Submissions to CRAN

Friedrich Leisch Friedrich.Leisch@ci.tuwien.ac.at
Mon, 4 Jan 1999 15:42:38 +0100 (CET)


HAPPY NEW YEAR to all of you!

To clarify some questions in the recent thread on the timeslab
package (sorry that I couldn't participate in the discussion, but I
normally don't read my email on weekends):

*) Nobody should be able to download from the incoming section at our
site. That it was possible (again) was due to recursive link and some
lazy sysadmins which now have some more things they can try to make
better in the new year.

*) Please `announce' only packages that are really available somewhere
(and an incoming directory should not be considered such a
thing). This could either be

  -) your private server: nobody stops you from putting there whatever
     you want

  -) submissions to CRAN only after they have been moved by one of us,
     especially if it is a first submission. experience shows that
     first submissions very often need a little bit of `polishing' until
     they conform to CRAN standards, hence there are often time
     delays (and the already announced package cannot be found on CRAN).

*) I know that the packaging standards for CRAN may seem a little bit
pedantic, but only this way the `R INSTALL' interface can work without
troubles. Hence, the standards make life a little bit harder for
contributors, but we think it makes life much easier for the average
user (and these are the ones with less knowledge on how to fix
things).

*) On copyright for ports: This seems like a neverending discussion
and as I'm a mathematician and not a lawyer I can only contribute my
2(brand new Euro-)c:

Original code under GPL or similar: Port also GPL and you're done

`free redistribution for non-commercial use': I think it's OK to put
such things onto CRAN ... CD-vendors probably run into troubles.

The StatLib submission guidelines say that all code needs:

      A statement (which should also be part of the submission) that
      explicity gives permission to redistribute the software.  You
      can retain the copyright jsut so long as you clearly state that
      the software can be freely used for non-commercial purposes and can be
      freely distributed.
      If you are not the original author of the submission please make sure
      you obtain the original author's permission before submitting
      the software.

Now what exactly `redistribute' means (only original file or
derivatives) seems an open question to me ... and I think we never
settled this ...


In any case, asking the original author seems like a good first step
to me.


Feedback appreciated (please only to r-devel!!!),
Fritz

PS: After this thread is over we will make official CRAN submission
guidelines (with special emphasis on Statlib ports) and put something
in the FAQ.


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._