To communicate with the teaching assistant, please send mail to com1205-grader@ccs.neu.edu.
The course home page is at URL: http://www.ccs.neu.edu/home/lieber/com1205/w01
The Demeter home page is at URL: http://www.ccs.neu.edu/research/demeter
My home page is at URL: http://www.ccs.neu.edu/home/lieber
In /course/com1205 you find a wealth of information in earlier versions of this course.
Required book:
Recommended books (depends on your interests):
A recent message from a reader of the course textbook summarizes what the course tries to achieve:
Date: Mon, 20 May 1996 15:49:38 -0400 (EDT) To: demeter@ccs.neu.edu Subject: Thanks for Demeter The subject really gives it away. Thanks for Demeter. A few weeks I stumbled across the book in a local bookstore and it somehow drew my attention. Now after having studied it I find myself employing the philosophy in the projects I am doing. ... I think this book for the first time provided really convincing material in favor of OO programming. I especially like the structure shyness of the approach. ... Although I follow the methodology I am more than eager to use the tools. Mark van der Voort May 20, 1996 Eindhoven, Netherlands
The following message indicates that what you learn is not only useful for Java/C++ programming. It is a programming language independent.
Subject: Java version of Demeter Thank you for all the work you have been doing. It has had a profound impact on my understanding of objects, and has helped me educate and train many Smalltalk students. Demeter seems to be that combination of the best science with the finest aesthetics in the service of a practical art which is good engineering. It deserves recognition. Are there any plans to port the tools to Java? This would be a very popular platform, I am sure. Steve Parker From sparker@well.com Fri Mar 8 01:23:37 1996
The course teaches the concepts behind adaptive object-oriented programming and allows you to exercise the concepts by using an implementation which is built on top of Java.
In this course you will learn how
The purpose of the Demeter Method is to teach the important concepts of object-oriented software development in a programming language independent way. Knowledge of the Demeter Method will be useful in the following ways after this course:
The Demeter Method is known world-wide through papers in OOPSLA '88, '89, '90, '91 '95 (workshop), 98 ECOOP '90, '91, '92, '94 ICSE '88, CSM '89, '93, ISOTAS '93, ER '90, CASE '92, IFIP '92, ISOTAS '93, IEEE Data Engineering '93, IEEE Computer '88, IEEE Software '89, IEEE Transactions on Software Engineering '93, IEEE Transactions on Knowledge and Data Engineering '93, Theoretical Aspects of Computing '94, Communications of the ACM '94, Journal of Software Engineering '91, Journal on Lisp and Symbolic Computation '88, ACM Transactions on Programming Languages and Systems '95, European Symposium on Programming '96, Foundations of Software Engineering '96, The Science of Computer Programming '97, ICSE '97.
The Demeter System is included or has been included in the course programs of National Technological University, OOPSLA, ECOOP, ELECTRO, Finnish and Swiss Summer School, MIT Summer Session. It has been taught at Northeastern University during the last 8 years.
The goal of the course is to significantly improve the software development skills of the attendees by allowing them to write their software at a high level of abstraction.
The course teaches object-oriented software development, specifically adaptive object-oriented software development with the Demeter Method during the first 5 weeks.
Object-oriented software engineering techniques are widely recognized as improving programmer productivity and reducing software maintenance costs. This course will introduce you to state-of-the-art techniques and concepts in object-oriented analysis, design and programming. Many of the concepts will be introduced in a programming language independent style; however, all your programming exercises will be done in Java. You will learn a novel and useful system for object-oriented design and programming, called the Demeter System which is developed at Northeastern University and you will have opportunities to influence its development. The Demeter System consists of two parts: the Demeter Method and the Demeter Tools. The Demeter Method provides the theory and a systematic approach to object-oriented programming and is supported by the Demeter Tools. The course is designed around behavioral objectives which make it easy to learn object-oriented design and programming techniques in a systematic way.
From file
ftp://ftp.ccs.neu.edu/pub/people/lieber/Demeter-interest
THIS IS A COLLECTION OF MESSAGES ABOUT
THE USE AND POTENTIAL OF THE ADAPTIVE
SOFTWARE TECHNOLOGY AND THE DEMETER TOOLS/C++
==============================================
The high-level approach of Adaptive
Software is very interesting. I believe the biggest benefit in using
Adaptive Software is the next level of abstraction up from Object-Oriented
programming.
Doug Sprague
IBM Microelectronics
Dec. 1995
===================================
From the ACM Computing Reviews, October 1995, page 528.
About the Communications of the ACM paper:
``Adaptive Object-Oriented Programming using Graph-Based Customization'',
by Karl Lieberherr, Ignacio Silva-Lepe and Cun Xiao,
{\em Communications of the ACM}, May 1994, pages 94-101.
Quote from review:
"If you have been writing object-oriented programs
and are interested in what comes next, this paper
is certainly for you."
Wayne Summers, Las Vegas, NM
Oct. 1995
=========================================
I might add that propagation patterns are a way to
specify flexible (C++) database programs which adapt to a broad class
of schemas or, another way to look at it, allow the application programmer to
specify code that does not need to know in detail the database schema.
Given that in practical C/S application for MIS, DSS, TP and in general
mission critical applications, the underlying database can have a very complex
schema, this has the potential, in my opinion, of greatly simplifying
writing C++ mission critical C/S applications.
Professor Ugo Gagliardi
Harvard University
May 30, 1995
======================================
From an anonymous reviewer of a funded NSF proposal on
"Engineering adaptive software"
(award for $295000 to Northeastern University)
I have followed the work of Professor Lieberherr for a number of
years, and have been consistently impressed by the approach and the
results. He and his group have had many good ideas in the area
of object-oriented programming and development methodology, and have
published them. The Law of Demeter is widely recognized as an important
stylistic guide. They also have an excellent record of going beyond
the ideas and their publication to implementation and use in
an educational environment and beyond. The Demeter System supports
at least two object-oriented programming languages, extending its
benefits to programmers in those languages without requiring them to
learn a new one. It has been used in many classes at Northeastern
University, and has been requested by many other organizations.
Several tutorials have been offered on the Demeter Method, and it
has been included in a survey of object-oriented analysis
and design methods undertaken by the Object Management Group.
This proposal deals with a critically
important area of research. The object-oriented paradigm is
widely believed to provide major benefits, and is becoming increasingly
widely used. Actually achieving the desired benefits of reuse,
extensibility and ease of maintenance, however, is no easy matter.
Simply using object-oriented style is not enough. Many researchers
therefore study object-oriented methodologies to find ways to
improve those characteristics. Many of the approaches, such as the
popular and valuable "frameworks" approach, allow the good designer to
produce a unit of software that is highly reusable in certain planned ways.
Such a unit of software will not easily be reusable in unplanned ways, however.
Only a few approaches to object-oriented software engineering
concentrate on supporting unplanned extensions. These tend to require
extensions to the object-oriented paradigm, so they are not currently
"mainstream", but it is my belief that these approaches will be of paramount
importance as object-oriented software engineering matures, and will
be instrumental in achieving its potential. Prof. Lieberherr's
"adaptive software" approach is one of these.
Anonymous NSF reviewer
=================================
It has long been a goal in the software development community
to link the design documents and the implementation code more directly.
The Demeter System is a powerful tool that helps designers
accomplish this goal. Because so much of the conversion from design to
implementation is made automatic by Demeter, when errors are detected, the
designer is encouraged to modify the design rather than tinker with the
low level C++ code. In the context of a software design course,
this methodology encourages students to think and design on an
abstract level rather than on the level of a
particular programming language such as C++.
In our view, this is precisely the kind of thinking
that computer science educators have been trying to instill in
students for more than 20 years.
...
Given the fact that students can learn only a single methodology
in detail [in an eleven week term], we believe that Demeter is a particularly
suitable one for teaching purposes. The reason is that Demeter is based
on theoretical studies of the optimal way to design loosely
coupled classes for maximum effectiveness and flexibility.
These theoretical principles can be used by students in whatever design
environment they happen to work in. Comments by students who have completed the
Software Design and Development course indicate that this is in
fact the case.
Richard Rasala
Professor and Associate Dean of Undergraduate Education
College of Computer Science, Northeastern University, Boston
May 1, 1995
=================================
My approach was to define the Library System class dictionary. Once the
class dictionary was designed, the next step was to define the operations
that would be performed on that class dictionary.
Lastly I designed the propagation patterns to implement those Library functions.
Although this class was my first introduction to Object Oriented Design
and to C++, I found that the Demeter System and adaptive S/W to be extremely
flexible and a very powerful technique. This flexibility was important
especially in this project when I had to make changes (of which there were
many).
I never had any problems with the basic Demeter tools: gen-imake, gen-make,
make, run, make clobber, and make clean.
Peter Monteagudo
Hardware Design and Development Engineer
Raytheon Company
Missile Systems Division Labs in Tewksbury, MA
Dec. 11, 1994
=================================
Overall, we learned that adaptive software allowed us to implement a fairly
complex software system in a fast, flexible, and efficient manner. It reduced
the amount of time devoted to actual coding and debugging, and allowed us to
redesign parts of the package with very little effort.
Rafael David
Sarah Sullivan
Dec. 11, 1994
=====================================
The adaptive
software work by Karl Lieberherr and his group is
also very useful here because it's about telling you
something about the way the internal data structures
are arranged, but not too much about how they are
arranged.
...
and again, adaptive software plays in here
because it is about being looser about the carving of the world.
Gregor Kiczales, Xerox PARC,
in his invited presentation at OOPSLA '94
(see http://www.xerox.com)
==================================
From an anonymous course feedback form:
Question:
Do you agree or disagree with the view that class dictionaries and propagation
patterns are a design notation which you can manually translate into
the programming language which your employer has decided to use?
Give an explanation.
Answer:
Agree. Learning the adaptive software notations ...
has done wonders for my object-oriented programming without
the Demeter Tools.
Undergraduate student in the Northeastern
Software Design and Development class
========================================
The project I am working on ... is developing an object-oriented power
system model in C++ for ABB Network Control in Vasteras.
When the implementation of the model was almost done, I read your article on
Demeter C++ in the Feb. 94 issue of the C++ report. It struck me that with
a method and tools as you describe in the article, I would have been able
to reduce the amount of time I spent on implementing and debugging my
power system model drastically. It seems to me that the Demeter C++
system addresses exactly those aspects of C++ development that make
implementing and comprehending a non-trivial C++
application tricky and time consuming.
Is a commercial product ready today?
Martin Falkevall
Magistratsv 55 K324
S-226 44 Lund
SWEDEN
================================================
I believe that the Demeter approach is a significant advance
in the art of software development ... because it allows the construction
of object-oriented systems, but reduces the dependence of
programming on the schema of the object dictionary.
Terry Glagowski
Washington State University
June 7, 1994
==================================================
I've found Demeter to be very useful in regards to compiler design. For this
homework, I developed a grammar using Demeter and 'sem-check -i' for an
object-oriented language that is LL(1) compliant. The translation to EBNF
form was then trivial. sem-check's analysis and violation reporting made the
design of the grammar much faster.
The grammar was then ready for input into the compiler-compiler LLgen:
...
The scanner definition, designed for flex, is very similar to that of Demeter's
generic parser. In fact, I used many of the patterns in lex-DEM.l for my
scanner.
...
I know Demeter is useful for many more things besides compiler design,
but that is my main interest. Traversal and transportation specifications,
along with wrappers, make coding such things as intermediate-code generation
and optimization analysis much easier than tools such as yacc or bison.
Andrew Purtell
College of Computer Science, Northeastern University
June 9, 1994
-------------------------
In response to the CACM paper, May 1994, pages 94-101.
... Adaptive software leaves ...
the issues of partitioning the world into objects
and navigation between objects flexible.
I completely agree with your perception that this is a serious issue.
...
It is also one that holds a lot of interest for the group I work in ...
...
We discussed your CACM paper today at our weekly paper reading group,
and came up with a couple of ways of looking at your system other than the
"adaptive" description:
"It's a tool that lets you do OO programming directly in terms of class
dictionary graphs, bringing programming one level closer to OOA/OOD."
"It's a tool supporting more powerful forms of object relative reference
in OO programming than just slot access."
John Lamping
Xerox Parc
around May 19, 1994
=========================
"I am very enthusiastic about the Demeter System, and feel it represents a major
advance in software engineering (a field where "hard" advances are rare).
I think it will also transform the way a number of areas of computer science
are taught."
...
I've been using the software fairly heavily and have a few
comments on it. ...
(I've been using it to teach a class
and my students are using Demeter to design and build a text-adventure
game). It is a remarkable system with many potential applications beyond
the scope of software engineering (for instance Compiler Design can be done
very effectively using the Demeter System.)
"This is an extraordinary powerful software tool that will undoubtedly
become a standard fixture in Computer Science curricula (and in several areas,
including Compiler Design, Software Engineering, Artifical Intelligence)."
...
"Extremely valuable --- I predict it will sweep the industry"
Justin R. Smith
Drexel University
May 19, 1994
=========================
From a discussion on the patterns news group.
The Visitor class reminds me of the Demeter work by Karl Lieberherr and his
associates at Northeastern University. They're taking an approach that
involves abstracting the "meat" of the behavior of a complex construct from
the "skeleton" methods that deal with managing and navigating the
construct. Their work also addresses keeping the abstraction valid across
changes to the implementation of the construct. To do this, they use a
graph notation to represent the class structure, and a "propagation pattern"
language that tells, in a general way, how to distribute the meaningful
methods and navigation methods across the structure. A propagation pattern
may be applicable to many class structures, thus allowing changing the
latter without needing to modify the former. A code generator does the work
of creating the C++ code. A good introduction to this is in an article
called "A Report on Demeter C++" in the February '94 issue of C++ Report, by
Ignacio Silva-Lepe, Walter Huersch, and Greg Sullivan.
If you can't get to this periodical,
you might try emailing them at {nacho,huersch,gregs}@ccs.neu.edu.
Don Dwiggins "Things should be made as simple as possible,
Mark V Systems, Inc. but no simpler"
dwig@markv.com -- Albert Einstein
April 7, 1994
============================
I really think you are on to something with Demeter.
I think propagation patterns are very powerful.
...
I think that the best part of Demeter is the "adaptive" part. Being
able to modify the interface of a class or change the class hierarchy
without major problems is a huge benefit of propagation patterns. C++
and OO languages in general do not address this problem.
Joseph N. Coco, Software Engineer
TASC, 55 Walkers Brook Drive, Reading, MA 01867
Dec. 20, 1993