[pkg-bioc] Re: Genetics Program

[Steffen Moeller]

Am Samstag 10 Dezember 2005 15:54 schrieb Rafael Laboissiere:
> * Dirk Eddelbuettel <edd at debian.org> [2005-12-10 08:16]:
> > Undoubtedly many of the CRAN/BioC packages would build and work just
> > fine. But the problem is that a fair number require hand-holding to, say,
> > properly translate CRAN dependencies into Debian dependencies, make sure
> > those are all present and installable, do some extra work to prevent
> > building (for the Windows only packages), do follow-up work (CIGwithR
> > wants interaction with webserver, gnomeGui ships as a package but is
> > "only" a glade file that gets installed somewhere else, RScaLAPACK needs
> > a mess of library arrangements (or, rather, an upstream autoconf patch)
> > etc pp.  It should *really* work, not *just work*.
> >
> > [...]
> >
> > But I think we need to work a few more details out before we charge
> > ahead. I may be in the minority here and don't mean to block or veto
> > this, but who exactly would do the work?  You have your hands-full with
> > Octave, Steffen seems busy with his dozen plus packages, I have Quantian
> > plus a few dozen packages --- so who is going to run this? It will take
> > one or two hours of daily hand-holding.

> Is it really that amount of *_daily_* hand-holding, or is it just a lot
> of work to set up things and then it runs almost automatically?

> At any rate, I am not willing to spend time *_daily_* on this project. I
> think we need a volunteer here, perhaps someone willing to become a
> Debian developer who is not yet maintaining other packages.

I want to thank Dirk for providing the detailed information. From my 
perspective I am mostly thinking about the BioConductor packages and the CRAN 
packages would be coming with it. For BioConductor, with most users executing 
>source("http://www.bioconductor.org/getBioC.R")
within R for an automated install, a preparation of prebuilt packages through 
Debian to my perception might be very very close or better than what a 
regular user or a novice expects. Debian would allow to omit the build 
environments and the libraries would be in place with no manual care. Dirk 
pointed out very vividly that this does not hold for several of the packages 
in CRAN. The gain is mostly for novices and power users who are distributing 
their software over many machines.

Would the following be acceptable? The script today caters for the distinction 
of the distributions in contrib and main, depending on the license. 
Analogously we could have a default distribution in experimental. We would 
still go through all the packages distributed in CRAN and BioConductor.org, 
but move any we do not consider as up to a certain standard for inclusion 
with a Debian main into experimental. A package's move from experimental -> 
main on alioth could then be initiated by the upstream maintainer, 
improvements to the script or the addition of requirements to Debian.

Today's script takes some time, indeed, though I am with Rafael, the Debian 
BioC maintainers shoud have things either mostly automated or should not be 
doing it. Improvements should mostly be done to the upstream packages 
whenever appropriate.

Many greetings

Steffen