Michael Gunn, a "student" at the University of Kent in Canterbury, UK, was caught plagiarizing from the internet just before his final exams. He now states his intent to sue the university for not clearly telling him that this plagiarism was wrong earlier (he claims to have done the same plagiarism for the whole three years at school.)
Surprisingly, Teresa Nielsen Hayden actually thinks the student has a point, because student handbooks are never read by the students. I say this is surprising, because I've always been impressed by Teresa's thoughts before, but this is a horrible defense. Yes, some college students do not read the academic handbook, or the plagiarism warnings on syllabi, or the citation directions on a writing assignment. But this is the university setting, where we expect the students to take responsibility for their own education. For the first time in their lives, these students have chosen where to go, and what to study. They also decide when to study, when to sleep, and when to party, usually making bad decisions. They also get to decide whether to inform themselves on university policies that can affect their chances to graduate. Michael Gunn is either baldly lying, or decided not to arm himself with all the knowledge he would need to do well at the university. That is his problem, not the university's.
On comments at both Teresa's post and a rebuttal by Kieran Healy, some professors talk about how difficult it can be to catch the clever plagiarist. The stupid ones are easy, as the writing style is markedly different and the sources are easy to spot (Google, usually). In my case, the tricky cases involve music theory assignments of partwriting or roman numeral analysis. When I get two students with the exact same (wrong) answers, I suspect that they were collaborating, but I cannot prove it. My solution to this problem is to allow/ignore collaboration on these simple homework sets, but to weight the exams higher. The homework is meant to help the students in mastering various topics. If the collaboration is genuine, then both students should do well on the exam. If one student was mindlessly copying from the other, then he/she should do poorly on the exams.
When I was in grad school a fellow TA experienced an amazing example of plagiarism. In our second-year theory class the students were required to compose a musical work that used specific harmonies and form. This composition was stretched throughout the semester, with numerous drafts and consultations with the teacher. One student was resistant to any changes her instructor (the fellow TA) made to her work, which can be understandable in creative activities. But her final draft ended up being an exact copy of a marimba piece by Evelyn Glennie. The plagiarism was discovered in a fluke, the TA just happened to pick up this set of marimba pieces because another student was performing them and asked a question. The first student had clearly decided to cheat on the project at the beginning of the semester, and this composition was a rather large part of the overall grade.
Finally, here at DePauw academic policy requires any case of cheating to be reported to the administration, and the penalty has to be more than just a zero on the offending assignment. I have not been through the process here yet, so I don't know what it is really like. But DePauw wants all professors to treat cheating with the same types of penalties, thus the requirement that all cases must be formally reported. I appreciate their sentiment, but it does remove most of the flexibility from the professor and makes it more likely for many professors to ignore plagiarism altogether, just to avoid having to write a formal report.