[R] The Origins of R AND CALCULUS

[Peter Dalgaard]

Mark Difford wrote:

> 
> It would have been very easy for Mr. Vance to have written:
> 
> John M. Chambers, a former Bell Labs researcher who is now a consulting
> professor of statistics at Stanford University, was an early champion. At
> Bell Labs, Mr. Chambers had helped develop S, THE PROTOTYPE OF R, which was
> meant to give researchers of all stripes an accessible data analysis tool.
> 

...except that it would be wrong in about as many ways. (In fact, 
referring to S (v.3) as "the prototype" was an internal R Core joke for 
quite a while.) Two major points:

- S-PLUS was at the time a strong commercial product, not a prototype of 
    anything, and calling it that would be disrespectful to quite a few 
people working for and with StatSci/Insightful/TIBCO and their 
international distributors, as well as the Bell Labs/Lucent Technologies 
group. It couldn't touch the "dinosaurs" SAS and SPSS, but it did reach 
a level of more than 100000 licenced users. It took several years for R 
to get to a credibility level where it was even known outside some 
narrow academic circles.

- S compatibility was not a primary goal of R. The original plan was for 
a Scheme-like language with "syntactic sugar" making in "not unlike" S. 
The potential for running existing S scripts with minimal modifications 
drove R much closer to S than originally anticipated.  This of course 
does not mean that the current R should not acknowledge its substantial 
S heritage, just that if you want to describe the early history of R 
accurately, you do need to choose your words rather more carefully.

-- 
    O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
   c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
  (*) \(*) -- University of Copenhagen   Denmark      Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)              FAX: (+45) 35327907