R-beta: S-Plus compatibility

Jim Lindsey (jlindsey@luc.ac.be)
Fri, 20 Feb 1998 11:59:15 +0100 (MET)

From: Jim Lindsey <jlindsey@luc.ac.be>
Message-Id: <9802201059.AA21347@alpha.luc.ac.be>
Subject: R-beta: S-Plus compatibility
To: r-help@stat.math.ethz.ch (r-help)
Date: Fri, 20 Feb 1998 11:59:15 +0100 (MET)

Readers of this list should be aware that compatibility is a one way
street. If you are writing new functions (or libraries) and you
imitate S-Plus, you lose two of the most important features of R
(besides its being freely available with source): speed and superior
scoping rules.
  Porting code from S-Plus is one thing; creating new libraries is
another. A number of people have tried to port some of the simpler
functions of my four nonlinear and repeated measures libraries to
S-Plus. No one has succeeded. They do not work in S-Plus.
  The second question is compatibility with which S-Plus? Two months
ago, I wasted about five hours trying to port a library called design
to R. A month later, learned that much of the incompatibility was
between versions of S-Plus, not between S-Plus and R. design worked
with S-Plus v3.1 but apparently must not work with current versions!
  S-Plus is constantly mutating. Do we try to emulate that?
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@stat.math.ethz.ch