M.I.T. DEPARTMENT OF EECS

6.033 - Computer System Engineering Handout 33 - April 20, 2000

Design Project #2: Peer review details

Your team's report on design project 2 is due in recitation on Thursday, April 27. Your group should hand in two almost-identical copies at your designated recitation. The difference is that one copy should contain a title page with the names of the group members, while the second copy will omit the names. Both copies should have on their title page a 6-digit code number (nonce) that you have chosen at random to identify your report.

The copy with your names on the title page will go to your recitation instructor for evaluation. The second ("blind") copy will be given a moment later to another team in your section, so that the other team can prepare a short critique of it for recitation one week later, on Thursday, May 4. You will thus get an opportunity to perform (and receive) what is called a blind review; a particular form of peer review in which the reviewer does not know the name of the authors, and therefore is (presumably) less likely to let personal biases influence any recommendations. Of course, if you do some sleuthing you can probably figure out whose paper you are reading, but that would not be in the spirit of blind reviewing, so restrain yourself.

In preparation for the Thursday, May 4, recitation meeting, your design team should carefully read over the paper you have been given to review, write a one page review, and prepare a 5-minute at-the-blackboard oral critique. Choose one of your members to deliver this critique. (If you wish, you can divide up the presentation among two or three members, but if you do that you will need to work skillfully to keep from tripping over each other in the short time available.) You can assume that your audience is very familiar with the problem to be solved, and even has a pretty good idea about the design space, so you do not need to spend much of your time giving background or explaining the design in detail. Instead, quickly sketch the high points of whatever makes the design unusual, and move on immediately to explain what your group thinks are its major strengths and weaknesses, and whether or not you would recommend to MIT that this design should be seriously considered, perhaps with some suggestions for improvement.

Peer review is a commonly practiced method of reviewing designs, proposals, and professional papers. The goal of peer review is to be constructive, not destructive. Comments along the lines of ``this design is trash'' are out of order. Instead you should try to come up with comments along the lines of ``this design is weak because it doesn't handle X, but by incorporating Y it would be more bullet proof'' or ``This design is excellent, because it meets all of the stated goals and manages to add two neat features, Y and Z.'' You should try to enhance this goal by maintaining a professional, even collegial attitude during the presentation and ensuing discussion.

Your group's grade on design project two will not take into account your design and your oral critique of the other group's design. Instead, this review and critique will be considered as part of your recitation grade. And don't worry about negative recommendations and misunderstandings by the team that critiques your design; they do not affect your grade.


Go to 6.033 Home Page Questions or Comments: 6.033-tas@mit.edu