News for hackers. Stuff that matters.
Re:The Shadow Scholar by skybrian on Saturday November 13, 2010 @07:15PM (#1902054)
Hmm, I suppose one way to go would be to require students to use a version-controlled word processor for their writing, and submit the version history. It could still be faked, but the difficulty of creating a good fake would make the economics less favorable.

If it's badly faked, then the version history would provide evidence that's more likely to hold up than a teacher's suspicion about the student's writing.

Re:The Shadow Scholar by robryan on Saturday November 13, 2010 @11:37PM (#1902543)
Make them write a smaller version in a test environment from only say a couple of pages of notes they can bring from their long paper. Style changes would quickly become apparent.
Re:The Shadow Scholar by forensic on Saturday November 13, 2010 @07:27PM (#1902072)
So the shadow scholar submits his version history. And then?
Re:The Shadow Scholar by hardy263 on Saturday November 13, 2010 @09:55PM (#1902369)
A 75 page paper usually takes longer than 48 hours for normal students to write. If the version history for the paper is submitted, but all the commits are within a very short time frame, then either the student stayed up all night to write it, or they paid someone else to stay up all night to write it. And if you ask them about what the wrote on on something they took all night to write, they should know it fairly well. But if they paid someone else to do it, they wouldn't know anything that's on the paper.
Re:The Shadow Scholar by gwern on Saturday November 13, 2010 @10:18PM (#1902407)
So then this big company hires a programmer to write a tool to randomize entries over a week or a month or whatever, and our shadow scholar has 1 additional step when he finishes up.

Basically, this idea is DRM in another guise, and terrible for all the same reasons. If you don't trust the student that much, then simply abandon take-home work in general.

Re:The Shadow Scholar by skybrian on Tuesday November 16, 2010 @10:15PM (#1912739)
Randomized entries would be obviously fake. To be a convincing fake, they have to make sure it doesn't conflict with the student's work schedule, class schedule, and so on. That means deciding when the student was supposed to have been writing the paper, which requires quite a bit more coordination and drives up the cost.
Re:The Shadow Scholar by mitcheme on Wednesday November 17, 2010 @03:20AM (#1913440)
You're assuming that the student doesn't own a laptop (or at least doesn't bring it to lectures) and that TPTB know the student's work schedule.