Principles of Programming Languages
CSCI 3155, Fall 2003

Table of contents

• Administrivia
• Course goals
• Text
• Course requirements
• Grading
• Collaboration policy
• Mailing list
• Calendar
• How to submit assignments
• Interesting links
• Worksheet (version 2): a set of questions prepared by Christoph to help you prepare for your exams
• Acknowledgements

Administrivia

Goals

You will learn concepts that will make it easier for you to learn new programming languages.

Why? Off the top of my head I've used the following languages, for real work: Alpha Assembler, awk, bash scripting language, C, C++, FORTRAN, Hypercard, Java, Lisp, Modula-2, Modula-3, Pascal, Python, SML, and VAX Assembler.  In many of these languages I've written tens of thousands of lines of code. Over the course of your career, you too will use many, many languages, and you will need to be able to pick a new language quickly, without somebody having to teach it to you. In this course, you will learn concepts that allow you to zoom in quickly on some of the common ways languages differ, so as to be able to relate a new language to ones you already know, and speed up the learning process.

 You will learn how to evaluate a language for a particular purpose.

Why?  Different languages have different strengths and weaknesses.  A language may be very suitable for one task but too cumbersome for another.  This is the reason why I've used such a wide variety of languages in my career!  Just as you’ll be learning many languages in your career, so you’ll be making many language choices. 

 You will learn how to design a new language.

Why? You may not realize it, but the chances are pretty good that you will design one or more languages during your career. I’ve designed many small languages during my career; you will see one of them in this class (called mystery).  Programming languages are pervasive and many large applications today have little languages of their own (e.g., for writing macros).  If you work in the software industry chances are high that you will design and implement a number of small languages during your career.

One of our faculty (Evi Nemeth) asked her former students what computer science courses they found particularly invaluable in their jobs.  Many of them named the language design and implementation courses they had taken at CU.

You will improve your ability to work collaboratively.

Why? "A recent survey of employers of graduates with bachelor's degrees in engineering found that of eight employee attributes, the one which the greatest number of employers valued highly was teamwork-- and engineering graduates were less prepared in this area than in math and sciences" [Bucciarelli and Kuhn, 1997].

On a more immediate note, researchers report that students learn more effectively and efficiently when they work together on learning with their colleagues.

You will improve your ability to learn from written materials.

Why? Learning can be easy (though often slow) if someone is available who can spell out for you everything you need to know. But in real life there’s often no-one available to do this, and/or no time for the teaching to be arranged and carried out. Especially in today’s information-rich world, you need to be able to get what you need from books and from the Web. You may already be very good at this, but some of you are not. If you are not, you’ll get better in this course.

You will improve your ability to learn and use abstract ideas.

Why? You can know a lot of things without seeing the big picture. In the world of programming languages people have developed concepts and terms that are convenient for representing the big picture, or at least parts of it. These concepts save you a lot of work, once you master them. But learning the concepts, and using them, is quite different from getting to be skilled in different specific languages. For some of you, the concepts will be easy to grasp and work with. Others of you will work at this and improve.

You will learn how to focus on learning, rather than just on satisfying course requirements.

Why? The point of having homework assignments and projects in a class is that you should learn from them. But it’s an easy confusion to think that the point of an assignment is to turn in what you are asked to turn in, rather than to learn. Here’s a common example: an assignment has some reading and some exercises. You only read the parts of the reading that is necessary to do the exercises. What’s wrong with that? You didn’t read all of the material, and hence you didn’t give yourself a chance to learn from all of it.

You will increase your level of interest and enthusiasm for Computer Science.

Why? While there are exceptions, few people excel at things they don’t enjoy.  There’s fun in understanding some of the bright ideas that show up in programming languages. There’s fun in the creativity of design. There’s fun in working together to learn powerful ideas that aren’t  obvious. We’ll all be working to share these experiences.

You will help improve this course and other CS courses, now and in the future.

How? My collaborators (primarily Bill Waite and Michele Jackson) and I, with funding from the National Science Foundation, have been studying new approaches to teaching and learning in a range of Computer Science courses, including this one. The methods and approaches we’ll be using this semester are based in part on this work, and studying what works and doesn’t work for you will help improve our courses in the future.

Methods and Tools

The Conversational Classroom. Sitting in a room listening to somebody talk is not the best way to learn (and much research backs this up). This course will de-emphasize listening to me talk and emphasize your learning ideas from resources out of class, and discussing problems and questions in class. We’ll try to make our classroom a place of active conversation, rather than passive listening.  You can find out more about the conversational classroom by reading this article.

The PL-Detective. The PL-Detective is a tool based on our (my collaborators and myself) experience in teaching computer science courses.  The PL-Detective aims to make learning programming language concepts fun while attempting to address many of the goals enumerated above.  Click here to get a paper describing the PL-Detective.  We will use the PL-Detective not just for assignments but also for classroom demonstrations.

Texts

Course requirements

There will be weekly assignments and several of the assignments will involve programming.  All assignments will be due at 12 noon on Tuesday (i.e., before class) unless the assignment specifies otherwise.  No extensions will be granted except in the case of emergency, in which case the student must provide appropriate documentation.

There will be two in-class midterm exams.  All exams are closed-book.

Students are responsible for the material covered in the readings, the material covered in the assignments, and the material covered in the classes.  I will expect you to have read the assigned pages before coming to class.

Grading

For the non-code parts of an assignment/exam we will consider the following:

• How well your submission answers the questions.  For example, a common mistake  is to give an example when a question asks for an explanation.  An example may be useful in your explanation but it should not take the place of the explanation. 
• How clear is your submission?  If we cannot understand what you are trying to say then we cannot give you points for it.  Try reading your answer aloud to yourself: that is often a great way to identify holes in your reasoning.  

For the code parts of an assignment/exam, we will consider the following:

• Does your code do what it is supposed to do?
• The quality of your code (documentation, structuring etc.).  This is particularly important if your code is incomplete or buggy: if your code is clean, well structured and documented, then we will be able to give you partial credit. 

Collaboration policy

We strongly encourage you to work together in learning the material.  As a matter of fact, most of the assignments will involve group work.  If a student outside your group helps you in your assignment, you should note it in your submission.  As a matter of fact, we encourage you to talk both to your group members and other students when learning the material or doing the assignments.  If your submission includes quotes from a book, paper, or web site, you should clearly cite the original source at the point of the quotation.  Bottom line: feel free to use whatever resources that are available to you as long as you cite them in your submission.

Mailing list

Please sign up for the mailing list by filling out this form. Feel free to post any relevant question/opinions to this mailing list by sending mail to: csci3155@cs.colorado.edu.  I often post important and time-critical announcements to this list so please make sure you get on it as soon as possible.  You can get to the archives of this mailing list by clicking on the "archives" link on http://www.cs.colorado.edu/mailman/listinfo/csci3155

Calendar

Date	Topic	Reading
8/26	Introduction
8/28	Evaluating languages, language paradigms	Chapter 1
9/2	Assignment 1 due
9/2	Syntax	3.1, 3.2, 3.3
9/4	Syntax continued
9/9	Assignment 2 due
9/9	Names and Bindings	5.1, 5.2, 5.3, 5.4
9/11	Types and type checking	5.5, 5.6, 5.7
9/16	Assignment 3 due
9/16	Scopes	5.8, 5.9, 5.10, 5.12
9/18	Data types, particularly Arrays	6.1 to 6.6
9/23	Assignment 4 due
9/23	Data types, particularly records and unions	6.7, 6.8
9/30	Data types, particularly pointers	6.9
9/25	Midterm 1 in class	Cumulative
10/1	Set types in recitations	Set type notes
10/7	Assignment 5 due
10/7	Evaluation order, overloading	7.1, 7.2, 7.3
10/8	Set theory in recitations	Set theory notes
10/9	Subtyping	Subtyping notes
10/14	Type conversions, short-circuit evaluation	7.4, 7.5, 7.6, 7.7, 7.8
10/14	Assignment 6 due
10/16	Iterative and conditional constructs	Chapter 8
10/21	Parameter passing modes	Chapter 9 upto and including 9.5.3
10/21	Assignment 7 due
10/23	Type checking parameters	9.5.4 upto and including 9.7
10/28	Midterm 2 in class	Cumulative
10/30	Implementing Subprograms Activation records: a place to put procedure-local data Dynamic links, return address, and static links: what they are and why they are needed What happens on a call and return How to access local and non-local variables on the stack	10.1 upto and including 10.4.2
11/4	Generic subprograms Motivation:  preserving type information while achieving code reuse  Examples	9.8 to 9.10, inclusive
11/4	Assignment 8 due
11/6	Abstract data types Motivation: why useful Grouping data and its operations + information hiding Examples in C++ etc.	11.1 to 11.4.3 inclusive
11/11	Object oriented concepts Inheritance: single and multiple Dynamic binding Subclassing = subtyping?	12.1, 12.2, 12.3 Also would be worthwhile to refresh your memory on classes in Java or C++
11/11	Assignment 9 due
11/12	Recitation: Java interfaces	Interfaces (read the pages referred to by the links "What is an interface", "defining an interface", "implementing an interface", "using an interface as a type"
11/13	Object oriented concepts in real languages Importance of separating interface from implementation  C++: issues with multiple inheritance Java: interfaces	12.4, 12.5, 12.6
11/18	Assignment 10 due
11/18	Implementation of O-O concepts v-tables and dynamic dispatch Challenges with multiple inheritance	12.10
11/20	Exception handling Motivation Examples in C++: try-catch	14.1, 14.3
11/25	Exception handling continued try-finally: why and what checked and unchecked throws clauses and subtyping	14.4
12/2	Introduction to functional languages Concepts: first-class functions, referential transparency, functional forms, recursion instead of iteration Simple examples in SML	15.1, 15.2, 15.3, 15.7
12/4	Standard ML (Type inference) Pattern matching Parametric polymorphism	Sections 2.1 upto and including page 39 of the SML/NJ tutorial
12/4	Assignment 11 due
12/9	Continuing with Standard ML More examples Equality types
12/11	Assignment 12 due
12/11	Review
12/13 7:30 a.m.	Final exam	Cumulative

How to submit assignments

All assignments should be submitted electronically.   Following are directions for submitting assignments.

• Click here to access the submission page.
• Log in using your login name (this is the same one you use to log in to CU machines) and password (the password is initialized to the last 5 digits of your student ID).  Use csci3155 as the course.  All students that were registered by 24th August are already in the system.  If you registered after that, please send mail to the instructor or TA with your information (student id, email, first name, last name).
• Click on the assignment you want to work on in "Pick the assignment you want to work on".  If it is an old assignment and you want to see your grade, click on that assignment in the "Pick an assignment to view feedback".
• Click on "fill out form" .
• On this page you will have a text window for each question.  For each question, we identify it by giving the first line of the question.  You will need to look at the assignment description to see the full question.  
• Paste your solution in text format to each question and click on "submit".  Note that you can also type your solution directly into the forms rather than pasting it in.  However, the advantage of the pasting is that you will have another copy of your solution on your local machine in case there is a web or network problem.

Some useful things to know about online submission

• The submission system will time out your session if you have been idle for one hour.  "idle" means that you have not clicked on any link or button (NOTE: typing does not change your idle status as far as the system is concerned).  This may be a particular issue if you are typing your assignment directly on the form: if you have typed up your assignment but haven't clicked submit or any other button/link for an hour, you may lose your assignment!
• Every time you click on "submit" it saves a copy of your solutions.  You can click it as many times as you want.  Your assignment isn't really "submitted" until the assignment due time passes.
• The system will stop accepting your assignment as soon as the due time passes.  

Interesting links

Acknowledgements

This web page and course has benefited greatly from the web pages and experience from earlier versions of this class taught by Michael Main, William Waite, Martin Hirzel, and Clayton Lewis