.edu/eddy/







More information about the Bio-soft mailing list