0

Suppose Dr. Smith at Stanford developed some open-source software there, and is getting paid from an NIH/NSF grant to continue doing it. But suppose he wants to move to Harvard.

While his software is open-source, Stanford owns the relevant trademarks, and possibly the website.

Would this make such a move difficult? Do software authors taking faculty jobs typically pre-negotiate contracts saying that they'd own the trademarks?

If the website was under github.com/smithlab/, would Dr. Smith get to keep that at least?


Clarifications:

  • This question is not about about the law. The law was perfectly clear to me when I posted the question -- Stanford owns the trademark
  • It's not about the copyright -- while Stanford owns it, the software is open-source, so anyone can distribute derived works (under a different name)
  • The trademark does not have to be registered in order to exist
  • The software, as the question implies, is important enough that labs get funded to work on it
  • Stanford might like to continue to have the funding and the credit go to Stanford

(All names made up for the purpose of an example)

6
  • 1
    What do you mean precisely by "relevant trademarks"? Do you mean the copyright?
    – user176372
    Commented Jun 30 at 19:23
  • @user176372 No, I mean trademarks. The name of the software can be considered a trademark.
    – MWB
    Commented Jun 30 at 21:59
  • 4
    Do people in your field often bother to trademark their software names? Commented Jun 30 at 22:01
  • @AzorAhai-him- You don't have to register them, in the US: uspto.gov/trademarks/basics/why-register-your-trademark You own them automatically. If you wrote a popular program called "FOO", anyone else doing this in the same domain will be in violation.
    – MWB
    Commented Jun 30 at 22:06
  • 3
    @MWB Okay, do people in your field bother about that? Commented Jun 30 at 22:32

2 Answers 2

5

I've done this several times, albeit for software that isn't especially high-traffic. Long story short: Don't enmesh your accounts and intellectual property with your university's. Keep software on accounts that you control, not your employer. Don't name your software after your employer. Don't include a bunch of your university's visual branding in your software. Etc. etc.

Trademarks...

Part of your question regarding trademarks is unusual to me. A general statement is that, if you have serious questions regarding intellectual property, you should obtain actual legal advice. I'm not a lawyer. I've never heard of research software containing a registered trademark. Don't, I suppose, include your university's trademarks in the name of your software.

In my experience, universities have generally not asked me to sign over copyright to any of my works, including research software. Even if they did, if the university had approved distribution under an open source software license, I'd be entitled to do whatever I wanted with it as long as I provided appropriate notice under the license terms. Most open source licenses are non-revocable and perpetual, meaning that the copyright holder couldn't decide later that I wasn't able to do what I want with it.

In this context, switching over is trivial.

possibly the website

Any time you move institutions, your websites with them are going to be shot. You either have to host with your institution and understand that leaving will require you to recreate the webpage at a new home, or else set things up so that you have personal control of the domain name people associate with your software.

If the website was under github.com/smithlab/, would Dr. Smith get to keep that at least?

As long as they still have access credentials, sure. The person who wrote the website owns the copyright on its content, except for content that they're using with permission from another owner. Maybe you'd have to strip university visual branding if that's embedded in the webpage.

7
  • 1
    And NIH or NSF grants likely have something to say about IP ownership in them.
    – Jon Custer
    Commented Jul 1 at 0:25
  • 1
    I think that this answer is poorly informed. Copyright and intellectual property by default resides with the employer unless you have made specific provisions to the contrary. Whether or not you slap a license on it is a separate matter: Even if your university agreed for you to distribute the software as open source, you cannot later turn it into a commercial project (perhaps dual-licensed) without your prior employer's approval. The point is that license and IP are orthogonal dimensions. Commented Jul 1 at 1:36
  • 3
    @WolfgangBangerth I've responded to your answer-- I don't think you're particularly well-informed on this question either, or is anybody who doesn't specialize in this area of law. Thus why I recommend getting legal advice if the situation is serious. Additionally, I intentionally suppressed getting complicated about licensing, but you seem to be saying that I suggested violating license terms. I expressly did not. Most open licenses allow very broad use of the licensed software, including commercial use. Many allow one to go closed source with derivative works, even.
    – user176372
    Commented Jul 1 at 9:08
  • 1
    @user176372 Well, but you do. You say "I'd be entitled to do whatever I wanted with it" which is simply not true. The software you wrote does not belong to you. It belongs to the university. Whether or not it is covered by an open source license is immaterial in this regard: The university may have agreed to distribute it that way, but it continues to belong to the university. The fact that it's under an OS license means that if you switch employers, the old university can't prevent you from distributing it that way, but they can prevent you from commercializing it. Commented Jul 1 at 20:17
  • 2
    @WolfgangBangerth: "they can prevent you from commercializing it" - "commercializing" and going by the terms of an open source license are not mutually exclusive. Above, you mention the exemplary scenario of dual-licensing the product later on, which the former employer can indeed prevent. However, if the software has been released under a suitable permissive open source license, I see no real reason why the author couldn't take a later version after switching jobs, add e.g. non-OSS modules, and commercialize that complete package. Commented Jul 1 at 21:43
1

@user176372 is correct in their answer that it's useful to ensure early on the lifetime of a project that you will be hosting websites, project repositories, Q&A forums, email lists, etc., on domains that you control. This is practical, because it ensures that you can move to another university, give the project to someone else (say, if you retire), or simply because universities change their minds what open ports they want to provide on their machines to host services -- generally, this has become much more difficult than it was twenty years ago when every employee could basically run whatever web service they wanted on their desktop.

There are also legal questions. At least in the US, the intellectual property (copyright) of work that employees produce -- including software -- rests with the employer. That means that, strictly speaking, when you move to another university, the rights to everything that has been developed up to that point remains with your old university. That pertains to the software itself, the documentation, the trademark (if any), etc. In this regard, you're really no different than any employee of a company: If you've worked on something for Google and you move to Microsoft, you can't just take all of the source code you wrote along and expect that to be ok. I will note that that is no different if the software was written for contracts funded by NIH, NSF, or other federal agencies: The Bayh-Dole Act says that even though the Federal government paid for it, the university owns the intellectual property.

Now, in practice, university generally don't have much interest in small software projects -- say, things for which the market potential is less than a few million dollars over the course of a few years. That's because to bring something to market costs money: marketing costs, writing user interfaces, supporting users, etc. As a consequence, most of the time universities will not care much one way or the other what you do with academic software, but that is not something you should rely on. Instead, have a conversation with the IP office of the university, explain the scope and size of the project to them, and see if they're willing to agree that the project can be distributed as open source (which, strictly speaking, you cannot do unilaterally without your university's approval). My experience having this kind of conversation is that it is typically not all that difficult to get an IP officer to send you an email saying "yes, we're ok with you distributing this as open source". In my case, this included explaining to them that I will likely bring in more money through external grants than we could by making things into a commercial product. This is useful because it essentially gives you a perpetual right to keep distributing the software as open source, whether or not you stay at that university.

Then, when you go to a new university, have that conversation with them before you sign the contract. Here, too, my experience is that it is not all that difficult to get them to agree up front that open source is ok with them.

The point I'm trying to make is that it is far preferable to have an email from someone in a position to make a legally binding determination over just "hoping for the best and otherwise feigning ignorance". In the end, this is a legal question. You should inform yourself of at least the basic tenets of who owns intellectual property, the relationship between intellectual property and software licenses, and adjacent questions.

1
  • 2
    I'm upvoting this because I don't think it's an unreasonable answer, but it's also very distinct from my experiences in a university setting. I'm not confident on the default copyright status of scholarly work, including software, created by faculty. That being said, my current and former university employers have had clear policies disowning any copyright interest in traditional scholarly works, including software, and defaulting it to the creator. The AAUP has a statement in that regard that they follow. aaup.org/report/statement-intellectual-property
    – user176372
    Commented Jul 1 at 8:58

You must log in to answer this question.

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