Is it ethical to ask for somebody else's research code?

  • Thread starter lambrtz
  • Start date
  • #1
4
0
Hi all,

This is my first post here. I had a hard time determining the correct subforum to ask this question, as this is not a technical issue, and is closer to academic problem. Having read the subforum guidelines, this post doesn't seem to suit the subforum's purpose, but let alone other subforums, so I decided to ask it here. If the scope really lies outside this subforum's context, please move accordingly, and my apology.

I am doing PhD in Computer Science, and being a fan of code re-use, I have been using (free) programming libraries in my project, like Boost, CGAL, and Ogre3D. Just one short question: is it ethical to ask for somebody else's research code which are not made available, unlike those libraries that I have previously mentioned? We need to compare our methods with existing approaches, so I think we can cut down the implementation time by re-using other researchers' codes, but, again, I don't know whether it is ethical to do so (or is it what other CS students do?). Let's assume that the researcher doesn't explicitly mention about the availability of the research code in her/his webpage.

Thank you.
 

Answers and Replies

  • #2
Simfish
Gold Member
818
2
Well, if they're willing to give you the code, then why not?

It's only unethical if you take it without asking.
 
  • #3
ZapperZ
Staff Emeritus
Science Advisor
Education Advisor
Insights Author
35,847
4,676
I am doing PhD in Computer Science, and being a fan of code re-use, I have been using (free) programming libraries in my project, like Boost, CGAL, and Ogre3D. Just one short question: is it ethical to ask for somebody else's research code which are not made available, unlike those libraries that I have previously mentioned? We need to compare our methods with existing approaches, so I think we can cut down the implementation time by re-using other researchers' codes, but, again, I don't know whether it is ethical to do so (or is it what other CS students do?). Let's assume that the researcher doesn't explicitly mention about the availability of the research code in her/his webpage.

Thank you.

If using other researchers' codes is part of your research work as comparison, then I'd say that if you receive the permission to use it, then it is ethical. You just have to remember to acknowledge the sources of the codes in your thesis and in any publication resulting from it.

Again, if you are unsure, consult your Ph.D supervisor. He/she should know more about what is acceptable within your field of study. Besides, shouldn't he/she be aware of what you are doing in any case?

Zz.
 
  • #4
Choppy
Science Advisor
Education Advisor
Insights Author
4,687
1,860
It's completely fine to contact someone and ask. Explain the situation and what you're doing. So long as they agree, there's no reason you can't use their code - provided you credit them accordingly.
 
  • #5
4
0
Thank you for your responses. Considering the positive replies, I will discuss this during our weekly meeting with my supervisor. I notice the crediting issue, so when I make a publication, I have to give credit to the respective researcher in the acknowledgment, and also refer to her/his work in the References section.
 
  • #6
6,814
15
I am doing PhD in Computer Science, and being a fan of code re-use, I have been using (free) programming libraries in my project, like Boost, CGAL, and Ogre3D. Just one short question: is it ethical to ask for somebody else's research code which are not made available, unlike those libraries that I have previously mentioned?

There's certainly no problem in asking. They can say no. The one thing that you should clarify is the exact conditions of using the code. Ideally, they would include an explicit BSD/LGPL license to the code.

I should point out that there could be a pretty big mess if the code is given do you without a clear understanding of the conditions of the code. The problem that you can run into is if they hand you the code, you make substantial additions and improvements. Without an explicit BSD or LGPL license on the code, the code becomes quickly becomes unusable because neither of you can redistribute the code.

The other thing is that I'd like to discourage people from using full GPL license, as using the full GPL makes it impossible to reuse the code in any commercial project. We very heavily use open source software in the company that I work at, and when the code is BSD or LGPL we've contributed major bits of code and bug fixes back to the maintainers. Curiously no one (including often the maintainer) knows that we are doing that since we want to keep our involvement low-key, so all of our open source contributions are made under personal e-mail accounts.

When something is full GPL, we can't use it commercially which means that we have to rewrite everything internally which is bad for everyone.
 
  • #7
4
0
Hi, sorry for my late reply.

The one thing that you should clarify is the exact conditions of using the code. Ideally, they would include an explicit BSD/LGPL license to the code.

Well, honestly, I haven't thought about this licensing issue, especially that I'm not currently interested in commercialising my project (unless my supervisor tells otherwise), so thank you for bringing this up, twofish-quant. I think I have to ask about this too.

By the way, I have discussed about this matter with my supervisor, and he just suggested me to ask the researcher if I want to use her/his code, so I think I will do this immediately.
 
  • #8
6,814
15
Well, honestly, I haven't thought about this licensing issue, especially that I'm not currently interested in commercialising my project (unless my supervisor tells otherwise), so thank you for bringing this up, twofish-quant. I think I have to ask about this too.

Most people don't. The trouble is that it causes massive problems later. The saddest thing is when you see perfect good code that is totally usable because of licensing issues. The problem is not that anyone objects to the code being used elsewhere, but if you have a piece of code with uncertain licensing, then you run into the problem that it's unclear who the "owner" is, and if title is unclear then no one can give "permission" to use for other things.

Oddly, you run into the same problem with foreclosed houses.

The economics, sociology, law, and politics of research code is a very fascinating topic. From a Marxist standpoint, you can think of source code as the "means of production" and how to think about source code in terms of Marxist economic theory is something that I've been thinking a lot about.

By the way, I have discussed about this matter with my supervisor, and he just suggested me to ask the researcher if I want to use her/his code, so I think I will do this immediately.

One thing I suggest that you do is to save a copy of the e-mail somewhere. You can run into the problem where person A gives you permission to use the code, then person A leaves, and then person B gets annoyed that you are using the code.

Also if you get a firm NO, that's a good thing.

One reason I'm thinking a lot about this is that I'm in the process of modifying some CFD code to use GPU's, and I'm getting hit by licensing issues. Again, it's not the fact that someone objects to using the code, if you get a firm NO, that's a good thing. The problem is that you have code in which no one can say YES or NO, and legally you can run into big problems unless someone says YES.
 
Last edited:
  • #9
4
0
Wow, now the problem looks more complicated than what I thought.

One thing I suggest that you do is to save a copy of the e-mail somewhere. You can run into the problem where person A gives you permission to use the code, then person A leaves, and then person B gets annoyed that you are using the code.

I have recently sent an enquiry to a researcher, asking whether I can get a copy of his code. I will save all correspondences with him, including the license which I will ask later, in my hard drive. Thanks for your suggestion.
 

Related Threads on Is it ethical to ask for somebody else's research code?

  • Last Post
Replies
9
Views
3K
Replies
8
Views
1K
Replies
2
Views
1K
Replies
28
Views
45K
Replies
21
Views
2K
  • Last Post
Replies
4
Views
3K
Replies
3
Views
2K
Replies
18
Views
13K
Top