COMP 302: Lecture 1 Course Overview and Introduction to SML Joshua - - PowerPoint PPT Presentation

COMP 302: Lecture 1 Course Overview and Introduction to SML

Joshua Dunfield

4 January 2010 McGill University

www.cs.mcgill.ca/∼joshua/302

1

Overview

• COMP 302: “Programming Languages and

Paradigms”

• . . . is not a trip to the zoo
• . . . is about thinking rigorously about

programming languages and programs

• What does a program do? How do you know?

What does it mean?

• How do you specify a programming language?
• [How do you implement a programming

language?]

2

Course Mechanics

• Prerequisites: COMP 250 or COMP 203
• If you have not taken either course,

talk to Liette Chin (Trottier 2070)

• We have TAs
• Your grade is 50% homework:
• 5 assignments, mixing programming and

written questions

• Midterm in class, worth 10%
• Final worth 40%
• You get two late days for homework

3

Textbook

• Robert Harper’s Programming in Standard ML
• Download it from the course website
• Also a version for printing 2-up
• You will not need to print the whole thing

4

Slide-Free Zone

• This may be the last lecture with slides
• Code and notes will be posted for many lectures
• You are responsible for what is presented in lecture

5

Don’t Copy Code

• It is possible to design exam questions that are
• moderately hard if you did the assignments
• very hard if you didn’t
• Copying is unfair to students who don’t
• Copying is likely to be noticed
• It’s against the rules

6

Don’t Copy Code

• If you’re worried that you will fail,

talk to us

• I would rather give some big hints in office hours

than have anyone copy code

7

Standard ML

• High-level, functional, and statically typed
• From the 1980s (revised 1997)
• No, it is not widely used in the “real” world
• Yes, you can get a job using functional languages

8

History

• PL/I was supposed to be huge

(“PL/I belongs more to the problem set than the solution set” —E. Dijkstra)

• C++ was going to dominate forever
• Java appeared

Moral: No one knows what languages will be popular in 10, 20, 30 years.

9

Thinking on All Cylinders

• “The tools we use have a profound (and devious!)

influence on our thinking habits, and, therefore, on

• ur thinking abilities.” —E. Dijkstra
• There is no more important tool when

programming

• Even in Java or C++, it can be useful to “think in”

another language

10

Values and Effects

• SML is (primarily) a functional language
• Evaluation, not execution
• Values, not effects
• Binding, not assignment

11

Types

• SML is statically typed
• Types describe data:

integers, strings, lists of strings, lists of pairs of strings and integers, vectors, arrays. . .

12

Types

• SML is statically typed
• Types describe data:

integers, strings, lists of strings, lists of pairs of strings and integers, vectors, arrays. . .

• Types describe the behaviour of functions:

integers to integers, strings to integers, lists of things to lists of lists of things. . .

12

Types

• SML is statically typed
• Types describe data:

integers, strings, lists of strings, lists of pairs of strings and integers, vectors, arrays. . .

• Types describe the behaviour of functions:

integers to integers, strings to integers, lists of things to lists of lists of things. . .

• Types are checked by the compiler
• You’ll need to write types sometimes,

but SML will often figure them out

12

Demo

[SML demo—basicSML.sml]

13

First Assignment

• Basic SML programming with lists and trees
• Out Wednesday, Jan. 6 (we hope)
• Due Wednesday, Jan. 20

14

Zeroth “Assignment”

• Get a CS account (if you don’t have one)
• Get access to the SML of New Jersey compiler:

www.cs.mcgill.ca/~joshua/302/faq.php

• Start reading Harper, chapter 2

15