mns@doc.ic.ac.uk (Sattary) (07/26/90)
As part of an automatic coursework marking system for Modula-2 programs in our undergraduate courses, we are trying to construct a style marker. We'd be very interested to hear what people in the Modula-2 community consider to be the most important criteria for style in Modula-2 programs, i.e. what should this program look for in a good/bad program. For example, average length of identifiers, use of comments, and use of local as opposed to global variables are some of the things we have thought of so far. On the same line of thought, we are also thinking of writing a program that detects plagiarism. (We have a large number of students => this is impossible to do by inspection!). We are not thinking of anything overcomlicated. A simple measure that catches most cases would be sufficient. The criteria could include number of variables and procedures (without regards to their names) and possibly the size of the resulting objects/executables, preferably encompassed into a single value. Any suggestions in this directions will also be gratefully accepted. Replies by email ONLY please (walls have ears and students read news). Thanks in advance, Mohammad. -- Mohammad Nejad-Sattary phone: +44 71 589 5111 X 7539 fax: +44 71 581 8024 Department of Computing,Imperial College,180 Queens Gate,London SW7 2BZ, UK Janet: mns@gould.doc.ic.ac.uk UUCP: mns@icdoc.UUCP (or ..!ukc!icdoc!mns) DARPA: mns@gould.doc.ic.ac.uk (or mns%uk.ac.ic.doc@nsfnet-relay.ac.uk)
mns@doc.ic.ac.uk (Sattary) (07/26/90)
As part of an automatic coursework marking system for Modula-2 programs in our undergraduate courses, we are trying to construct a style marker. We'd be very interested to hear what people in the Modula-2 community consider to be the most important criteria for style in Modula-2 programs, i.e. what should this program look for in a good/bad program. For example, average length of identifiers, use of comments, use of local as opposed to global variables are some of the things we have thought of so far. On the same line of thought, we are also thinking of writing a program that detects plagiarism. (We have a large number of students => this is impossible to do by inspection!). We are not thinking of anything overcomlicated. A simple measure that catches most cases would be sufficient. The criteria could include number of variables and procedures (without regards to their names) and possibly the size of the resulting objects/executables, preferably encompassed into a single value. Any suggestions in this directions will also be gratefully accepted. Replies by email ONLY please (walls have ears and students read news). Thanks in advance, Mohammad. -- Mohammad Nejad-Sattary phone: +44 71 589 5111 X 7539 fax: +44 71 581 8024 Department of Computing,Imperial College,180 Queens Gate,London SW7 2BZ, UK Janet: mns@gould.doc.ic.ac.uk UUCP: mns@icdoc.UUCP (or ..!ukc!icdoc!mns) DARPA: mns@gould.doc.ic.ac.uk (or mns%uk.ac.ic.doc@nsfnet-relay.ac.uk)