Is it possible to trace text from Word documents?
If i fx copy/paste a word from a copy of someones document, and pastes it in my own document, is it then possible for a program to trace it back to his document?
My school have a program which can detect if you have copy/pasted something from the net / classmates, i assume it does so by looking and comparing similarities, but the thought hit me, that it might be able to actually trace the texts source via code or some shit.
Enlighten me please, for i am blatantly retarded at this.
Name:
Anonymous2012-05-22 18:13
Not sure if you understood what im saying, but yeah... i'll explain it better if anyone need.
Name:
Anonymous2012-05-22 19:36
If you mean as an image, yes. Optical character recognition.
If you mean as some sort of readable text that is on the internet or in their database (e.g. including other students' work), yes.
If you mean using metadata in the file itself, maybe. Word documents are mostly comprised of cruft and formatting data, but they also include things like the author's name. Save a blank Word document, open it in a text editor (if it's a .doc) or unzip it then open it in a text editor (if it's a .docx) and search for your Windows username and real name. They're in there.
If you simply copy the file from George, it's possible the file will still contain George's name somewhere. If you copy and paste the content of George's file, it's a maybe. I don't know how Word documents work (and neither does anybody else, not even Microsoft).
Whether all these plagiarism programs go to such lengths, I don't know. They'd rely a lot on human judgment. I've always assumed they're just glorified automated googling programs sold to stupid institutions for ridiculous license fees.
Whether all these plagiarism programs go to such lengths, I don't know. They'd rely a lot on human judgment. I've always assumed they're just glorified automated googling programs sold to stupid institutions for ridiculous license fees.
I know stupid amounts about TurnItIn's `system` (as a TA), and you are right. It is basically a shitty Google. It seems to try to search for `statistically unique phrases` and long passages as plagiarism, but it's not uncommon to have it detect ``See spot run.'' type stuff as plagiarism, even though it is basic grammar that every paper would have. It also can't tell the difference between a quote and plagiarism, so the grader has to check if it is sourced and cited correctly.