[Rd] Packages in R & Java
A.J. Rossini
rossini at blindglobe.net
Fri Feb 27 19:23:14 MET 2004
"Marsland, John" <John.Marsland at CommerzbankIB.com> writes:
> Clearly, I disagree in some areas. But you raise an interesting, but perhaps
> unintended, question:
>
> Do we think that R is a programming language in its own right where we seek
> to develop code and then build on that code for example in a object oriented
> way?
My answer is: "we think that R is a programming language in
its own right where we seek to develop code and then build on that
code". Whether object orientation is a good approach is a completely
different issue. (note: I've spent significant amounts of time with
S4 classes on 2 projects, this may color my attitudes in non-obvious
ways).
> Or, is R a mechanism for combining lots of C/C++/Fortran libraries (written
> by suitably qualified experts) at the command line in an easy to use
> interpreted fashion in order to crunch the numbers and display them nicely?
For the general setting, think Python and SWIG. Why not both?
> Surely, if it's a speed issue we should try to help to make R faster for
> these sorts of problems rather than just bailing out into C at the first
> opportunity?
Are you volunteering? I'll never find the time in the near future.
Generalized optimization is very hard (or: high risk for possibly
little gain). Specific examples/contexts are easy, of course, with
the ease narrowing as the scope of speedup narrows.
R has improved beyond where I thought it would go back when I first
started using it (8 or 9 years? its been a while), but
across-the-board improvements in performance have always taken time.
--
rossini at u.washington.edu http://www.analytics.washington.edu/
Biomedical and Health Informatics University of Washington
Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email
CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}}
More information about the R-devel
mailing list