Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!usc!rpi!batcomputer!llenroc!cornell!uw-beaver!milton!ogicse!pdxgate!parsely!percy!m2xenix!news
From: news@m2xenix.psg.com (Randy Bush)
Newsgroups: comp.lang.modula2
Subject: Re: Oh My! Modula-2 Book
Message-ID: <1991Mar16.070351.3692@m2xenix.psg.com>
Date: 16 Mar 91 07:03:51 GMT
References: <2177.27DB3A01@puddle.fidonet.org> <2441@sumax.seattleu.edu> <STUCKI.91Mar14120723@amygdala.cis.ohio-state.edu> <1991Mar14.191513.20184@iitmax.iit.edu>
Organization: Pacific Systems Group, Portland Oregon US
Lines: 30

> It is somewhat regrettable that Niklaus Wirth left the issue of libraries so
> open-ended that programmers cannot rely upon a minimal set of libraries from
> implementation to implementation.

I do not mean to put words into his mouth, and thus the following should be
taken as my insufficiently humble opinion only.

Professor Wirth often stresses that what one leaves out of a language is often
as or more important than what one puts in.  Somewhere, I also think he said
that one should only include what one knows the [one] simple correct way of
doing.

Thus, Modula-2, the language, omits that which is contentious, unclear, or has
multiple [good] solutions.  As Modula-2 provides for a library, what better
place to put all those things?

And, as someone who has done a few libraries, and watched the standardization
process's attemtps at libraries, I assure you that they're contentios, unclear,
and admit of many solutions (but few good ones).

At the last ISO Modula-2 meeting, I suggested that

   FROM SYSTEM (* or whatever *) IMPORT printf, ...;

would be far better, was already a standard, is quite easy to implement, and
will be more acceptable to the mass of users than anything the committee has
come up with to date.  Although few took me seriously (well, we were in the
pub), I fear that it is the sad truth.
-- 
Randy Bush  / news@psg.com  / ..!uunet!m2xenix!news