[Bioperl-l] BioPerl CPAN

Nathan (Nat) Goodman natg at shore.net
Wed Jul 2 07:22:37 EDT 2003


Hi Folks

This is a great discussion.

Jason & Aaron and Heikki gravitated to the two extremes of what might be
done with the repository.  The Jason/Aaron thought, in Jason’s words, is
that this would “be a way for people to get their own code out there sooner
without it being blessed by 'bioperl'.”  Heikki thought the idea was to
“split up the current bioperl-live”.

I think the Jason/Aaron end of the extreme would be extremely valuable in
its own right. If this is all we accomplish with CPAN it would be worth
doing, because it would make it easier for the community of BioPerl ‘users’
to contribute code to the effort even if it isn’t general enough to be part
of the main BioPerl distribution.

I think the repository can also be helpful with efforts to simplify the
existing BioPerl class structure by providing a place to move code that no
longer fits, or code that just isn’t central enough to warrant a spot in the
main distribution.  It can also help the BioPerl distribution grow in a more
coordinated fashion by providing a place for new code to live before it gets
integrated into the main structure.

Each of these uses is valuable.  I suggest we test drive the CPAN concept
with one example of each type.  In other words,

1) one contribution that represents code that will probably never migrate to
the main distribution
2) one contribution that represents code from the main distribution that
needn’t be there
3) one contribution that represents code that might migrate to the main
distribution someday, but isn’t ready for prime time yet.

I’m happy to volunteer my new Biblio code for type (3).  Some of the
existing tools stuff could possibly be used for type (2).  I’m sure there’s
tons of great BioPerl-based code out there that could be used for type (1).

In terms of the ideas for restructuring the core: my only modest suggestion
is that it would make sense to defer significant renovations until the
technical evaluation that’s been discussed recently is carried out.  This is
to avoid the “ready, shoot, aim” phenomenon.

Best,
Nat



More information about the Bioperl-l mailing list