Web Informant #248, 11 May 2001: Cheating vs. collaboration http://strom.com/awards/248.html [I]n an era when people swap music over the Internet, forward email messages and send texts to each other with a single keystroke, the lines between collaboration and theft have blurred. -- NY Times article yesterday The front-page article in the Times (and the original story in the Richmond paper, with a link below) blames the Internet as a factor in suspected widespread cheating of University of Virginia students taking an introductory science class taught by Lou Bloomfield: http://www.timesdispatch.com/pda/MGBK2CCABMC.html What attracted me to the story was the role played by Bloomfield, who developed a computer program to scan all of the papers he received to look for similar word combinations. The program found 122 matches in scanning over 2000 students' papers. And these weren't just a few words here and there: Bloomfield told me that all of the papers matched at least 500 or so words or about a third of their length. UVa has a strict honor code and the students could face expulsion from the university, or their degrees revoked in the cases of students who have already graduated. The cases are now being decided by the UVa Honor Code Committee. Bloomfield's scanning program went back several semesters, examining all the papers he received for the class. Buried deep within the story is one very interesting fact: all of these papers were REQUIRED by the professor to be submitted via the course web site. If you go to the site, you can learn about class assignments and read the very precise instructions for how the students should prepare their papers (including admonishments about not collaborating on the work). Bloomfield began requiring electronic submissions for the students' papers because "the paperwork was exhausting for me. Putting it on the web made the bookkeeping trivial. After all, all of my students do their work on computers already, and instead of printing the papers out, I get them electronically." http://rabi.phys.virginia.edu/106/ So my question is, did this encourage a climate for cheating? It certainly made it easier for the prof to scan the papers with a program, even though his program took several days to run through all the texts. (He told me that he originally wrote the program in perl but it was too slow, so he rewrote it in C++.) I can't be sure. It certainly is something to think about. Of course, many colleges already require their students to come to school with a computer as freshmen, so perhaps this requirement isn't so unusual. And those that don't have computers make extensive use of them to type up papers anyway. Now, don't get me wrong. If any students copied another's paper, they should be punished. But what about the student who originated the work and sent a copy via email? Is it wrong for someone to send someone else one's work? After all, the original author could claim he didn't compel any student to make a copy and submit it with their name on it. And Bloomfield mentions that several students have contacted him, saying that they were innocent bystanders and angry that someone lifted their works without their knowledge, perhaps from public computers or other places around campus. "If they are guilty of anything, it is trusting their fellow students too much," he said. I'd have to say that I feel that the original author should go free. The problem becomes, how do you prove who is the original author? Do you scan the web site logs for the class to determine the order of submissions? Fortunately, the pattern of cheaters is easy to detect, since the copied papers were always submitted in a later semester than the original works, according to Bloomfield's analysis. When I was researching this issue, I recall something back when I was working for PC Week at the beginning of my journalism career. Imagine my surprise when a competitor's paper lifted a quote out of my story and ran it without attribution and as their own piece. How can I prove this? Simple: both stories misspelled the source's name the exact same way. Hard to believe that someone could get the same exact quote and not get the guy's name right. And I still remember the anger that I had, after getting that initial scoop in PC Week, to see my hard-earned work printed elsewhere, without any credit. The seeds for these cheaters are sown early on. I don't have to look farther than my own daughter's research for her middle school projects to see it happening, as she liberally uses the cut and paste functions to lift stuff from various web sites for her assignments. I wouldn't call her a cheater per se -- but it is a difficult line to draw in the sand, or in the word processor, between wholesale lifting and just plain paraphrasing. In the old days with pen, paper, and the World Book Encyclopedia we had do a bit more work, but we still copied those passages just as if we had a cut and paste function. Granted, neither my daughter nor my plagiarist is the same level of offense as the UVa students we are talking about here. And what makes the whole situation interesting to me is not that the Internet makes it easier to steal ideas or entire assignments but that the way people create and collaborate on their original work has changed significantly. As Bloomfield says, "Students have forgotten the great art of going to the library and sifting through real books and looking stuff up. If it isn't on the web, it doesn't exist." This whole thing is somewhat unsettling to me, a writer who makes his living sending his work around the world via email. The blurry lines between theft and collaboration indeed. As Bloomfield told me, "Perhaps what everyone can take away from this experience is to be careful with your own intellectual property." And you can copy that thought down and share it around. To subscribe, send a blank email to: informant-subscribe@pez.oreillynet.com To unsubscribe, send a blank email to: informant-unsubscribe@pez.oreillynet.com Entire contents copyright 2001 by David Strom, Inc. David Strom, david@strom.com, +1 (516) 944-3407 938 Port Washington Blvd., Port Washington NY 11050 Web Informant is (r) registered trademark with the U.S. Patent and Trademark Office. ISSN #1524-6353 registered with U.S. Library of Congress.