Subject: Re: Why typing?
From: Erik Naggum <erik@naggum.no>
Date: 1995/09/24
Newsgroups: comp.software-eng,comp.lang.functional,comp.lang.lisp,msu.cps.misc
Message-ID: <19950924T022023Z@naggum.no>

[Matthias Blume]

|   When I still was a Scheme addict I used to believe in this, too.  But
|   after programming extensively in SML I now have to say that this is not
|   true at all.  Early _complete_ error checking leads (at least for me)
|   to a huge decrease in development time.

I have to modify my previous comment.  I was mainly addressing "manual
strong typing" (à la C++), which I still think is horribly inefficient in
terms of programmer time, has been oversold many times over, and is
unlikely to yield any actual benefits when all costs are tallied.  the
"inferred strong typing" of SML, where the strong typing is not specified
by the programmer, therefore yielding the benefits previouly restricted to
dynamically typed languages (some compilers of which also do a healthy
amount of type inference), while relieving the programmer of the tedium of
spelling out things that will, of necessity, have to be changed later.

although I have been (lightly) exposed to SML, I must admit to considering
it as being in a wholly different ballpark.  maybe that's wrong, but as
long as (cop-out alert!) the common understanding of strong typing is that
of "manual strong typing", the only difference I (think I) have to make is
to qualify my comments as applying to the "manual" aspect of strong typing.

|   I don't know where you get the strong conviction that let's you express
|   your views with such emphasis.

simple answer: by not differentiating "strong typing" from "lots of typing".
(sorry.)

#<Erik 3020898016>
-- 
computers -- the only field where knowing the past
will make you less able to appreciate the present.