Massachusetts Institute of Technology
Dept. of Electrical Engineering and Computer Science
Spring Semester, 2008
6.01: Introduction to EECS I
Comments on exploration 2
We apologize for returning exploration 2 so late. You'll see a score
on your paper, but there are almost no comments. The scoring is as
follows:
- A: prime number sieve
- B: numerical integration
- C: ordinal and cardinal preferences
- D: experiences with your program
- E: modularity
Each part was worth a maximum of 2 points, to make points maximum
possible for this exploration.
Most people who tried parts A and B got results. To get 2 points, you
needed to show test cases; otherwise the answer got 1 point, even if
it was correct.
The most common error on the primes problem was programs that simply
tested numbers n for divisibility by all numbers between 1 and n-1.
That gives the primes, but it's not the sieving method: you
need to remove composite numbers from the set of test divisors as you
go along.
For integration, many people write the programs, but didn't include any
discussion of the error. In particular, the Simpson's Rule results
can be confusing, because the error in the numerical approximation
algorithm can be masked by the error in Python's arithmetic and
printing for small dx.
Answers to the behavior priority questions were mixed -- many people
did extraordinarily good jobs here. The main reasons for losing
points were failing describe how that program behaved (part E:
exploration 7).