Comments on: Scuppering the Program Pirates http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html Science Blog from Freelance Science Writer David Bradley Sat, 07 Nov 2009 00:41:44 +0000 http://wordpress.org/?v=2.8.5 hourly 1 By: David Bradley http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html/comment-page-1#comment-577680 David Bradley Tue, 11 Nov 2008 07:42:11 +0000 http://www.sciencebase.com/science-blog/?p=2609#comment-577680 Yes, a reward/penalty system might work, but if we could entrench the idea earlier on that its wrong to plagiarise that would be better. I saw that <a href="http://www.guardian.co.uk/education/2008/oct/31/facebook-cheating-plagiarism-cambridge-varsity-wikipedia" rel="nofollow">news item</a> just after I posted this post and was going to reference it. Yes, a reward/penalty system might work, but if we could entrench the idea earlier on that its wrong to plagiarise that would be better. I saw that news item just after I posted this post and was going to reference it.

]]> By: Ian Kemmish http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html/comment-page-1#comment-577528 Ian Kemmish Mon, 10 Nov 2008 22:16:01 +0000 http://www.sciencebase.com/science-blog/?p=2609#comment-577528 In this age of "nudge" economics, why not simply offer extra credit for the fastest code (rewarding individuals and small collaborative teams), and demerits for code which runs no faster than the average (if you copy and don't know how many others copied, you'll most likely lose out)? Or the code which stands up longest to pseudorandom input? Etc.... Incidentally, a recent survey at Cambridge found 49% of students admitted to plagiarism, with lawyers being the most likely to copy! In this age of “nudge” economics, why not simply offer extra credit for the fastest code (rewarding individuals and small collaborative teams), and demerits for code which runs no faster than the average (if you copy and don’t know how many others copied, you’ll most likely lose out)? Or the code which stands up longest to pseudorandom input? Etc….

Incidentally, a recent survey at Cambridge found 49% of students admitted to plagiarism, with lawyers being the most likely to copy!

]]> By: David Bradley http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html/comment-page-1#comment-576243 David Bradley Fri, 07 Nov 2008 17:13:04 +0000 http://www.sciencebase.com/science-blog/?p=2609#comment-576243 @Sundeep - interesting approach @Bob - thanks for the input, yes I am sure there are more stringent assessment methods and definitions in the industry than in academia, it's a shame there isn't more collaboration between the two camps. @Sundeep – interesting approach

@Bob – thanks for the input, yes I am sure there are more stringent assessment methods and definitions in the industry than in academia, it’s a shame there isn’t more collaboration between the two camps.

]]> By: Bob Zeidman http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html/comment-page-1#comment-576236 Bob Zeidman Fri, 07 Nov 2008 16:55:09 +0000 http://www.sciencebase.com/science-blog/?p=2609#comment-576236 My company, Software Analysis & Forensic Engineering Corp., has created CodeSuite for determining source code correlation, the first step in detecting software plagiarism. Although our papers on the topic have been largely ignored by the academic community, the software is commercially very successful because it is many time more accurate that anything out of academia. You can even download a copy and use it for free on small code sets -- something that hundreds of users are doing including many university professors. Also if you're looking for a definition of software plagiarism that is much more rigorous and testable than anything out of academia, check out the papers references on the site. We define source code correlation and the define the steps necessary to determine whether correlation is due to copying or due to one of 5 other reasons. My company, Software Analysis & Forensic Engineering Corp., has created CodeSuite for determining source code correlation, the first step in detecting software plagiarism. Although our papers on the topic have been largely ignored by the academic community, the software is commercially very successful because it is many time more accurate that anything out of academia. You can even download a copy and use it for free on small code sets — something that hundreds of users are doing including many university professors.

Also if you’re looking for a definition of software plagiarism that is much more rigorous and testable than anything out of academia, check out the papers references on the site. We define source code correlation and the define the steps necessary to determine whether correlation is due to copying or due to one of 5 other reasons.

]]> By: Sundeep http://www.sciencebase.com/science-blog/scuppering-the-program-pirates.html/comment-page-1#comment-576074 Sundeep Fri, 07 Nov 2008 08:41:25 +0000 http://www.sciencebase.com/science-blog/?p=2609#comment-576074 The solution adopted by my school was to have final exams which required coding a small program by hand on paper (it's harder than it sounds). Another idea is to mangle their source code and require them to debug it in class. That would require a working knowledge of the programming language and debugging techniques. The solution adopted by my school was to have final exams which required coding a small program by hand on paper (it’s harder than it sounds). Another idea is to mangle their source code and require them to debug it in class. That would require a working knowledge of the programming language and debugging techniques.

]]>