[BioC] Installing Biocondutor under Debian r

Johannes Hüsing johannes.huesing@ruhrau.de
Sun, 1 Dec 2002 11:54:44 +0100


On Fri, Nov 29, 2002 at 05:28:41PM -0800, A.J. Rossini wrote:
> >>>>> "johannes" == Johannes Husig <derwisch@panix.com> writes:
> 
>     johannes> Cheers,
>     johannes> after installing Bioconductor on my Debian system I was
>     johannes> surprised to see that the new files were installed in
>     johannes> the /usr/lib/R tree. 
> 
>     johannes> Under Debian, this is considered offensive behaviour.
>     johannes> Debian assumes that the /usr tree (save /usr/local) is
>     johannes> only touched by the distribution, and that non-distributional
>     johannes> packages are to be installed in the /usr/local tree. 
> 
>     johannes> So people who use Debian and related distributions should
>     johannes> be advised to change the paths according to their system.
> 
>     johannes> Many things run automatically under Bioconductor so sometimes
>     johannes> users are tempted to perform things that are not in line with
>     johannes> the philosophy of the underlying system.
> 
> 
> This is actually true of ANY R INSTALL under the Debian system, unless
> you specify a local install.  Doesn't matter if it's Bioconductor or
> not!  It is a well known problem to the Debian maintainer, and
> discussion as to the "right" solution is still on going...
>  

Yep, it's a bit like TeX, where distributions are nicely packaged which
makes a re-packaging by the Debian distribution slightly redundant. I can
sense a problem where the solution is not trivial.

> Sounds like you installed under root, eh?  (or via sudo).  Else, you'd
> get an error.

Yes.

> 
> I've got Debian packages for Bioconductor (takes care of some of the
> associated requirements, such as PostgreSQL for AnnBuilder, etc).
> I'll make the archive available as soon as they get vetted by a
> developer I've been working with (the debian R maintainer)....

That'd be very helpful. Thanks a lot!

Greetings


Johannes
-- 
Johannes Hüsing   There is something fascinating about science. One gets
hannes@ruhrau.de  such wholesale returns of conjecture from such a 
                  trifling investment of fact.                Mark Twain