jump to navigation

It’s about the work, dummy June 18, 2009

Posted by gordonwatts in university.

I read this in an article on ars technica. First the setting:

When San Jose State University student Kyle Brady published the source code of his completed homework assignments after finishing a computer science class, his professor vigorously objected. The professor insisted that publication of the source code constituted a violation of the school’s academic integrity policy because it would enable future students to cheat. Brady stood his ground as the confrontation escalated to the school’s judicial affairs office, which sided with the student and affirmed that professors at the university cannot prohibit students from posting source code.

And second, the thing that made me decide to write this post:

Cory Doctorow shared his thoughts about the issue on Thursday in a blog post on BoingBoing. Doctorow suggests that assignments are ultimately more valuable to the students when the work that they produce can have broader purpose than merely fulfilling academic requirements. He also rightly points out that peer review of source code and studying existing implementations are both common practices in the real world of professional software development.

These are both compelling points and they illustrate how traditional academic sensibilities can be detrimental to the intellectual development of students.

Give me a break. This has nothing to do with any high end ideals. It has to do with work. In lower division courses there are only so many types of homework problems you can write without making something really complex, and in upper division courses creating a good problem that is hard, solvable, and interesting takes an immense amount of time. The professor of the course just wants to be able to re-use the homework problems – and cut/pasting the answer from the web is something he/she wants to make as hard as possible.

I wrote a bunch of problems for my graduate course this last year – they took a lot of work – I spent hours on them. I’d very much like to be able to reuse them – or reuse the core of the problem. If the solutions were widely available then that means I have that much more work I need to do next year.

I think, on its own, the answer to the question about a student posting their source code is clear: they should be allowed to do it. But the issue isn’t black and white when you get right down to it – that solution is a product of both the students and the professor’s sweat. Finally, actually saying that you can’t post code is totally unenforceable in this day and age (e.g. RIAA). There is also the basic fact that a student that decides to cheat is only cheating themselves… boy, that sounds kind-a lame, doesn’t it? 🙂

I don’t have an opinion in this particular instance. But I think the overly simplistic view that is taken is a bit sensationalist. Cory Doctorow wrote a much more nuanced bit:

But the convenience of profs must be secondary to the pedagogical value of the university experience — especially now, with universities ratcheting up their tuition fees and trying to justify an education that can put students into debt for the majority of their working lives. Students work harder when the work is meaningful, when it has value other than as a yardstick for measuring their comprehension.

I disagree. It isn’t just about convenience of the profs – it is about having a good course for the next student. It is about the professor learning what worked to teach the students this time and refining it next time – these things have to be factored in. Both the students and the professors, it seems to me, have a shared responsibility here.

Though a bit later on he seems to go off the rails (no pun intended):

I’ve always thought it was miserable that we take the supposed best and brightest in society, charge them up to $60,000 a year in fees, then put them to work for four years on producing busywork that no one — not them, not their profs, not other scholars — actually wants to read.

Well, gee. When you start learning something you have to start with the basics. You can’t start with Quantum physics – you need to understand a bit about mechanics, E&M, and other things – after all, quantum had better devolve to those in the macroscopic world! You can start out students on cutting edge research – we all do it when we take on undergraduate researchers – but they have to learn the basics too. It is, sadly, a fact of life. 😦 If you don’t know how to start a program, how can you learn how to write cool code!?!?


BTW, if you read down, you can find a response from the student involved in this case.



1. Mefisto - June 18, 2009

I think the problem here is not the fact that the professor would or would not reuse the assignments, he/she could reuse them without problem, the key is having a way to evaluate what the student did and learned, it’s just like reading a new contribution to science, you never thought about it but you could learn it and thats what you need.

2. gordonwatts - June 18, 2009

I was with you right up until the last phrase – “it’s ust like reading…” – I didn’t get that bit. But yes, definately – HW serves two purposes – one for the student to learn about the course work and two so the prof can get some feedback on how the student is doing. Tests suck – if you make a mistake your grade gets nicked. HW you can work on, discover you made a mistake, fix it, etc. – which is the way to learn. So, usually there is no time to learn on Tests, but there is time to learn on HW. 🙂

3. Michael Schneider - June 18, 2009

I don’t see how this is a problem. For example, the solutions to Jackson E&M problems have been on the web for years but my E&M class still had a range of grades on the homework and exams. Honest students will do their best to solve the problems on their own to really learn the material. Being able to check solutions on the web at the end of the process should be saving work for the Professor. Dishonest students will simply copy the solutions, but this is often obvious and they will fail the exams.

If it is a problem to have homework solutions available, doesn’t that mean the evaluation metrics are not measuring what students have actually learned?

4. gordonwatts - June 18, 2009

I don’t give exams in my graduate courses. I do in the lower division ones in part for this reason. But not all profs agree with this approach to teaching. From their point of view this may feel like a real limitation of their options – and, generically, mean more work for them.

5. jeremy - June 18, 2009

I don’t think it’s such a big deal–as you said, the students who don’t learn it are only hurting themselves. I think the students who are interested will learn just as well regardless of if there are solutions and regardless of if they look at solutions beforehand.

Good students will think about the problems they’re solving, try to find clever ways to do it, think about what their answers mean, etc. For them, I don’t think that having solutions available online is any different than having other students available to talk to about their solutions.

The students who aren’t interested, if they can’t google solutions, they’ll find someone else to work with and use theirs, or write something confusing and hope the TA will give them credit (and they will, because TAs usually don’t have as much time as they should to grade anyway!)

So I don’t think it matters, in the big picture. The good students will learn in the end, the bad students won’t.

In terms of feedback, I think talking to the students who ask questions in office hours, or talking before/after class with students can be a more realistic gauge of where problems are than HW scores. And often talking can show deeper misconceptions that solving a specific problem won’t show. Problems tend to focus on mechanics as opposed to the big picture. Mechanics are useful, but are only a part of the story.

6. Aleksie - June 18, 2009

I’ve been in classes with students who do resort to internet searches for answers. When it comes to exam time, they really do hurt themselves.

7. gordonwatts - June 19, 2009

Jeremy – I wish more students would come by to office hours! But that might just be me! 🙂 And what you say is true, Aleksie – though some advanced classes don’t have exams. But by then people would be expected to be self motivated. Well, I would hope! The only danger is that the University should not be giving degrees to people that haven’t learned the material.

8. Gordon Stangler - June 20, 2009

Dr. Watts,

All my classes have been the opposite. I didn’t have homework (or if we did, it was never graded by the prof), but we did have a final exam; or a midterm plus final exam.

9. gordonwatts - June 20, 2009

Gordon – what kind of classes were these?

10. gordonwatts - June 20, 2009

Gordon – the ads on your site make it veyr hard to look at your blog!

11. Can students post their homework on internet before due date? « Paviavio           - June 27, 2009

[…] write about this incident. Google search, however, still led me to few of them. Here is one of few: It’s about the work, dummy. From Lightspeed, I found the student’s blog posting CS146 Code Issues that indicated he […]

12. anon - June 29, 2009

I don’t see how seeing the solution to a problem stops learning. When I was an undergraduate working on problem sheets if I got totally stuck, then I went to the library and looked in textbooks for similar problems. Usually I would find one, see the solution to the bit of the problem I was stuck on and be able to solve the rest of the problem on my own.

And of course I would remember how I solved the problem the next time and not need a textbook to help, so I had learnt something!

Obviously if you just copy the solution blindly without understanding what you are writing you don’t learn much, but thats the students choice and probably they will then fail the exam at the end of the course. I guess if the problems are assessed and count towards the final mark doing this is cheating (but in my degree the only assessment was the exam at the end of the course, so actually simply copying the entire solution would be a complete waste of time, as you might as well just not bother doing the problem sheet!).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: