[Bioperl-l] release numbers

[Jason Stajich]

On Wed, 14 Aug 2002, Hilmar Lapp wrote:

> Hm. That automatically bumps all other packages to maturity? Well, if
> it saves a lot of hassle for the actual user ... does it really?
>
I don't know, it was suggested to me that people find it very confusing to
know which version of pkg X is necessary with extra-add-on pkg Y when
things are different version numbers. As developers we know the history
so it isn't a big deal, for a new person trying to get things working,
all these versions are confusing and most people will just pick the
highest number from each category.   Clearly this has been solved before
so I think we ought to follow a standard protocol here.

Alternatively we can just let each pkg keep its own version scheme and
make a big PKG.TABLE which lists which packages are compatible.
I don't know how you guys want to do the bioperl-db release wrt to the
schema, does than need to be a separate pkg that the user gets or should
the pkg maintainer put the latest table.sql in the correct place before a
release is done?

I'm also worried we don't give people an easy way to detect which version
of a pkg they have installed since we don't use the $VERSION variable in
any of our modules expect Bio::Seq.  So we can't do auto-checking of
which version someone has installed.


-jason



> 	-hilmar
>
> > -----Original Message-----
> > From: Jason Stajich [mailto:jason@cgt.mc.duke.edu]
> > Sent: Wednesday, August 14, 2002 2:18 PM
> > To: Bioperl
> > Subject: [Bioperl-l] release numbers
> >
> >
> > Shall we release the bioperl-run, bioperl-db, and
> > bioperl-pipeline pkgs
> > with the same version number as bioperl (1.1).  We can do it s.t.
> > bioperl-pipeline 1.1.1 is compatible with bioperl 1.1 but
> > bioperl-pipeline
> > 1.2 is not necessarily?  This sort of means we have to do the
> > major pkg
> > releases in lock step with the bioperl core releases.  Is
> > this okay or is
> > it going to be a major pain later on?
> >
> > I don't want us to get into confusing situtations for users - as it is
> > already a pain upgrading perl modules w/o doing with a pkg manager.
> >
> > -jason
> >
> > --
> > Jason Stajich
> > Duke University
> > jason at cgt.mc.duke.edu
> >
> > _______________________________________________
> > Bioperl-l mailing list
> > Bioperl-l@bioperl.org
> > http://bioperl.org/mailman/listinfo/bioperl-l
> >
>

-- 
Jason Stajich
Duke University
jason at cgt.mc.duke.edu