The best thread you'll ever see in a Computer Science forum

[Kyteland]

https://secure.csse.uwa.edu.au/run/help2140?p=np&a=197&all=y

A cautionary tale to all of you aspiring computer scientists "working in groups." They are watching you.

The graph in the second post is classic.

Edit: Link evidently went down. I pulled this from my cache. Graph

 

[her209]

Pwned!

 

[rcpratt]

Hahahhaha omg hilarious.

We were always threatened with similar software for our MATLAB work in our intro to engineering class.

 

[Scouzer]

Australians are such shitty ass typers

 

[gimmewhitecastles]

seriously i dont understand how those checkers are accurate. everybody gets the same assignment and they expect everbody's code to have no similarities? keep drinking koolaid.

 

[alkemyst]

What's funny is they never said what the threshold was, yet most are getting defensive and throwing out "I don't know them".

Thing is I have seen printouts of A work code being passed around...it was so bad in one class I asked to do it in VBScript instead of C++. It was the workings of the CPU/Registers.

In my database class, 10 dumbasses didn't even bother to change the code at all, right down to the non-standard variable names the original student used.

They were publically removed from the class.

Cheaters suck.

 

[alkemyst]

seriously i dont understand how those checkers are accurate. everybody gets the same assignment and they expect everbody's code to have no similarities? keep drinking koolaid.

They do...not 100% though.

 

[Gibson486]

seriously i dont understand how those checkers are accurate. everybody gets the same assignment and they expect everbody's code to have no similarities? keep drinking koolaid.

depends what the assignment was....

if it's something like create an ASCII chess game, yeah, this software would work.

 

[Saint Nick]

Lol!

 

[ivan2]

they shouldn't even bother with those 90%. 100% basically means the idiots will go in and change the variable name of the program. my professors used to threaten those people into dropping the class instead of failing them.

what noobs don't realize is that they still have the hw in the past few years, they will compare yours with those too.

 

[eldorado99]

Was hoping for more lulz, but not bad.

 

[gimmewhitecastles]

depends what the assignment was....

if it's something like create an ASCII chess game, yeah, this software would work.

when i was taking cs courses the assignments were straight forward enough and related to recent lectures that i could see some people getting the same solutions.

but posting the users who supposedly got caught is crass. he totally deserved that facebook ownage. :awe:

 

[Gibson486]

when i was taking cs courses the assignments were straight forward enough and related to recent lectures that i could see some people getting the same solutions.

but posting the users who supposedly got caught is crass. he totally deserved that facebook ownage. :awe:

Definitely, If it was something like, create code to filter a signal, yeah you will get lots of similar results.

 

[Leros]

Definitely, If it was something like, create code to filter a signal, yeah you will get lots of similar results.

You might get similar results, but I doubt you'll have solutions that are 80-100% the same.

 

[slayer202]

the graph is down, damn

 

[apac]

How exactly do these checkers work? Is a 100% a word-for-word copy? Even working in groups, the profs usually make it explicitly clear that you can work together to solve a problem but must code up the solution yourself. Individual work is individual work.

If a 100% means the program structure is too similar, that's stupid. If a 100% means a binary diff of the files is exactly the same, ok yeah that's blatant plagiarism.

 

[ObscureCaucasian]

The graph in the second post is classic.

Link doesn't work (to the graph).

 

[Leros]

How exactly do these checkers work? Is a 100% a word-for-word copy? Even working in groups, the profs usually make it explicitly clear that you can work together to solve a problem but must code up the solution yourself. Individual work is individual work.

If a 100% means the program structure is too similar, that's stupid. If a 100% means a binary diff of the files is exactly the same, ok yeah that's blatant plagiarism.

A good checker would be a text-based diff ignoring whitespace, comments, and symbol names (students often copy code and change variable and method names).

 

[PimpJuice]

What's funny is they never said what the threshold was, yet most are getting defensive and throwing out "I don't know them".

Thing is I have seen printouts of A work code being passed around...it was so bad in one class I asked to do it in VBScript instead of C++. It was the workings of the CPU/Registers.

In my database class, 10 dumbasses didn't even bother to change the code at all, right down to the non-standard variable names the original student used.

They were publically removed from the class.

Cheaters suck.

Didn't you cheat on your wife?

 

[alkemyst]

Didn't you cheat on your wife?

Get the fucking story right dumbass.

I told my wife I was going to get a girl friend.

That is not cheating...what a tool.

 

[AstroManLuca]

Link is severely lacking in lulz.

EDIT: drama removed

 

[ivan2]

How exactly do these checkers work? Is a 100% a word-for-word copy? Even working in groups, the profs usually make it explicitly clear that you can work together to solve a problem but must code up the solution yourself. Individual work is individual work.

If a 100% means the program structure is too similar, that's stupid. If a 100% means a binary diff of the files is exactly the same, ok yeah that's blatant plagiarism.

the checker is basically half of a compiler, it is much more sophisticated than diff. it parses the code into a tree and then compares them. 100% means the parse trees is the same. The the output (binary) won't necessary be identical though, as you can set your constants differently.

 

[Printer Bandit]

anyone have the graph?

 

[Possessed Freak]

During my senior year we truly group collaborated. The members of our group made 4 different sets of solutions in 4 different classes and shared them with our members. This way I didn't have to study 3 of the 4 classes during the semester. Come exam time we just got together and the person who did the work would tutor us to pass the exams. I did the java programming class. It was fun actually coming up with 4 different solutions to the same problem, using loops/subroutines or nested if statements. I did reserve the right to use the most efficient method for myself (as did the others who took the best reports/whatnot for themselves).

Eh, got me through 23.5 credits in one semester (+internship).

 

[Kyteland]

Link doesn't work (to the graph).

Fixed and added to OP.