1

I'm pursuing an industrial PhD within a company in the context of an European project, having two directors: one from the company and another from the university. The topic of my PhD is related to the use of artificial neural networks, although the company where I work has no relation to this field at all, neither they plan to be related to it. Actually, I had to "sell" them that it was going to be useful for the company so they allowed me to select my PhD topic while I work in some other stuff. Anyway, my results belong to them as far as I know.

The problem I foresee is that:

  • In order to have some real impact with my research in AI, I think it would be convenient to publish the code as Open Source so other researchers can access and check my code.
  • If I don't publish some code, I will not have good proves of my work after the PhD as my thesis will probably remain confidential for some time.
  • At the same time, it would probably be convenient for the results not to be completely lost, as I foresee that nobody is going to use this work once I finish my PhD in the company.
  • Lastly, the project is funded with public resources and I believe unfair to privately keep developments that could be somehow useful for the scientific community (more if I achieve something interesting, and even more if they are going to waste my effort).

The question then is, what should I say to convince them to publish the code of my developments as Open Source?

Additionally, they don't have a public Github repository and so that, I think it would also be ideal to publish on my own repository (maybe specifying that they own the commercial rights?). This way, they could not remove the code whenever they want...

I insist, as far as I know, they own all the work I do. That's why I want to convince them. I'm not trying to do it without letting them know.

6
  • 1
    Well, better to work this all out now with the university and the company. I did a lot of my PhD research at Bell Labs, and published it all, so it is not unheard of in the least. Do you know anyone else at your company that has done a PhD in this way?
    – Jon Custer
    May 25, 2022 at 13:33
  • @JonCuster thanks for your reply. Sure, there are plenty of people who has made the PhD thesis this way in the company, but I kind of the first one doing something related to software instead of materials or biology stuff. So I don't think I can use other people as reference
    – JSL
    May 25, 2022 at 14:42
  • 4
    I guess the point I'm trying to make is that the company (say the IP lawyers) have experience with the requirements for PhD research material to be published. That may require you to limit some of what you publish (leaving out the important-to-the-company bits), but likely the other folks did similar things.
    – Jon Custer
    May 25, 2022 at 14:44
  • Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking.
    – Community Bot
    May 25, 2022 at 17:06
  • "industrial PhD within a company in the context of an European project" have you looked into what the project/cooperation contract between your company and the university says? May 26, 2022 at 19:31

2 Answers 2

2

A very short example: MacOS is released under a BSD license, but you can still charge others for any products you deliver if produced using a MacOS-based computer.

Put the focus on "we do not sell the software, we sell the use of the software and the training to use it" and good luck.

1

When you undertake a PhD program with outside sponsorship from a private organisation (whether it is an industrial firm or some other private firm) the university will need to negotiate an appropriate arrangement for IP and publication expectations up-front. In order for this kind of arrangement to work, the university must be satisfied that you will be able to publish scholarly work and have the appropriate evidentiary backing available to defend that work, so that you can meet the requirements of your PhD. If open source code is necessary (or even valuable) for providing evidentiary backing to your work, then the university might consider it to be an important aspect of your research. If the sponsor imposes restrictions that prevent this, then the university should not accept the arrangement. On the flip side, if you are going to receive private sponsorship for your degree, you will generally expect that the sponsoring firm ought to get something out of it. Consequently, it is necessary to negotiate an arrangement where you have the ability to meet the expectations of your PhD (inncluding publishing your work) but you also give something of value to the sponsoring firm. If those conditions can't be met, then this kind of "joint venture" ought not be undertaken.

As to how to negotiate this, you shouldn't feel like you need to do this yourself. Universities have experienced academics and legal advisors who are highly knowledgeable about these kinds of arrangements and negotiations --- ask them to help you negotiate an appropriate sponsorship arrangement with this industrial firm. If the university is doing its job, it simply will not accept an arrangement that prevents you from publishing your work and meeting other requirements of a PhD program. (Often in these cases, a sponsoring firm gets the benefit of having academic research created that helps it to do its work, even if they hold litte to no IP in that research; in some cases, they might negotiate a "head start" on knowledge of the material before it is published. In any case, the university ought to be pretty good at negotiating an outcome that can satisfy both institutions.)

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .