COD Rules on Seniors in 1.00
By Irene C. Kuo
The Committee on Discipline concluded the cases of seniors accused of turning in duplicate code on problem sets in Introduction to Computers and Engineering Problem Solving (1.00) last Wednesday, according to course professor Nigel H. Wilson SM '70.
Committee chair Sheila A. Widnall '60 refused to discuss the outcomes of the 10 cases, except to note that there would be a lot of "familiar faces" at Commencement and that the affairs of 68 other students in 1.00 remain to be examined in September.
As the COD will eventually present to the faculty what it has learned about unauthorized access to programs, committee members have welcomed the chance to examine the 78 cases, Widnall said. She did not foresee policy changes resulting from the COD's review.
The discovery in late April of duplicate code in 1.00, a class with
an enrollment of 240,
marked the largest instance of student cheating uncovered in MIT's recent history.
(Continued from page 1)
"We were actually very conservative in interpreting duplicate code," said Wilson of the computer program written to screen problem sets. "The cut-off was not arbitrary."
Duplicate code included identical statements, functions, number of functions, content, and individual statements within each function. As problem sets involved fairly long computer programs, Wilson ruled out the possibility that two students could have individually submitted identical programs.
Students get zeroes
on problem sets
Approximately 90 people were caught after all problem sets were examined. These students picked up letters from Wilson's secretary telling them of his intention to submit names to the COD and the number of zeroes they had received for problem sets in which duplicate code appeared.
Wilson did not require these people to meet with him, but about 80 did. As a result of the meetings, 15-20 names were dropped from the list Wilson was about to forward to the COD. He also became aware of a range of situations that could lead to duplicate code.
The most "benign" kind of behavior, according to Wilson, was working jointly on a problem set. "The educational value of doing problem sets is preserved," he acknowledged. "However, I had said that it was not appropriate to do joint programs."
For these students, Wilson modified the policy so that they received half credit. He still submitted their names to the COD, but with a note that they had collaborated on problem sets.
At the other extreme were those who had stolen other people's programs, Wilson said. A few students believed he was referring to people who had stolen hard copies of problem sets from trash bins in Project Athena clusters.
The middle ground, according to Wilson, was the situation in which one individual had done the work and others had turned it in as their own. As it was "obvious" which person had been responsible for the problem set, that individual received half credit, and the others got zeroes, Wilson said.
Changes in 1.00 likely
Wilson said that some changes to 1.00 were in order as a result of his experiences last term. "I went to each of the 18 recitations at the beginning of the term and told them what was expected of them individually and what constituted cheating," he noted. "In the future, I will write the policy in boldface, though one feels sheepish having to do both."
He also realized there are benefits to working with others and said that in the future, he would designate two or three out of the 10 problem sets as ones on which collaboration was allowed.
Wilson expected that the professor who teaches 1.00 this fall will use a program for spotting duplicate code on problem sets.
"I'm upset and the students were upset that we discovered duplicate codes so late in the term," he said. Some people had to petition to drop the class because Wilson announced his discovery to the class after drop date.
"Had I known in week three, I would have said something," he concluded.