As Computer Coding Classes Swell, So Does Cheating (nytimes.com)

An anonymous reader shares a report: College students have flooded into computer science courses across the country, recognizing them as an entree to coveted jobs at companies like Facebook and Google, not to mention the big prize: a start-up worth millions. The exploding interest in these courses, though, has coincided with an undesirable side effect: a spate of high-tech collegiate plagiarism. Students have been caught borrowing computer code from their friends or cribbing it from the internet. "There's a lot of discussion about it, both inside a department as well as across the field," said Randy H. Katz, a professor in the electrical engineering and computer science department at the University of California, Berkeley, who discovered in one year that about 100 of his roughly 700 students in one class had violated the course policy on collaborating or copying code. Computer science professors are now delivering stern warnings at the start of each course, and, like colleagues in other subjects, deploy software to flag plagiarism. They have unearthed numerous examples of suspected cheating.

Comments

Nothing new here

[Anonymous Coward]

Borrowing, or reusing code, has always been the norm and is the basis for libraries of routines and procedures. Blatant ripoffs should be obvious but smaller scale plagarism (your word) is hard to determine.

Re:Nothing new here

[gweihir]

Often it is extremely easy to determine, namely from patterns of mistakes. I once, as a TA, had a case where I refused points for some exercises to several students. They complained, and then I showed them how a comma had become a dash and then had become a quote over several generations of copying (with the dash and the quote making absolutely no sense in the given context). That shut them up pretty fast.

Of course, for simple and correct code fragments you cannot actually detect cheating reliably. The easy solution is to make the exercises more complex. Students are coddled far too often in coding courses anyways.

Re: Nothing new here

[Anonymous Coward]

Sums it up. Or they get a friend to write it for them. I remember in college I was the only one in my computer science class to get a week long project right and have unique code. Everyone else either got it wrong, or all had the same code.

But I agree, kids are coddled in class. This is a very tough mental field. The campaigns to teach everyone to code make it seem like it's house work. Anyone can do it. I think they are finally realizing it's not so.

Re: Nothing new here

[gweihir]

This is a very tough mental field. The campaigns to teach everyone to code make it seem like it's house work. Anyone can do it. I think they are finally realizing it's not so.

I sure do hope so.

Re:Nothing new here

[XxtraLarGe]

Often it is extremely easy to determine, namely from patterns of mistakes.

It's even easier when it's a direct copy. I've had students who copied homework directly, including the same errors, and the same comments. One time, I had a student use another student's file and they didn't even bother to remove the other student's name out of the comments section. They both got 0's for the assignment, and it was a pretty hefty assignment to get a 0 on.

Re:Nothing new here

[XxtraLarGe]

Borrowing, or reusing code, has always been the norm and is the basis for libraries of routines and procedures. Blatant ripoffs should be obvious but smaller scale plagarism (your word) is hard to determine.

In my classes, the students are expected to complete the assignments using the techniques covered to date. That is because what we learn in the next class often relies on understanding what we covered in the previous when. So when they use a solution that is beyond what they have already learned, they are putting themselves even further behind the curve.

You can't imagine how many times I've had students make a hamfisted attempt to insert something they found on Stack Overflow that uses techniques not yet covered in the course. Since they don't understand what that code is supposed to do, it often doesn't work correctly, or crashes because they thought they could copy & paste without even compiling. I have no problem giving those students a 0 on their assignments.

Re: Nothing new here

[Anonymous Coward]

Anyone can code, it just needs to compile. Almost no one can program. Crap code is breaking in odd ways because of huge amounts of undefined behavior. Most of the time the programmer can't even tell you why their code worked in the first place, yet alone make it work against once it breaks. Large amounts of determinism allows most to plow through their lack of understanding by throwing code at the wall and seeing what sticks.

Re: Nothing new here

[pr0fessor]

Today everything is color coded and can't be installed in the wrong place or backwards, not like years ago when installing a stick of ram could fry everything if you put it in backwards... The barrier to entry is much lower for just about everything computer related and as kids now grow up with computers in their pockets from the time they are little, that barrier is going to just get lower. Eventually there will be few real programing jobs and a bunch of drop in ide code everywhere and it's not far off at all.

Re: Nothing new here

[DuckDodgers]

Bullshit. This is not a tough field. When CS101 and CS102 classes are hard, all you're doing is weeding out the kids who didn't have any coding experience in high school.

My family didn't own a fucking computer before I got to college. I wanted to work in the industry, but I new nothing. I passed my first three classes by cheating. The kids that aced had taken programming classes in high school and owned computers since they were in elementary school. By junior year I was all caught up. I'm still working as a developer, almost twenty years later. Java, Python, Perl, Scala; git; Maven; Jenkins; shell, vim, sed, piping. And I would have never got here if they did a better job catching cheaters at school.

All you're doing is weeding out the people that didn't start young.

Re:Nothing new here

[A. Craig West]

Really? When I was doing programming in university, I had already been programming for six years before that. I started very young...

Re:Nothing new here

[timholman]

In my classes, the students are expected to complete the assignments using the techniques covered to date. That is because what we learn in the next class often relies on understanding what we covered in the previous when. So when they use a solution that is beyond what they have already learned, they are putting themselves even further behind the curve.

And unfortunately for them, once they start down that path it is nearly impossible to turn things around. Every semester, I warn students not to copy homework answers from a downloaded textbook solutions manual. Inevitably I wind up giving failing grades to several students with excellent homework grades who have miserably failed every exam.

Some students start out with the best of intentions, then get behind and copy an assignment while promising themselves they'll catch up later. Of course, it never happens. Every assignment is simply more incomprehensible than the one before it. A month later, they're hopelessly lost.

Re: Nothing new here

[gweihir]

While I thing that this will still require quire some time and I think most frameworks make application coding actually more difficult (at least if anything goes wrong), I do agree that eventually only a few real coders will be left. But that is the same in other fields. For example, there are much more electricians than EEs (working as EEs), and the like. Of course, the only good jobs will be the ones where you write real code and everybody else will be easily replaceable.

Re: Nothing new here

[gweihir]

That, if true, is very far from the standard case. The standard case is that the "coder" is clueless because of lack of capability for insight. That is not a question of "doing the exercises".

Re:Nothing new here

[gweihir]

Ah, yes. Some students truly aim to attain new heights of incompetence. Leaving the name in is pretty "special" ;-)

Re: Nothing new here

[steveg]

Reusing code for some sort of production is good sense.

Reusing code that you are supposed to be writing in order to learn *how* to write it using certain techniques short circuits the learning process. You may get the code to do what you want, but the important result (you learned how to do it) doesn't happen. Nobody needs a program to pick from a menu to multiply two numbers, or to print out a Fibonacci sequence, or whatever. What they do need is a programmer who can create these things, because that programmer is on their way to create things that will actually be useful.

So reusing code for school assignments is stupid. You may or may not get caught, but either way, you didn't learn anything. So why even take the class?

Re:Nothing new here

[Anubis IV]

Same here. I used to have a group of students (from a part of the world where plagiarism is more culturally accepted) sit in the lab and work together. With me sitting just a few feet from them, I watched them as they copied code back and forth from each other's screens, so I made it clear that while they could work together in figuring out the algorithm, it was up to each of them to actually implement the algorithm. They nodded, made a gesture of following what I said, and then went right back to it again.

So, since their code ended up being identical, I gave them 0s.

After that, they stopped copying from each other while sitting in the lab, and instead took their cheating out of class. Thankfully, that was even easier to detect, since they'd have signature traits, such as the same number of irrelevant trailing spaces at the ends of specific lines. They started changing variable names and intentionally adding whitespace in irrelavant places, just to make their code look distinct. What they didn't know is that I had a plagiarism detection tool that accounted for those sorts of changes, and sure enough their code always came out as a 100% match for someone else in the class, while everyone else's would be around 20-50% (which was expected, given that there are only so many ways to implement simple tasks).

Much to my disappointment the professor refused to act on any of it, so the best I could do was give them 0s when it was abundantly clear they were cheating. Eventually they started changing the implementation enough that the tool I used no longer showed a 100% match, but at that point they were effectively re-implementing it themselves anyway, which was exactly what we wanted from them in the first place.

Re:Nothing new here

[Dan East]

Back in the early 90s I was an assistant to a CS instructor, and one of my jobs was to grade assignments (if you ever have the opportunity I strongly recommend doing so - it's amazing to see the very different ways students go about solving a problem, as well as the vast difference in coding styles, etc). It was blatantly obvious when students had collaborated, and I would point out to the instructor which students appeared to have shared code with one another.

Also at this point in time in the 90s, the ability to find source code for use in very specific programming assignments was very difficult if not impossible. The WWW was in its infancy, and the only real options were gopher or Usenet. In the case of the latter, it was very clear when people were asking others to do their assignment for them, and typically several posters would call them out on that.

Re: Nothing new here

[pr0fessor]

The use of frameworks, ide, and markup is common already and you don't need to be a rocket surgeon just need to have played with a model rocket... it's easy and fast though you are right if anything goes wrong only someone that really knows what they are doing can fix it.

Re: Nothing new here

[rogoshen1]

"if they made computers easy to use like the TV then people in the community could compete against the whites and the chinese" -- some random dude in oakland complaining about the lack of diversity in IT.

Apparently reality does not meet up with that gentleman's (or the idiots pushing the idea that everyone should code! #kodewithkarlee -- aka, ls ls ls ls cd .. ls ls) perception.

Because coding IS hard, and is something you have to be passionate about to be worth a god damn; cheating is the logical outcome. (don't even get me started on 22 year old snowflakes who gimped their way through school, and finally land an entry level job, then proceed to complain about how hard for loops are.)

Re:Nothing new here

[istartedi]

Yeah, it was about 4 or 5 years for me before university. If somebody is using techniques that weren't taught, they might be copy-pasting or they might be more experienced. I couldn't pass the AP test to get out of CS, so I still had to take it but I knew more than beginners.

I have mixed emotions about some of the down-grades I got. One time I got marked down for using the "wrong" kind of state machine by a TA. He showed me a cookie-cutter process (that we hadn't even been taught) and told me it was the expected procedure to use for generating the machine, and marked me down for not knowing that. Thinking about that still pisses me off. I deemed it unwise to waste the prof's time arguing about it. That's about as much politics as you get in engineering school.

OTOH, I turned in a program one time with self-modifying code because I thought it was elegant. I got slammed for that, and it was the right call. When I was that age, I thought it was "cool" that a program could modify itself. I guess it still is in a game or something; but not in secure, maintainable code.

Re:Nothing new here

[Hognoxious]

plagarism (your word)

Wrong. His word was plagiarism.

Re: Nothing new here

[The Snowman]

Eventually there will be few real programing jobs and a bunch of drop in ide code everywhere and it's not far off at all.

If Salesforce is any indication, the "drag and drop" model of designing a system is nowhere near ready for prime time. Even simple implementations require a real programmer to make them work, despite what they promise.

Re:Nothing new here

[maestroX]

It's even easier when it's a direct copy. I've had students who copied homework directly, including the same errors, and the same comments. One time, I had a student use another student's file and they didn't even bother to remove the other student's name out of the comments section. They both got 0's for the assignment, and it was a pretty hefty assignment to get a 0 on.

All nice and dandy, I remember getting this "copy" mark once without foul play (some frequency analysis flag) and it was, to say the least, pretty fucked to get at least the *opportunity* to redo the assignment, let alone the effort. . Be fucking sure it is foul play before condemning students--

Re:Nothing new here

[TimHunter]

With that level of stupidity a 0 for the assignment is letting them off easy. I'd just show them the door.

Re:Nothing new here

[pjt33]

Sounds like you started quite old. I'd been programming for ten years when I started university, and I'm sure someone's about to one-up me...

Re: Nothing new here

[pr0fessor]

I was thinking more along the lines of visual studio... but yeah some the drag-n-drop stuff out there is a nightmare.

Re:Nothing new here

[OzoneLad]

When I was a marking assistant, I would be assigned a portion of a class assignments to mark. Once, I had to mark two assignments where the code was identical, save for variable names. Both assignments had full documentation and printed out (this was the early 90s) test runs, which was a neat trick because one of the assignments had missed a variable when doing the search and replace, and shouldn't have been able to compile at all.

Re: Nothing new here

[barbariccow]

IDE? How the hell does anyone program anything without vim? I don't get it. Like... once you learn how to record (either via 'q' for complex recording, or '.' for "repeat last action" simple recording).... how do you ever go back to basically a colorful notepad?

Re: Nothing new here

[barbariccow]

Programming is just a dictation of logic. If you understand how you think and solve problems, you recreate it for a computer to do much quicker. Encapsulate like your mind encapsulates objects as distinct and concepts as distinct. Then, interact between the objects.

It's double-ended too. The more I program, the more I understand thought patterns. The more I think about thought patterns, the better I understand code. Hell, I can look at someone's code and know exactly how they see the world and interpret transformations.

Remember... all the knowledge of the universe runs on any wire at any time. It's only useful to us if we follow transformations and display information that a person can then understand. That's the core of programming.

Re: Nothing new here

[serviscope_minor]

Because coding IS hard, and is something you have to be passionate about to be worth a god damn;

No it ain't and no you don't. Being a great coder is hard and to be the best you generally need passion.

You don't need to be the best or great to be useful however. In fact there are plenty of people who have no particular love of coding but see it quite reasonably as a means to an end. For example the coding is required to control this or that piece of kit or to perform the right kind of data analysis etc etc.

Re:Nothing new here

[barbariccow]

That always got me bad grades. I had already been programming professionally for many years before I decided to try for a CS degree. I already knew everything, and often did things in ways the TA's didn't understand, or that was beyond where we were, and would get marked down for it. The ONLY CLASS IN THE WORLD where you can get marked down for over-achieving. In fact, I used to sell my work to girls, I'd put errors in my code and change things around. They'd get better grades.

Re: Nothing new here

[K. S. Kyosuke]

How the hell does anyone program anything without vim?

Using Emacs, obviously. ;)

Re: Nothing new here

[Time_Ngler]

What's a "fucking computer"?

Re: Nothing new here

[Quirkz]

For sure. I've written code in order to solve word puzzles, or to create them. Realizing I could apply programming to a dictionary file was an eye-opening experience for me.

Re:Nothing new here

[networkBoy]

My ex got flagged for plagiarizing from *herself*.
Basically a sophomore class and later graduate class on the same topic assigned the same writing prompt about the same historical site.

While my ex *did* write a new essay, the source material didn't magically change in two years, nor did her writing style or propencity for fscking up affect/effect, then/than, their/there/they're in the same ways, so...

FLAGGED!

Re:Nothing new here

[networkBoy]

Secure? yes*, maintainable? not so much.

*WRT polymorphic viruses.

Re: Nothing new here

[mallyn]

Hold on!!!! You might be correct with some connectors. But you forgot about static sensitivity and power supply voltages.

In the old days, TTL logic was 5 volts and components were larger and more robust.

Now, I understand that some components are either 3 volt or even 1/5 volt logic. You put 5 volt logic into a 3 or 1.5 volt logic device and you have just lost the device. This has happened to me.

Try using the Raspberry PI with some 5 volt components. You just lost the GPIO of the Raspberry if not the entire Raspberry device.

With many devices, you must wear the static electricity handcuff. If you don't, and if you are careless on a carpeted floor, you can kiss the device good bye.

I do agree that devices are less likely to kill you. Lethal voltages (anything over 12 volts) are mostly gone from newer devices.

I hail from the vacuum tube days where you quickly learn to keep one hand in your pocket and do not work in a damp basement with a concrete floor. In fact, I am still practicing those procedures at the Spark Museum while working on their vintage radios and electronics!

Re: Nothing new here

[DuckDodgers]

You're weeding out the people whose secondary education program didn't have adequate computer science offerings. Even in 2017, that describes most secondary schools in the US.

Re: Nothing new here

[DuckDodgers]

How do you know what the standard case is? My oldest kid at eleven was writing better code than I could at 19. He had access to better teaching materials and an experienced mentor. If he ever takes CS101 at a college, he'll sleep through it. That doesn't mean he's got some special 'capability for insight'.

We take for granted all of the concepts you have to handle to be a developer. We do abstract visualization of algorithms - and not just the fancy O(n logn) stuff but even simple things like "open the file, read each line, extract the entries between the third and fourth commas, convert from string to integers, sum the values". We know syntax rules, and won't get tripped up by errors like "if (condition);" (the semi-colon means the if condition has no result). We understand concepts like explicit and implicit type conversions. We understand variable scoping. We understand passing by value and passing by reference. We understand compilers and interpreters. We understand static types and dynamic typing.

And none of those concepts are difficult. Really, they're not, as long as you learn them one at a time, in isolation. If you throw three or four at a novice at once, that person is going to choke even if their raw intelligence is really high. I had a tough time with the learning process and cheated to pass the first few classes. When I was in my 20s, my father tried to make a career switch into software development. His CS101 class had C++ with some libraries and web services. The old guy had a full scholarship to college the first time around and graduated with a degree in math, so he's not stupid. But the only people who passed that class were the ones that were experienced developers that just wanted to put a degree on their resume. Every true novice flunked.

Re:Nothing new here

[XxtraLarGe]

In my classes, the students are expected to complete the assignments using the techniques covered to date.

That is the worst sort of professorship. Once I had an SQL class that required a right join returning exactly three rows in an arbitrarily large dataset, sorted by specific descending index and date range values all before we were taught how to do right joins, indexes, date ranges, and top selection.

I think you may have misread what I wrote; it sounds like the assignment you're describing is the exact opposite of what I do.

Re:Nothing new here

[XxtraLarGe]

That always got me bad grades. I had already been programming professionally for many years before I decided to try for a CS degree. I already knew everything, and often did things in ways the TA's didn't understand, or that was beyond where we were, and would get marked down for it. The ONLY CLASS IN THE WORLD where you can get marked down for over-achieving.

At the start of every semester, I give my students a survey that goes over their computer & programming experience. None of my copy/paste all-stars have ever had the kind of experience you have.

Re:Nothing new here

[SandWyrm]

Yeah, by the time you've seen the third project with the exact same spelling, spacing, and punctuation mistakes, you know you have a problem. In my case six students out of 24 had copy-pasted their code. Cue crying and angry parents saying that it was supposedly OK because they all got together as a study group and manually copied their 100 lines of code verbally (yeah right).

Re: Nothing new here

[SandWyrm]

"This is a very tough mental field. The campaigns to teach everyone to code make it seem like it's house work. Anyone can do it. I think they are finally realizing it's not so."

Yeah, nobody outside the field understands that the language is only part of it. The real work is breaking down a problem into a series of logical steps. A vast majority of the population simply will never be able to do this.

Re: Nothing new here

[peawormsworth]

In high school, I wrote code that was copied by 3 other students. The teacher said he would take the 90%ish grade on it and divide it between all of us. I stood my ground and said it was unacceptable to get dinged because others cheated and the teacher told me that it was my fault for not protecting my own code and didn't really have a way to tell it was mine. But I was the only one that was so adamant about it.

Later in the year, it was discovered that one of the students had written a fake login screen for the computers and was collecting user passwords. He was also selling commercial software and games that he had "cracked" to other students.

I couldn't understand why someone so obviously smart in computing would not just write the code for the assignments. But some people are just drawn to the dark side of computing. I would imagine that he is working for the government now, breaking into yours and my computers and phones right now. That is where the darkest of darkest people get jobs.

Re: Nothing new here

[K. S. Kyosuke]

There are other text editors?!

Re: Nothing new here

[DuckDodgers]

How many biology 101, history 101, or Mandarin 101 professors expect the students to enter the classroom with a decent background in the topic? "101" course levels imply introductory material. Until computer science classes are a standard, non-optional part of high school education in every school in the country, college CS101 students should be expected to know nothing about programming when they get to their first class.

Re: Nothing new here

[Hognoxious]

Fucking twaddle. How many highschools run classes in medicine, law, accounting, engineering...

Re:Nothing new here

[Hognoxious]

Fucking softy. I'd have given the little sadpacker an F.

How many different ways to solve problems?

[grasshoppa]

How many different ways can you solve a college level problem in a course assigned language? If you have 700 students, I guarantee successful assignments are going to look like they copied each other to varying degrees.

Re:How many different ways to solve problems?

[Anonymous Coward]

There may be a limited number of ways to solve the problem, but there are an unlimited number of variable names that can be leveraged to solve a problem.

No two should be identical. Though all will be similar.

Re:How many different ways to solve problems?

[king neckbeard]

Yes, there are unlimited possible variable names, but ideally, variable names make sense. The number of variable names that make sense is much more limited.

Re:How many different ways to solve problems?

[pr0fessor]

There also some commonly used variables for loops with int i and st as stream and they are repeated all over but generic variables are a different problem.

Re:How many different ways to solve problems?

[beelsebob]

The key is that there's nearly unlimited ways to solve a problem incorrectly, but when the exact same incorrect solutions keep coming up within the same sets of students it becomes very obvious to the teaching assistant what's going on.

Re:How many different ways to solve problems?

[grasshoppa]

The key is that there's nearly unlimited ways to solve a problem incorrectly,

Point of fact, there are a very small number of ways to solve simple, college level problems. Obfuscation is not a coding methodology we want to be encouraging, afterall.

You have a valid point about incorrect assignments, however.

Re:How many different ways to solve problems?

[Anonymous Coward]

if something like that happens, you just pull both students in independently and ask them to explain how the code works and ask a few pointed questions. In general the students who copy do it because they don't understand the problem or how to solve it, some won't even understand the language they are supposed to be working on.

Funny story in college I had an instructor end class early because he was so enraged by a cheating student. I think it was the complete lack of effort that did it, the student turned in the 1st assignment from last year for this years first assignment, despite the fact that the problem had changed.

Re:How many different ways to solve problems?

[Anonymous Coward]

I used to run diffs on everyone's code, piped to wc -l to see who had the smallest number of different lines. I'd check those. Usually they were identical down to the whitespace with only variable names changed. Sometimes only comments. Rarely, but it happened, no changes except the filename. Students heard that I automated grading so they assumed I wasn't checking. it was more work to make their code look different than to write it from scratch.

Re:How many different ways to solve problems?

[grasshoppa]

Heh, I'd have been the dick who wrote an automated obfuscator. Flip the indentation from spaces to tabs ( or tabs to spaces ), randomly change ctime/mtime ( within acceptable range ), camelcase to underscore ( or reverse ), use a dictionary to change variables to their synonyms, add generic comments ( ala "palm reading" ), randomly placed returns ( where language appropriate ).

Figure that wouldn't take more than an afternoon to code up.

Could get even crazier by adding the ability to swap out loops ( foreach to while/for ), but those can impact overall grade and is language specific so it'd have to wait for v2.

Re:How many different ways to solve problems?

[twistedcubic]

Some things you only realize when you have to grade students' programs. Clueless programmers cheat in clueless ways, of course, because if they were clever at hiding that they copied, they could just use this skill to write the programs on their own. Also, the anti-cheating programs we use are way more sophisticated than inept programmers. I use the excellent VPL Moodle plugin (https://moodle.org/plugins/mod_vpl) which has this built-in.

Re:How many different ways to solve problems?

[computational super]

an unlimited number of variable names

What? There are only 26 letters. a = b; b = c; and now I've already used up 3 variable names.

Re:How many different ways to solve problems?

[pr0fessor]

I like how you point out it's the assistant that is doing the grading of assignments.

Re:How many different ways to solve problems?

[uncqual]

Back in the day, we all used i , j , and k and none of us got accused of copying because of using the same variable names.

Get off my lawn.

Re:How many different ways to solve problems?

[HornWumpus]

Back then, the type was implicit in the first letter of the name and only two characters counted.

Re: How many different ways to solve problems?

[grasshoppa]

I was a horrible student, but CS courses were where I excelled ( except, notably, in the MS courses. Go figure ). I never pulled the auto-cheater thing outlined above, but I did other things throughout my college life which rivaled it.

I got the distinct impression that my professors were glad to see the end of me in their classes. One actually flat out told me so ( he later hooked me up with a great job ).

Re:How many different ways to solve problems?

[freeze128]

The summary doesn't mention what language the classes are teaching, but I wonder if there really is an unlimited number of variable names. In Microsoft C++, the variable names are limited to 2048 characters.

Re:How many different ways to solve problems?

[freeze128]

If you were a student, and you wrote that obfuscator just so you could copy someone else's code, then I would say that you deserve to pass your programming class.

Re:How many different ways to solve problems?

[Dan East]

Point of fact, there are a very small number of ways to solve simple, college level problems.

From a algorithmic sense yes, from an implementation sense, no. You are to write a function to calculate the nth number in the Fibonacci sequence, using only operators inherent to the language (IE no math libraries). What will you name the function? What will you name your variables? Will you use an iterative method, and thus will you use a for loop, or a do...while loop. Will you count up or count down? Or will you use a recursive function? Or perhaps you do some research and discover Binet's formula which calculates a Fibonacci number without iteration.

I guarantee you if you ask 20 developers to write that function, you will get 20 different implementations that differ quite enough to tell they are not copies of one another.

Re:How many different ways to solve problems?

[kleinesRaedchen]

Heh, I'd have been the dick who wrote an automated obfuscator.

Your obfuscator would be entirely useless against professional plagiarism programs.

Flip the indentation from spaces to tabs ( or tabs to spaces ), randomly change ctime/mtime ( within acceptable range ), camelcase to underscore ( or reverse ), use a dictionary to change variables to their synonyms, add generic comments ( ala "palm reading" ), randomly placed returns ( where language appropriate ).

spaces to tabs: whitespace usually is ignored
ctime/mtime: why should this be considered anyway?
camelcase to underscore: identifier names are ignored
change variable names: identifier names are ignored
add generic comments: comments are ignored
randomly placed returns: whitespace usually is ignored

Professional-grade plagiarism programs see
int a = b
+ 2;
as token sequence
INT INDENT ASSIGN INDENT PLUS 2 SEMICOLON

All of your obfuscations are snake oil.

Figure that wouldn't take more than an afternoon to code up.

Could get even crazier by adding the ability to swap out loops ( foreach to while/for ), but those can impact overall grade and is language specific so it'd have to wait for v2.

Swapping code parts woul indeed challenge plagiarism programs. But, as you conceded, this is hard to implement without introducing errors.

Re:How many different ways to solve problems?

[grasshoppa]

...and now I'm back to wondering how you might hope to catch cheaters when you have a program which starts by abstracting away the minutia when you're dealing with simple college level coding problems.

There are only so many ways to solve the same problem.

Re:How many different ways to solve problems?

[rhsanborn]

I like the anti-cheating programs, but I never saw that they stopped much cheating, it just helped me identify the students who did (a lot). I switched my classes to have challenging, hand-written programming. Simple mistakes weren't punished (missed semi-colons) but it forced the students to do the work without the aid of the internet or their peers in a controlled environment. The students who cheated all semester failed out on the tests. The students who completed the homework usually breezed through the tests.

Re:How many different ways to solve problems?

[beelsebob]

Well, I used to be said TA, so... yeh :P. It was really fucking obvious when people plagiarised. Not always easy to prove it, but very easy to tell.

Re:How many different ways to solve problems?

[kleinesRaedchen]

...and now I'm back to wondering how you might hope to catch cheaters when you have a program which starts by abstracting away the minutia when you're dealing with simple college level coding problems.

The "minutia" are worthless because they can easily be altered for fraud. However, it is not easy to change the overall structure of a program. The effort is similar to write the program by oneself from the start. We use professional-grade plagiarism software (throwing away "minutia") and it works very robust. The most important parameters are maximum length of identical token sequences and percentage of identical n-gram token sequences. The main difficulty is to find a balanced parametrization. A low levels of false positives leads to a high level of false negatives and vice versa. Following "in dubio pro reo", a low FP-level is prefered.

There are only so many ways to solve the same problem.

Ca. 20 code lines with freely to be written logic are required. You would be astonished of how many ways there are to solve even small programming problems.

Re:How many different ways to solve problems?

[dgatwood]

What will you name the function? What will you name your variables?

Proper cheating detection has to ignore function and variable names, because changing them is the most obvious way to cheat.

Will you use an iterative method, and thus will you use a for loop, or a do...while loop. Will you count up or count down? Or will you use a recursive function? Or perhaps you do some research and discover Binet's formula which calculates a Fibonacci number without iteration.

In college programming classes, the Fibonacci sequence is invariably being taught as an example of how to write a recursive function, which eliminates all the iterative implementations right off the bat. It also eliminates Binet's formula (and also, because you said no math libraries, computing a square root isn't really possible, at least in C, making that solution unavailable anyway).

In an introductory programming class, catching cheaters is useless, because their code should always be pretty much the same; if it isn't, they're probably doing it wrong and will fail at the end of the term anyway. And if they're just copying and aren't learning from what they're copying in the process, they'll fail at the end of the term anyway. This is why it is so important to require students to actually write code in class for the final exam. You immediately know with certainty who learned the material and who didn't.

I guarantee you if you ask 20 first-year CS students to write that function, given the usual criteria for a first-year college test question, even without cheating, the differences will be whitespace, variable names, function names, comments, and function order (whether function A is before function B or vice versa), none of which tell you anything about whether the student understands how to write code. And that's one problem with CS; by the time you get to the point where students either truly get it or don't, they've spent a lot of time in classes—at least a semester/quarter, and arguably more.

What we really need to do is come up with some sort of aptitude test for computer science, in which we somehow magically determine whether a person is going to be able to wrap his or her head around writing software. I don't have any idea what that might be yet, though, or else I'd have written it already, and I'd be rich, so there you go.

You say cheating, I say cargo cult programming

[Opportunist]

It's not like that's something new. Nor is it new that those that do it are not really the ones that will become the 7-digit-earners at Google or found million dollar startups.

All we get is more code monkey squeezing out insecure code. Or in the terms used by IT security consultants, job security.

Re:You say cheating, I say cargo cult programming

[gweihir]

Ah yes, for example the Java "programmers" that can only call existing functions and methods and can only do it with code copied from the web. The job-security aspect for IT security consulting is correct though, I am benefiting from that. Although when I have to explain to a web application programmer how to find out what his code does that is really demented. It is much more satisfying to do this in an academic setting, there I can just fail those that do not understand what they are doing.

Re:You say cheating, I say cargo cult programming

[Opportunist]

In a consulting environment I can make sure that his boss notices that the dud should be fired. Out of a cannon, preferably.

Just kick them out

[gweihir]

Just give them two warnings and on the third time caught kick them out. And of course even on the first time caught, fail them for that course.

Oh wait, this is education for money, i.e. that form were even the most stupid cheater has to make it in order to keep the money flowing. Well, why not make all courses optional and just sell that degree directly?

So many students...

[Infiniti2000]

...of his roughly 700 students in one class...

Can you effectively teach a class with that many students? That's a ridiculously high number.

Re:So many students...

[HornWumpus]

Pretty standard. Large lecture (likely two times), small labs. TAs do 99% of the teaching.

Re:So many students...

[phantomfive]

At Berkeley they have a professor who lectures a huge class, then a bunch of graduate students who work more closely with the individual students. So you rarely talk to the professor, but you frequently talk to your graduate student advisor.

I can't speak to the effectiveness of this system, but the purpose of course is to save money.

Re:So many students...

[frank_adrian314159]

You can teach that many if you divide the class into multiple sections with a main lecture and the sections led by junior faculty or teaching assistants. Lower-level classes have been handled this way forever in large universities. Or you just teach it online with multiple instructors covering various subsets of the students. You don't think educational methods scale, too?

Re:So many students...

[gweihir]

Welcome to industrialized "education" for money. Where I teach (on the side), they have an absolute class size limit of 30 for all classes, enforced by there simply not being more space in the classrooms. With that, you get actual interaction and that is incredible valuable. I think that somewhere beyond 150 participants or so you lose basically all of the advantages of direct teaching. But the university makes more money and that seems to be key here.

Re: So many students...

[phantomfive]

Yeah it's a problem. You are likely to be better off going to UC Merced and getting more personal attention at a lower cost.

On the other hand, I've known some good programmers to come out of Berkeley, so maybe they're still getting the basics right. If you're learning a linked list, a graduate school advisor can probably teach it almost as well as Donald Knuth.

Re: So many students...

[parkinglot777]

If my class was 700 students I'd walk out and complain. The most students I've had in one class was 100 and that was for Econ 101.

I highly doubt that 700 students meant to be in a class room at one time. I believe they are in different time schedules; thus, they are in different classes but all are in the same course. To me, the way TFA (and summary) is written is to exaggerate what is going on, so it looks more interesting to read...

Re: So many students...

[steveg]

s/free labor/cheap labor/

This doesn't really change your point much, but TAs *are* paid.

Re: So many students...

[phantomfive]

If you go to a 'lesser' university, it's a good idea to get a feel for what a graduate in your major is supposed to learn. That way you can supplement your education by reading books from the library or doing personal projects, in case one of your professors turns out to be a dud. Then you truly can get a world-class education at a lesser university.

Re:So many students...

[Obfuscant]

Welcome to industrialized "education" for money.

To be fair, there are some courses that are core requirements for a large number of degrees, like Math 101, Chem 101, Physics 101, etc. There are so many people that need to take those courses every year that it only makes sense to have "classes" of several hundred people. I think the lecture hall used by Chem 101 held 300 people when I was an undergrad, and it was used for several sections per term.

Trying to have enough staff for such large classes would easily drive tuition up to where it was unaffordable. It was much cheaper to have one professor and a dozen TAs instead of a dozen professors.

CS101 certainly fits that mold.

Re: So many students...

[Areyoukiddingme]

I heard that in one of his last semesters of teaching Donald Knuth was teaching an introductory programming class. Which would have been cool for the students, but is still a horrific misuse of the man's abilities.

You do realize he chose to teach that class, right? Nobody told Donald Knuth he had to teach an intro course against his will. College administrators are greedy and dumb, but they're not completely blind stupid (yet). I'm sure he wanted to teach the class in order to get in touch with the novice perspective again. Lots of smart people who do something revolutionary for the field forget what it was like not to know all the things they know. Donald Knuth is smart enough to know about that problem, and do something about it. You don't write textbooks for expert programmers (most of the time). You write them for programmers who are learning. As a legendary textbook author, Mr. Knuth needed to keep his finger on the pulse of the novice, and teaching an intro course is a good way to do that.

Paired

[Archangel Michael]

Paired Nicely with the article just below it. Excellent placement of plagiarized example articles.

A+

CHEATERS!

[Lumpy]

All of them steal my code!

function main()
{

}

and even using IF statements and DO WHILE! The freaking copycats!

If an instructor expects 30 students to do a bubble sort 30 different ways, he needs to be fired as a CS instructor.

Re:CHEATERS!

[Megane]

I'll show them!

int main(int argc, char * const argv[])
{
printf("Hello world!\n");
return 0;
}

(WTF? &nbsp; doesn't work anymore? Even in <code> mode?)

Re:CHEATERS!

[lgw]

int main(int argc, char * const argv[])
{
    printf("Hello world!\n");
    return 0;
}

Assuming preview isn't lying to me, you want tt, not code, and just use two spaces for indent.

Practice vs proving knowledge

[davidwr]

The purpose of labs and other homework is to practice. If you cheat, it's a waste of time.

Proving your skills to the instructor needs to be done in a controlled environment. "Controlled" doesn't always mean audited: In a perfect world a student's sense of ethics would be enough self-control that take-home tests or graded projects would be allowed as nobody would collaborate or use others' work beyond the bounds imposed by the instructor.

If ethics is a problem, either do a better job of teaching ethics, make your program and admissions process such that unethical people won't want to apply, or arrange the material so all work that makes a significant contribution to the final grade is audited in a very-hard-to-defeat manner.

Re:Isn't it all cargo cult programming?

[zlives]

the issue is not of creativity but of understanding. same applies to math.

Name and Shame!

[h4ck7h3p14n37]

Any student caught cheating should have their name announced/posted in a prominent location so all of their classmates know who the cheaters are.

The rest of the students work too hard to allow cheaters to remain anonymous. They deserve to know who's trying to screw them over.

Re:Name and Shame!

[JeffTL]

It sounds like an effective idea in theory - until someone sues the instructor and institution for violation of student records privacy (in the United States, FERPA is the relevant federal legislation), depending on the jurisdiction and details, potentially libel.

The x and y of cheating...

[__aaclcg7560]

When I was learning to program after the dot com bust, we had a class assignment to pair up and work together. But the code had to be written individually. A pair of students submitted identical code except one used the x variable and the other used the y variable. That got a good laugh out of the class when the instructor mentioned. The students got a slap on the wrist for not submitting their own work.

I had an instructor "cheat" decades ago

[macraig]

Back in the early Eighties, I took a programming course from a particular instructor who I later learned from a friend had "cheated" and used some of my work from the class as an example in later classes. Humor aside, cheating is hardly a new thing. Neither is programming.

Re:I had an instructor "cheat" decades ago

[i.r.id10t]

Why is why I submit my homework with a copyright header and GPLv2 it. Written works like term papers are submitted and released under a CC license with no commercial use, explicitly being kept by TurnItIn.

Honestly, I don't think it will do any good, but it makes me feel just a little tiny bit better.

Re:I had an instructor "cheat" decades ago

[macraig]

That's great, but the GPL wasn't even a gleam in anyone's eye at the time of my anecdote. Copyright certainly existed, but trying to exercise it in that fashion back then would have been nothing but trouble.

Comment removed

[account_deleted]

Comment removed based on user account deletion

asian culture is geared so that test / school chea

[Joe_Dragon]

asian culture is geared so that test / school cheating is very common and the schools really don't want to kick out full cost paying (some time more then out of state rates) foreign students.

Re:Isn't it all cargo cult programming?

[davidwr]

Show me someone doing something creative today in code.

Those obfuscated- and underhanded-programmer ng contest entries come to mind.

Big Deal

[diesalesmandie]

So what if they cheat; if they ever apply for a coding job they'll get caught out in the technical interview, moving on...

Re:Big Deal

[sizzzzlerz]

And even should they get through the interviews and receive an offer, their lack of ability will become quickly known with their first assignment. Unfortunately, you will always have other companies down the food chain who'll hire the dregs because the good people won't work there. It's my theory that this is how companies who make IoT devices get their programmers and why those devices are the shits.

Re:Big Deal

[PPH]

And after the third or fourth applicant from that college is found to be an incompetent fuckwit, HR will stop looking at any applications from that institution. Pretty soon, word will get out and a degree from there won't be worth squat. And nobody will enroll in their CS program anymore.

That's why colleges worry about stuff like this.

Re:Big Deal

[ausekilis]

While not cheating per se, my old employer stopped looking at my alma mater purely because they stopped teaching C/C++ and moved to Java. Makes sense too, since one of my professors wasn't happy with the change either. He calls Java a "training wheels language".

Re:Big Deal

[diesalesmandie]

And after the third or fourth applicant from that college is found to be an incompetent fuckwit, HR will stop looking at any applications from that institution. Pretty soon, word will get out and a degree from there won't be worth squat. And nobody will enroll in their CS program anymore.

That's why colleges worry about stuff like this.

Good point, didn't think of that

Some things should be more team based as some of t

[Joe_Dragon]

Some things should be more team based as some of this goes back years of the ivory tower ways of being be hide the times and with professes who at times have little idea of the real world uses of what they are teaching.

Example code

[fahrbot-bot]

From TFA, two students submitted work with the following lines of code.

boolean done = true;
while (!done) {
...
}

In a class of about 450, they were the only ones who made that fatal mistake. “This is pretty strong evidence that one had copied the other,” Mr. Dunsmore said. “They later both confessed to collusion.”

Re:Example code

[davidwr]

boolean done = true;
while (!done) { ...
}

In a class of about 450, they were the only ones who made that fatal mistake.

That's no mistake, that's a test of the compiler's optimizer.

Re:Example code

[parkinglot777]

If you really read a bit further down, you would see that the professor you are talking about is from another college. The one which is talking about 700 students is just a paragraph or two down...

Um, Ah....

[SuperKendall]

Students have been caught borrowing computer code from their friends or cribbing it from the internet.

I wonder if the article writer is familiar with what professional coders do, or GitHub in general???

Re:Um, Ah....

[geniepoo]

Penalize inferior wheel re-invention as industry does. Perhaps the thedailywtf.com could tell us how many of their posts come from original code snippets.

Re: Um, Ah....

[SuperKendall]

Just because it's on Github doesn't mean you can always just make a copy of it.

That is true, but the proof of the copy quality lies in the result.

That is to say, if you are copying the wrong things you'll get a bad result, or the program will not work, leading to a bad grade or failing.

If you are copying code well, that would seem to be a POSITIVE feature for future productivity as a professional.

Now if you are copying wholesale, that may be an issue, but even there how is that different from evaluating frameworks to solve a problem?

I know students need to learn the basics as well, but it seems like coding could be tailored to solving unique problems you could not copy code wholesale for (i.e. change projects between semesters), and also expect and support ethical code copying/reuse. That is much more realistic.

Re: Um, Ah....

[PPH]

but it seems like coding could be tailored to solving unique problems

The point is to learn how to do the work. On your own. The point is also to teach students the same curriculum and not demand that instructors create novel problems for each class.

First grade arithmetic: "Hey Miss Smith! Why should I have to get the answer for 1 + 1? Last years class already did that work."

Re: Um, Ah....

[SuperKendall]

The point is to learn how to do the work. On your own.

Doing the work, on your own, *in real life*, often means making use of other people's code.

That may not even be GitHub; it could be simply co-workers code.

The point is also to teach students the same curriculum and not demand that instructors create novel problems for each class.

That sounds nice for the professors but obviously does not scale (something CS grads should know about right??) as every class after the first one will have wholesale examples easily replicated, I thought you said the point was for them to do the work and not just copy projects? Yet you are setting up an environment where copying is guaranteed.

If you want a specific result you have to do something where the likelihood is the result you seek (i.e. students learning to program), not abject failure (students realizing they can save time by copying previous solutions because the professor was equally lazy).

Hey Miss Smith! Why should I have to get the answer for 1 + 1?

Here we see the core of the problem; you lack understanding of what education does that is helpful to a student. It's not to tech you facts a calculator can regurgitate. It's to teach you how to approach a problem. Similarly learning how to code means learning concepts and the language.

Re: Um, Ah....

[Hognoxious]

If you are copying code well, that would seem to be a POSITIVE feature for future productivity as a professional.

Not on a college course.

It's more like taking a driving test by hiring a chauffeur while you sit in the back.

Re: Um, Ah....

[SuperKendall]

Not on a college course.

Why not?

Perhaps a good starting point for coding would be to learn how to evaluate code, to learn how to use other code to put together to build something useful.

You always end up needing to use glue code, why is that not a good place to start? It's also more exciting as you get something more impressive together sooner than if you are truly starting from scratch.

It's more like taking a driving test by hiring a chauffeur while you sit in the back.

No, the correct metaphor is, that blindly copying code to solve an assignment is like watching Talladega Nights and then driving a car for the first time. It's going to go badly if you don't ACTUALLY know what you are doing.

Re: Um, Ah....

[PPH]

Doing the work, on your own, *in real life*

This isn't real life. It's school. You are expected to learn and demonstrate skills that a previous class has already completed.

It's not to tech you facts a calculator can regurgitate.

That's why calculators are not allowed in basic arithmetic classes.

I'm not sure what the point would be

[Skuld-Chan]

I mean at some point in your career you're going to have to figure out some problem on your own - and if you cheat during all that high priced training you might as well not even have gone.

Re:I'm not sure what the point would be

[Kielistic]

To a certain extent maybe. Keep in mind that a lot of these "intro to programming" tasks are things that no one would bat an eye at lifting whole-sale from wikipedia or stackoverflow in the professional world. I don't mean in a "I don't understand how to write this" type of way. I mean in a "I know this is a solved problem and it is quicker to google it than to write it myself with all these people screaming in the background".

Though I agree that lot of the people caught cheating in these things will never be good programmers I do not believe that everybody caught cheating is a bad programmer.

"Open Source" is human nature

[TheOuterLinux]

Whether it's fire or code, our survival has always depended upon sharing, or "cheating" if you're part of the proprietary problem. We copy code without completely understanding the "how" all the time, it's called a "library" and as technology advances, what is an advance script today may just be the next "library" tomorrow.

Re:"Open Source" is human nature

[Obfuscant]

We copy code without completely understanding the "how" all the time, it's called a "library"

And this is often why we wind up with complete nonsense output from a simple program. A large part of reusing code requires knowing when the assumptions that went into writing that code are valid and when they are not. It is hard to know when you shouldn't use a certain kind of sort algorithm if you don't know what that algorithm is and how it works. It is hard to know when not to use certain numerical functions if you don't know what they are.

That's why you learn what things are and how they work when you are in school, so that when you get into the real world you can make better choices and not so many mistakes.

Re:"Open Source" is human nature

[TheOuterLinux]

You may learn how to make your own library in school, but very few actually to go beyond the README to see how it works and most teachers don't care if they do or not. Programming is a lot like design; the ends do justify the means and no employer cares how you get there as long as he/she gets the most bang for their buck, degree or not. If I "draw" better than you, whether or not I'm using the tools correctly, the poor guy in college dept doesn't have a chance. Learning to program is too easy these days to be bothered with the fundamentals, which is why, unfortunately, most go for cloud computing because if something doesn't work, its not their fault since they are only using an API like everyone else. So, we either let them collaborate to encourage future FOSS desktop developers or we risk having a bunch of "interface designer" API slaves because either way, the current generation is going to half-ass things. I'd rather have 100 idiots on a unique project than another damn webkit/electron app on my phone.

India anyone?

[yodleboy]

If these schools and profs want to get an early look at the kinds of problems they can expect when a massive swell in IT courses happens, look no further than India. What problems do they have? How do they address them (or fail to)? Seems there are regular articles on mass cheating in Indian IT curriculum. For the sake of fairness, they could research the MSCE schools in the early 2000's. I guess my point is that as surprised as they seem to be, there is still time to get ahead of the most obvious problems if they do some research.

Re:India anyone?

[Trailer Trash]

If these schools and profs want to get an early look at the kinds of problems they can expect when a massive swell in IT courses happens, look no further than India. What problems do they have?

My experience would suggest that their problems are lack of money and the inability to write computer code.

Well why not?

[computational super]

borrowing computer code from their friends or cribbing it from the internet

Shit, as long as I've been working, I get yelled at for trying to write my own code instead of "just googl(ing) it!" or "just get Ramesh to explain it to you!" These guys are going to be more prepared for the actual workforce than dorks like me who did 6 years of computer science working through everything the way you're supposed to.

Re:Well why not?

[Anonymous Coward]

Don't underestimate Ramesh. Unbeknownst to you, might even have a PhD from some unheard Indian U. And is actually stealthing as a moron while he builds up his billion-dollar startup and buys a permanent US residency by investing a half mil and hires you and your pals.

Re:It doesnt matter

[HornWumpus]

IIT is an exception. Some of their grads are competent, but not all. Much better than the average Indian diploma mill.

I've never interviewed a moron from CalTech or MIT, but aside from those two schools they _all_ let some 'bad ones' through. I bet they do too, just very few.

What should be expected?

[Marc_Hawke]

The majority of posts here talk about how 'collaboration' and 'code recycling' are the absolute standards in professional programming. I agree of course. My question is how should the class be structured in order to allow for that, but still teach the class and measure the students grasp of it?

Does it come down to writing code by hand on paper? (Something I've never liked.)

Does the teacher have to 'warp' the assignments? "Do this project, but get the answer wrong in exactly this way that I've just randomly selected."

Do we ignore actual code as insignificant and just have short-answer or multiple choice questions on concepts? (I know I go through 10 languages a day, I'm competent because I know concepts rather than syntax.)

Or do we say, "There's no cheating in this class. If you're able to get the project done here...you'll be able to get it done in a work environment as well."

Stopping cheating on programming tests

[maiden_taiwan]

I taught programming at a well-known university in the 1990s. To prevent cheating on exams, I created three different versions of the exam. Call them A, B, and C. They had the same questions, but with different numeric values (and therefore different answers). I distributed the exams in the order A, B, C, A, B, C, .... So no matter where a student was sitting, the other exams around him/her were different. I did not reveal this to the students.

Everyone who cheated from his/her neighbor got caught, because their exam (say, "A") would have exam "B" or "C" answers on it. Those students instantly failed the course.

For homework, my advice was: you can talk about assignments in general terms, but you cannot show each other your code, because you are being graded on your work. That was where I drew the line. Still, a half dozen students (out of 150) would get caught cheating on their homework each semester. It made me sad, because none of the cheaters had ever come to my office hours for help. If only they had....

Re:Stopping cheating on programming tests

[hamster_nz]

BEST COMMENT EVER

That solves that problem.

Re:always been that way

[computational super]

buy my coursework

Wait, for how much? I may have found a new avenue for employment...

Self-defeating

[uncoveror]

Maybe you can cheat your way through a coding class, but one you are hired, then outed as a fraud, you will be fired. You can't hold a job with skills you do not possess.

Re:Self-defeating

[hackel]

This is not true at all, in my experience. Particularly in the proprietary/Microsoft world. Unless you're in a large enough organization to have other people who actually know enough to check your work, people get away with this for many, many years. Of course they do learn as they're copying things from Stackoverflow, which of course isn't a bad thing, but they never seem to really "get" it. And small companies are never any wiser. As a contractor, I've had to clean up so many shitstorms that resulted from this exact situation.

Same teacher same standards

[Dareth]

You can usually group college code submissions into groups based on who took which instructor for their introductory programming language course. They had the same lessons from the same teacher under the same coding standards.

Or they get promoted

[Dareth]

Or they get promoted and get to tell the other programmers, "how it is done".

Re:Isn't it all cargo cult programming?

[Opportunist]

Sorry. No.

Yes, more and more high level programming languages abstract away more and more complexity by offering standard libraries. That doesn't mean that you needn't understand what's going on under the hood, though, if you want to write secure code.

The question "does it work" can easily be determined even by a novice programmer. Run the program, see if the output matches the expected output, if yes, it works. The question "is it secure" is more tricky, for if you don't know what would make it insecure, if you do not understand the implication of, say, creating a static character array on the stack and fill it with variable user input, you cannot even determine whether it is safe. Actually, you cannot even define a test case that could determine whether it is.

Fake it till you make it

[rsilvergun]

a lot of kids are trying to work full time while going to school. Except for a few freaks of nature who don't need sleep and the occasional genius that doesn't work. Most end up dropping out. A few fake it till they make it to a real job and aren't trying to do 40 hours/week of course work + 40/week waiting tables because they don't have enough money for food, shelter _and_ tuition...

How dumb can students be?

[bradley13]

I see this among my students as well.

First, on the side of the students: It is perfectly fine to copy code snippets. How do I safely hash a password? Unless it's a computer security course, students shouldn't be reinventing code like that. That's when you go to StackOverflow and find the canned answer from an expert. Some students (and professors) are confused about this.

Ok, with that out of the way: When plagiarism does happen, it is generally pretty blatant. Two solutions submitted, identical except for the renamed variables. It's almost insulting, that they think I won't notice. Alternatively, they pay someone else to write the program, and then cannot answer even the simplest questions about how it works.

But even if they manage to sneak a plagiarized solution through: how stupid can you be?!?! If students aren't writing the programs themselves, they will fail the exam, where copying isn't an option any more. Or, even worse, they manage to scrape through the first year exams. If they get into their sophomore year, they are allowed to fail a course and repeat it a second time. This is horrible, because they drag out the pain for 3 or 4 or 5 years before failing out of the program.

What a waste of their lives. If they can't handle the material, they're only doing themselves damage by dragging things out. Plagiarism in a technical field, where ultimately you either have the skills or not - and this will be discovered - is just unbelievably dumb.

Re:How dumb can students be?

[hackel]

As a programmer who studied physics and has never taken a computer science course, how do you handle exams? Do you actually expect people to hand-write code? Pseudocode? Or do you just let them use a computer and hope that they don't cheat?

I've never understood the concept of cheating on a course you are PAYING for. It just boggles my mind. No one is making you be there!

Re:How dumb can students be?

[tommeke100]

This was in the 90s and mostly in Scheme (lisp) but yes, we had to hand-write the code and explain it in writing, analyze the performance in Big O notation and such.
I think other exams in C/C++ were also hand-written
Exam questions were usually in the form of combing a few abstract data-structures. Or implementing some kind of derivate of an abstract data structure like a special type of hashtable, or some queue with some other special ability.
In Scheme you could usually implement this in maybe 10-20 lines, so was do-able to write it down.

Re:How dumb can students be?

[PPH]

That's when you go to StackOverflow and find the canned answer from an expert.

Where do these experts come from?

Re:How dumb can students be?

[Ken_g6]

I've never understood the concept of cheating on a course you are PAYING for. It just boggles my mind. No one is making you be there!

Probably most of the students aren't paying for their classes. Either their parents are paying or student loans are "paying". And if you don't think people can run up debt without thinking about it, you haven't been paying attention.

Of course, the kids will wind up "paying" for cheating sooner or later. But if it's not sooner, they probably think about it just like they think about the student loans.

Re:Not Cheating!

[computational super]

encourage people to use their brains rather than googling

Heck, I'd settle for just not actively discouraging people from using their brains and doing some actual research before just "asking the guy who knows this stuff".

Then they'll be ready for a programming job!

[ne1av1cr]

My hardest class was a class in Translators where we were told at the beginning of the semester to copy and collude and do any and all things we could and if we made it to the end of the class we'd get a 100%.

Randy Katz?

[Hognoxious]

Randy Katz? Throw a bucket of water on them!

The wrong message!

[hackel]

Students shouldn't be called out for "cheating." Instead, they should be taught *how* to collaborate and properly give attribution to the source of their code and adhering to licensing requirements. This is how it works in the real world. It's not "cheating." Programming is supposed to be collaborative! Open Source is the way, and Free Software should be the mantra of every academic computer science programme.

Of course cheating on an actual test is terrible. But for an assignment? As long as the code runs, and the parts taken from others are properly attributed, it should be permitted. Just as long as it's not a 100% copy of the entire code base, but rather copying an algorithm here or there, using an existing library, whatever.

Of course the vast majority of developers should NOT be studying computer science. They should be doing some kind of software engineering course that is more practical. Computer Science should remain just that—a science. Mostly theoretical, based on research.

Re:The wrong message!

[serviscope_minor]

They should be taught both, but that means cheating will still be a problem. Collaborating etc is all well and good but students also need to know how to actually write code themselves. Not everything can be assembled from libraries and stackoverflow code with trivial glue logic.

Re:The wrong message!

[hyades1]

Collaboration is great, and everything you've said about it is true. But what happens among students often isn't really collaboration.

My work takes me into a lot of classrooms, and I've noticed something you might want to consider. Often "collaboration" among students involves one kid doing all the work and a bunch of lazy and/or stupid sponges hanging around him and just copying. Then they all share in the group's mark.

This doesn't help the sponges learn the subject. It doesn't help the smart kid, either, because all he's learning is that his way is the only way that matters. And as you point out, genuine collaboration is an excellent way to make a project really successful.

There are limits to originality

[hyades1]

When it comes to coding, there's more than one way to skin a cat...but though that number may be large, it is not even close to infinite.

MOD PARENT UP

[damn_registrars]

... I see what you did, there.

Re:Some things should be more team based as some o

[Ken_g6]

The problem being that a team can appear as smart as its smartest member.

I suppose pair-programming exercises could be good, though, if the pairs are randomized each time. Any bigger than that and it becomes hard to figure out which member of a team is earning the grade.

Re:I took some coding.

[fisted]

The fuck does Trig have to do with Boolean algebra?

Lower project score to exam score

[tgibson]

In addition to implementing several ideas posted above, I have a policy in my syllabus that if an exam score is <=30%, I reserve the right to lower all project scores to the exam score. A student must essentially be learning nothing to score that low on my exams (many blank answers, others answered with random snippets of code, etc.). This catches those who turn in projects that are stellar, which an impossible feat given their exam performance.

Exceptions are made for those whose project scores are already poor, and those who are regular visitors to office hours--evidence of their earnest effort.

It Didn't Matter

[n329619]

The AI grading machine will still give you a D- regardless of your effort unless you've managed to exploit a zero-day on the program to change the grade.

Edit

[poofmeisterp]

...about 100 of his roughly 700 students in one class had violated the course policy on collaborating or copying code....

Edited: ...about 100 of this roughly 700 students in one class had been discovered to have violated the course policy on collaborating or copying code....

Cheating in Programming is a bad idea

[foxalopex]

I know cheating has always been an issue in schools but I can't imagine students who do are doing themselves a favour when they eventually reach the workplace. A lot of programming jobs are screened by requiring you to write code on paper at the Interview without any aids. So if you don't know what you're doing because all you've done is cheat, I'm not sure how you expect to be hired. And as great as stackoverflow is for looking up solutions, sometimes they're wrong, written incorrectly or you're trying to figure out how someone else's code works. Good luck doing your job then.

It's not cheating...

[thomn8r]

..it's inheritance.