Path: utzoo!attcan!uunet!clyde.concordia.ca!news-server.csri.toronto.edu!helios.physics.utoronto.ca!ists!yunexus!oz@sprawl.yorku.ca
From: oz@sprawl.yorku.ca (Ozan Yigit)
Newsgroups: comp.lang.scheme
Subject: Re: Reference request:  Prog.Lang, An Interp-Based Appr.
Keywords: review
Message-ID: <1990Sep19.111312@sprawl.yorku.ca>
Date: 18 Sep 90 14:43:12 GMT
References: <1990Sep18.213646.5209@athena.mit.edu>
Sender: news@yunexus.YorkU.CA
Reply-To: oz@sprawl.yorku.ca (Ozan Yigit)
Organization: York U. Communications Research & Development
Lines: 46

I don't know if I posted this micro-review in this newsgroup, but the
latest discussion of lisp syntax, and a strong scheme/lisp orientation
of this book makes this a worthwhile re-post.

enjoy.	oz
--------------

Programming Languages: An Interpreter-based Approach
Samuel Kamin
Addison-Wesley	ISBN 0-201-06824-9

Source archive-site: cs.uiuc.edu [128.174.252.1]
Source archive-directory: pub/kamin.distr

This is a book that details programming languages through a study of Apl,
Lisp, SmallTalk, Scheme, Prolog, Clu and Sasl. [some my favorites :-)] 

The really unique thing about the book is that it presents a simple (?)
interpreter for each language that highlights the functionality of that
language. Each interpreter is a modification of a simple language 
[functions, control structures, variables] interpreter with a lisp-like
syntax.

The choice of languages is especially interesting: these really cover an
extraordinary collection of concepts, though there are some similarities
between some of them, and they are bundled where they are related: scheme
and sasl, clu and smalltalk etc.

Discussion of each language is done in its lisp-like form [which nicely
hides all the "real life" syntax crud that sometimes tends to overshadow
the really interesting concepts -- nowadays known as C++ syndrome ;-)],
with plenty of *complete* examples highlighting the characteristics of the
language, followed by a section on implementation [i.e.  modifications to
the basic interpreter - each interpreter is written in pascal, and full
source code included in the appendix] and a section on "the real language".
The discussion of the languages is very thorough. The author does not
short-change the reader just because there is complete source somewhere in
the book. I was pleasantly surprised to find an excellent discussion of
scheme compilation, garbage-collection strategies with source (mark-sweep,
stop-copy, reference-counting), a section on lambda calculus, ML, C++ and
few other gems.

In my view, An excellent book. Good mix of theory and practice, and is
understandable to boot, all with source. It is highly recommended.

--------------