Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.6.2.16 $; site ima.UUCP
Path: utzoo!decvax!cca!ima!compilers
From: compilers@ima.UUCP
Newsgroups: mod.compilers
Subject: Re: Compilers digest Vol. 1 Number 4
Message-ID: <136300025@ima.UUCP>
Date: Mon, 6-Jan-86 22:14:00 EST
Article-I.D.: ima.136300025
Posted: Mon Jan  6 22:14:00 1986
Date-Received: Wed, 8-Jan-86 02:57:36 EST
Lines: 40
Approved: compilers@ima.uucp
Nf-ID: #N:ima:136300025:000:2090
Nf-From: ima!compilers    Jan  6 22:14:00 1986


[from rmc at DECVAX/GENRAD/PANDA/TEDDY]

Organization: GenRad, Inc., Concord, Mass.

	For a long time, the two best / only books on Lisp implementation
seemed to be Allen's "Anatomy of Lisp" (Academic Press?) and Henderson's
"Functional Programming:  Application and Implementation" (Prentice Hall,
CAR Hoare series with red bindings).  I found Allen's text somewhat dry.
The main focus was on memory allocation techniques and dynamic binding
techniques (important questions to be sure, but...)  Henderson's text was
less detailed in such areas, but it also did a better job on lazy
evaluation and stream support.  It also has a very good definition of the
SECD machine, which is one of the abstractions used to build reasonable
compilers for Lisp.  You can also look in Darlington, Henderson & Turner
"Functional Programming and its Applications" for some nice articles by
Wise and Sussman on optimizing interpreters (the book itself contains
general articles from a U of Newcastle on Tyne seminar, and includes
information on typed ffp by Guttag and other goodies).

	Recently (Nov 1985?) MIT Press published Gabriel's "Performance and
Evaluation of Lisp Systems".  It really has much more low level
implementation information available than the title indicates, but i have
only skimmed this one and don't know it thoroughly.

	A major problem with most of the above is that they don't give you
any details about lexical scoping.  You don't really want to do a Lisp with
dynamic scoping any more - there are just too many programming techniques
that this makes difficult.  Sussman and Abelson's "Structure and
Interpretation of Computer Programs" (MIT Press again) is a freshman text 
introducing programming, but does show some of the discipline.  A better
source are the Steele and Sussman papers on Scheme from the MIT AI Lab (try
AI TR 453 may 1978 "The Art of the Interpreter",  527 nov 1979 "The Dream
of a Lifetime:  A Lazy Scoping Mechanism" and/or Steele's thesis AI TR 474
"Rabbit:  A Compiler for Scheme").

	Enjoy!
					R Mark Chilenskas
					decvax!genrad!panda!rmc
--------