R-beta: design && hmisc && CRAN

Kurt Hornik hornik at ci.tuwien.ac.at
Sun Aug 23 16:20:26 CEST 1998

On Fri, 14 Nov 1997, Jim Lindsey <jlindsey at luc.ac.be> sent an email to
r-devel describing his first shot at porting Design and Hmisc to R.  A
few days later, his port snapshot of Design was uploaded to CRAN's
src/contrib/devel area, and has since been available from there.

A few weeks ago, Andreas Weingessel <A.Weingessel at ci.tuwien.ac.at> had a
look into porting Hmisc to R (we needed some code for imputation).  When
he checked with the author, he was not given permission to ``port'' the
code.  (I guess everyone interested has meanwhile had a look at the most
recent version of the Hmisc copyright.)  As, on the other hand, Jim's
port was made from a version of Design which was obtained from Statlib
and had been released under the GPL, CRAN went on carrying the port.

We will now remove this port, so that within a few days it should have
disappeared from all CRAN sites.  Maintainers of CRAN mirror sites, pls
check.  We will do this mostly as there seems to be no point in having
an (even incomplete) R version of outdated S(-PLUS) code.

I would also like to support Bill Venable's advice NOT to try building R
versions of the packages from scratch.  In the past, porting code has
not only helped in increasing R's computational power, but also in
getting us substantially closer to the plain S (v3) standard.  With a
little effort (and we remember Paul Gilbert saying that it is worth it
anyway) one can have S packages which work under various S-PLUS versions
and R.  Several key developers are now providing R ``versions'' of their
code themselves.

What I am trying to say is that in the future, the R community should
spend time devoted to ``porting'' code wisely.  One criterion is whether
the author of the original code is willing to incorporate compatibility
changes if necessary and in the future ``maintain'' compatibility
through defensive programming.  Another is whether the software is
really needed under R.  Altogether, it might be better to come up with
entirely new solutions.

r-help 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-help-request at stat.math.ethz.ch

More information about the R-help mailing list