Newsgroups: comp.object,comp.lang.eiffel,comp.lang.c++,comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!howland.reston.ans.net!ix.netcom.com!netcom.com!NewsWatcher!user
From: hbaker@netcom.com (Henry Baker)
Subject: Re: Why is one OO language more productive than another?
Message-ID: <hbaker-1608951540100001@192.0.2.1>
Sender: hbaker@netcom16.netcom.com
Organization: nil organization
References: <40t027$7j9@hardcopy.ny.jpmorgan.com> <40tk13$qms@wcap.centerline.com>
Date: Wed, 16 Aug 1995 23:40:10 GMT
Lines: 49
Xref: glinda.oz.cs.cmu.edu comp.object:36912 comp.lang.eiffel:10408 comp.lang.c++:144314 comp.lang.smalltalk:27271

In article <40tk13$qms@wcap.centerline.com>, chase@centerline.com (David
Chase) wrote:

> > Conceptual integrity.
> 
> I'll assume that conceptual integrity means that a few concepts can guide
> your understanding of the entire language.  Whether this matters, in terms of
> productivity, I do not know, but I know that I like it.  For instance, I like
> BCPL and Scheme very much, because both languages are very regular and
> predictable in their mapping of syntax to semantics.
> 
> I think that "conceptual integrity" also is correlated with a concise
> language description.  Modula-3 was designed with an arbitrary constraint on
> the length of its description (goal of 50 pages) that forced the definers to
> aim for a more concise (and hence, in my opinion "conceptual") view.
> I think that this is perhaps good for productivity, if only because
> I can RTFM, and then keep the whole FM in my head.  If I fail to keep
> it in my head, I can look it up and refresh my memory quickly.
> 
> I'm mystified at the length of the descriptions of Ada, Eiffel, C++, and
> Common Lisp -- I can only surmise that the people working on defining these
> languages became so involved in their work that they failed to appreciate how
> complex the languages were becoming.

I think you have put your finger on the problem.  The people that seem to
gravitate towards standards committees seem to be those that can keep a
much bigger FM in their head than I can.  I am constantly amazed by their
ability to reel off a 3 paragraph nearly-word-for-word quotation from some ANSI
manual from the top of their head.  I think that this same kind of intelligence
is required to come up with the IRS tax code (and also comply with it).

Unfortunately for the rest of us, we either can't keep the entire FM in our
heads, or don't want to waste valuable neurons doing so, in which case a
50-page manual is at the upper end of reasonableness.

You can add lots of _libraries_, but the core language needs to fit
within 50 pages, max.

----

Perhaps the ANSI/ISO committees should only meet after each member has become
high on cannabis.  The well-known property of cannabis to reduce the famous
psychologist George Miller's 7 +- 2 items in short term memory to perhaps
1 +- 2 items might reduce some of the stuff that gets dumped into ANSI/ISO
standards.

-- 
www/ftp directory:
ftp://ftp.netcom.com/pub/hb/hbaker/home.html
