Copyrights for R contributed libraries

Douglas Bates
23 Jan 1998 16:03:58 -0600

I have been packaging both R and the R contributed libraries for the
Debian GNU/Linux operating system.  Right now if anyone installs
Debian Linux on an Intel machine or a DEC Alpha they will have the
option of installing R as a regular part of the operating system.  I
think this is a good way of getting wider distribution of high quality
statistical software.  The Linux community is growing rapidly and is
one of the highlights of the free software effort.  I have heard from
several people who started using R because it was included with 
Debian Linux.
Since vendors are encouraged to make CD-ROM's from the entire
distribution, the Debian organization is very careful about copyrights 
for the code.  The base R language is covered by the GNU Public
License (GPL) but some of the contributed libraries have other
copyright statements.  As a result I have divided the libraries into
r-cran, the contributed libraries that are covered by a copyright that 
follows the Debian Free Software Guidelines (DFSG), r-cran-non-free,
the contributed libraries that do not seem to be DFSG-compliant, and
r-mlbench.  That mlbench library is split into its own package because 
it is large and not everyone wants to install it.

I have just been informed (see below) that the copyright statement I
made about the r-cran-non-free package is insufficient.  I have to
clarify the copyright status of every library included in this
package.  I'm a little pressed for time right now and may not be able
to do this by myself.  My only option then would be to withdraw the
package.  I think that would be a shame because the package contains
very useful libraries like the survival analysis library.

I believe Kurt and Fritz are going to start to ask for contributed
libraries to include a statement of copyright as part of the
submission to CRAN.  That would help me a lot.  Could anyone
converting a library or writing a new library please take care to
include a copyright statement?  If you are converting a library and
the copyright status of the original code seems murky, please try to
contact the original authors and get the copyright status cleared up.

If we could quickly establish a format for the copyright file for a
submitted library and contributors could update their libraries, that
again would help a lot.  I don't like making work for others but I
think there are great benefits from this.  I believe that it is
helpful to have good cooperation between the Debian organization and
the R developers.

I personally prefer the GPL for code but the "Artistic" license under
which perl is distributed or the BSD-style license are also acceptable
for the regular Debian distribution.

Feel free to contact me if I can provide further information.
Douglas Bates                  
Statistics Department                    608/262-2598
University of Wisconsin - Madison
------- Start of forwarded message -------
Subject: Bug#17407: r-cran-non-free: incomplete copyright file
Reply-To: Christian Schwarz <>,
Resent-CC: Douglas Bates <>
Resent-Date: Fri, 23 Jan 1998 07:33:01 GMT
Resent-Message-ID: <>
Date: Thu, 22 Jan 1998 18:23:40 +0100 (CET)
From: Christian Schwarz <>
To: Debian Bugs <>
Message-ID: <Pine.LNX.3.96.980122182112.6087A-100000@monet>

Package: r-cran-non-free
Version: 0.61-2

The /usr/doc/r-cran-non-free/copright file reads:

Individual modules, installed under /usr/lib/R/library/, contain their
own copyright statements.  These modules considered not to be Debian
Free Software Guidelines (DFSG) compliant.

According to current policy, the copyright file should contain the
complete license texts. (This is important, since we are planning to set
up a license/ directory on which contains the copyright
files from all non-free packages.)



--                 Christian Schwarz
Do you know,,
Debian GNU/Linux?,
Visit                  PGP-fp: 8F 61 EB 6D CF 23 CA D7  34 05 14 5C C8 DC 22 BA

------- End of forwarded message -------
r-devel mailing list -- Read
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: