Paul Johnson, at 01:03 +0200 on Sun, 1 Apr 2001, wrote:
Without commenting on main theme of this thread, although I have plenty
of opinions on that too, and not wanting to open too many cans of
worms, may I simply mention that I hope we are not trying to cater too
much to the average programmer? There are already plenty of languages
that will do that.
I'll accept this.
By using a correct term, although it may be unknown to the average
programmer, the programmer is presented with an oportunity to learn, and
may even be exposed to a completely new programming paradigm.
This is true, if "pure" can be "correctly" applied to the notion of
"nosideeffects" in the context of standard Perl programming. While the
term "pure", surely can be deemed "correct" in the context of functional
programming, it cannot in standard Perl programming. This is not to say
one can't do functional programming in Perl, or anything of the like.
However, considering context in which most Perl is written, "pure" has no
meaning, and hence I wouldn't consider it "correct".
--
Frank Tobin http://www.uiuc.edu/~ftobin/