Re: [S] scope rules

Douglas Bates (bates@stat.wisc.edu)
25 Jun 1998 14:36:02 -0500


"David Nelson" <daven@llnl.gov> writes:

> Coming from the LISP community, I've always found the scoping rules for S to
> be quaint in the extreme, and a constant source of irritation to me. I have
> always wondered if the developers of S were aware of the research at the
> time on lexical scoping, but deliberately chose not to incorporate modern
> scoping techniques because they would conflict in some way with S's lazy
> evaluation and delayed commitment paradigm. Or, were they just unaware of it
> and were trapped in old-fashioned FORTRAN thinking?
>
> In any event, there is no conflict between structured programming,
> object-oriented programming, and lexical scoping (see, e.g, the work on
> Common LISP Object System (CLOS)). And the scoping rules surely were not
> used to enforce "security".

The R language, which its creators characterized as "not unlike S",
uses lexical scoping. See
http://www.ci.tuwien.ac.at/R/doc/misc/lexical.tex
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news