Lazy evaluation

Question 1

How are expressions in Haskell evaluated?

Answer 1

Lazy evaluation

Question 2

What are the benefits of lazy evaluation?

Answer 2

• Avoids doing unnecessary evaluation
• Ensures termination whenever possible
• Supports programming with infinite lists
• Allows programs to be more modular (separates control and data)

Question 3

If the program terminates, what can we say about the ways of evaluating the same expression?

Answer 3

Anyway of evaluating the same expression will give the same result provided it terminates

Question 4

What is a redex?

Answer 4

A reducible expression

Question 5

What are the two main evaluation strategies?

Answer 5

• Choose a redex that is innermost (doesn’t contain another redex)
• Choose a redex that is outermost (is not contained in another redex)

Question 6

Compare termination with innermost and outermost evaluation

Answer 6

Outermost evaluation may give a result when innermost evaluation fails to terminate

If any evaluation sequence terminates, then so does the outermost, with the same result

Question 7

Compare the number of reduction steps with innermost and outermost evaluation

Answer 7

Outermost evaluation may require more steps than innermost

Question 8

How do we reduce the number of steps in outermost evaluation?

Answer 8

Using pointers to indicate sharing of arguments

Question 9

What is lazy evaluation?

Answer 9

Lazy evaluation = outermost evaluation + sharing of arguments

Question 10

What does Lazy Evaluation ensure?

Answer 10

Termination whenever possible, it never requires more steps then innermost evaluation

Question 11

What is modular programming?

Answer 11

Seperating control from data
LE allows this