[pkg-bioc] CVS %p

[Steffen Moeller]

Hallo Dirk,

On Saturday 19 May 2007 20:23, Dirk Eddelbuettel wrote:
> On 19 May 2007 at 20:01, Steffen Moeller wrote:
> | Maybe you can transfer some bits to the UseR paper. When is the
> | respective deadline? Do you have a particular emphasis in mind?
>
> Last Monday was the deadline for paper, but the organisers are friendly.
> Worst case it'll be a poster...

Uh. I feel booked until that paper is updated. I can devote Wednesday to it, 
though, if this is not too late.

> | The other issue is with how we present ourselves to the public now. We
> | should have some nice web page, but is there somebody volunteering for
> | it? Or should it just remain with the Wiki?
>
> Fine by me until we really have an automated service that would call for
> more visibility.

The casual visitor of the pkg-bioc.alioth.debian.org page should feel safe. 
From my observation we are already fairly complete. Those packages that are 
missing (or that are having external build dependencies added in cran2deb) or 
not working as anticipated (CGIwitR) may be of interest for the paper.

I personally do not think that every package should be available as a Debian 
package. From the BioC I have removed those Berkeley genomes that happily 
take a couple of 100MBs in size.

> | In a vain attempt to install Axis under Tomcat I have apparently crashed
> | my machine which I can only reboot on Monday. What comes to mind... does
> | your heart very much hang on the .tex files that are distributed with the
> | packages? I very much wished these would be left either in a
> | documentation packages or ... for simplicity ... omitted rightaway. If
> | you'd agree to start with the omission, should we add this to our
> | cran2deb.pl or rather to your cdbs magic?
>
> You mean for every CRAN/BioC package?  It so happens that there was just a
> discussion about 'vignettes' on r-devel. I think the Sweave-style, R and
> latex based documentation is here to stay.  We should include __where
> feasible__ If it doubles the pbuilder time just for texlive, then don't. I
> hate it now for R builds because it adds a lot of time. For my (inside
> Debian) r-cran-*, no additional load is added -- so I am unsure why it
> affects pkg-bioc. Don't we just call 'R CMD INSTALL -l ...' as well?

We do. I do not recall though how dynamic the dependency to latex is 
determined. Also, I am not speaking against building Sweave/TeX documents, 
but against distributing the raw files themselves with the binary package. It 
think that it saves quite some disk space and looks better.

Many greetings

Steffen
-- 

Dr. Steffen Möller
University of Lübeck
Institute for Neuro- and Bioinformatics
Ratzeburger Allee 160
23538 Lübeck
Germany
T: +49 451 500 5504
F: +49 451 500 5502
moeller at inb.uni-luebeck.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.alioth.debian.org/pipermail/pkg-bioc-devel/attachments/20070519/194dcad7/attachment.pgp