I'm an IP lawyer, but I'm not YOUR lawyer and so don't take this as legal advice about your particular situation; if you do consult a lawyer, this might give you some background info for use as a head start in your discussion.

1. U.S. patent law says patents belong to the inventor(s), absent either (i) an agreement to assign the patent rights, or (ii) an implied-in-law duty to assign, such as arises when an employee is "hired to invent" or "set to experimenting."

If your university has some sort of written policy about undergraduate- or graduate work belonging to X, you might well be held to have agreed to it by applying and/or enrolling, either in the school or in your particular course. (pg has it exactly right on this point.)

2. U.S. copyright law says that the copyright in original works of authorship (which might or might not include elements of your project) are owned by the author(s) UNLESS (i) there's an agreement to assign the copyright, or (ii) the author's authoring activities took place within the scope of his employment, or (iii) the work fits into one of a comparatively few specific categories -- translations, contributions to collective works, and some others -- AND the parties agreed in writing that the work would be a work made for hire.

3. U.S. trade-secret law might be a little trickier, because it varies state by state. The pretty-much-universal rule, though, is that at least some degree of secrecy is a sine qua non of trade-secret rights.

Points 1 and 2 are sort of fascinating to me. As far as I'm aware, patents and copyrights are defined in the constitution as monopolies granted to "Authors and Inventors". Do you know how it is that the laws, which are supposed to build on the constitution, have extended ownership of copyrights and patents to corporate entities? It seems to be directly opposed to what the constitution allows.

The constitution grants them to inventors, who can assign them to others via contract. How is that unconstitutional?

I can see how inventors can grant _rights_ to their patents to other entities. I don't see how they can grant _ownership_ (whatever that really means) of their patents to third parties. The constitution allows the government to grant the monopoly to the inventor, and the inventor can license other entities to exploit the patent, but it seems like the ownership of the patent has to stay with the inventor. Perhaps it's just a semantic argument, but it's one that's been bothering me for a while.

Whatever definition of "ownership" you want to work with, you can come up with a grant of rights that equates to it, and a reasonable argument that people should be able to trade those rights for something more valuable to them.

So, in other words, I have no idea what your point is.

I don't think I've made any point yet; I'm mostly fishing for information from people who actually do know things, and trying to figure out if what's on my mind makes any sense. Two possible advantages (for people) of individual ownership of patents over corporate ownership:

Contracts require consideration. If I own something and sign a contract with you that allows you exclusive control over it, then I have to be getting something in return. I'm not a lawyer, but I believe that there's some restrictions on consideration that can render a contract void if it is incredibly lopsided. I don't know for sure, but I'm guessing that a contract exchanging an invention earning billions per year for a five figure salary and a position that can be terminated on a whim might not be a valid contract. I really don't know about that though.

The other nice thing about personal ownership of copyrights and patents is that it makes the whole retroactive copyright extension and post-mortem copyright assignment transparently illegal. If only the creator of a work gets the monopoly over its distribution, then the termination of the creator must entail the termination of the monopoly. I'm very much not a fan of eternal copyright, so that alone would make me a bit happier.

I think I've probably threadjacked this way too far, so I'll just shut up now. Maybe I'll start a blog somewhere where you can rip holes in my half-assed ideas without us bothering anybody else :)

Only extremely rarely does a billion-dollar invention arise. When an employee signs an invention-assignment agreement with his/her employer, the employer is on the hook (usually) for a known salary, benefits, etc., in exchange for which the inventor is giving up his rights in inventions of unpredictable number and -significance. That usually gives the employer the bargaining power.

It's not always that way: I once had a software-company client that paid commissions to its product architects; the architect of one of the company's flagship products ended up making a ton of money, as did its shareholders. Sadly, companies that can do this successfully are few and far between.

In addition to this, don't forget that corporations are legal persons.

Edit: In the US and England at least.

Continuing my numbering of before:

4. 'Ownership' is to a certain extent a legal fiction - it's simply a label we use to indicate when the government will use force, if necessary, to allow the 'owner' to do X and/or to prevent others from doing it.

5. In that sense, a patent or copyright conveys 'ownership' only in that the government will (upon a proper showing) enforce specified penalties against non-owners who do things they're not 'supposed' to do.

6. Congress long ago stated, in the relevant Acts of Congress, that patent- and copyright 'ownership' can be transferred to others, including corporations and other entities.

7. So basically, the brute fact is that 'ownership' of a patent or copyright can be transferred because: (i) Congress said so, and (ii) the guys with the black robes and gavels follow what Congress said, and (iii) the guys with the guns and badges go along, and (iv) ultimately, the guys with the BDUs and M-16s and tanks go along too.

8. The above doesn't automatically prove that Congress made the right call on this point -- but I don't think there's been much of a controversy about it in the past 200-plus years.

Every patent application that I have been involved in, I and my co-developers are listed as the inventors, not the company. All inventions are assigned to the company by our employment agreements.

That's helpful information. I'll have to dig up what I signed when I agreed to go to school here and find out what it said about research.

Most universities have some kind of IP policy. E.g. Harvard says that software you write using their money or facilities belongs to them:

http://otd.harvard.edu/resources/policies/IP/

In practice I've never known IP to be an issue for undergrads. So I wouldn't annoy your professors by bugging them a lot about this.

You are many, many times more likely to lose because you built something people don't want than because your college sued you over IP. So if you're going to worry about something, worry that what you're building won't seem desirable to customers.

Bugging your profs will only show that you couldn't be bothered reading up on your university's by-laws and regulations. It could result in a down-mark.

It is really good to be confident, but in reality your undergrad work might be a lot less brilliant than you believe it to be. Some universities will actually co-venture with you if your work is patentable, etc and although you end up sharing revenue, they might foot the up-front legal costs (which can be rather substantial) - again, research what applies in your particular case.

Good Luck with your project and subsequent ventures.

Unis should also have less interest to pull a dickish move than a "mystery" cofounder may. So even if you were technically using their facilities to write your code, if a uni went after you for that it could give them tens of millions in bad PR.

Reputation matters a lot for them. Very different from some random individual who may have little to lose in suing you for stealing his idea for a social network.

When I turned a university project into an internship with them, we made it very explicit to everyone concerned exactly when the project started becoming "something other than class work" and what consequences that had for all parties involved.

I tend to think that 999 out of 1,000 legal worries people have are groundless. Want to sleep better at night? Draw up a simple letter of understanding between you and the University and ask for your professor to sign it. After doing so, take it to the university counsel or IP office (major research institutions have one) and ask them to sign off on it. If they do, congratulations. If they don't, consider your options.

This. Especially regarding your tech transfer office. Usually, you've signed something somewhere that says the university owns everything that you do that relates to them in some way. So you've gotta get them to either give you a release on the IP, or an exclusive license.

Thank the Bayh-Dole act.

Hm, I hadn't realized that the tech transfer office was now routinely dealing with undergraduates. Ouch.

They aren't. This is why I recommend not worrying about it. However, if you're the worrisome sort, "Look, you and I both know you don't really care about undergraduates. How about you sign this piece of paper? You aren't giving away anything you care about, and it gets me out of your office quickly!" works wonders.

Actually, most of us are undergraduates, and the project isn't affiliated directly with the university. However, we do get together and use thier lab for development, so they can still claim rights, since we used their stuff.

EDIT: Also, apparently I went to his school.

Any tips then if I wanted to get around this? In my case I won't be using anything from the University for the project. To be honest, I might do work on it in the library from time to time or something, but I could even avoid this if absolutely necessary. I'd literally just be doing coding on my laptop and using my own hosting.

I'd just like to see the language behind the agreement I signed I guess.

I found some documentation online about several policies surrounding this: http://www.pitt.edu/~offres/policyu.html

It's not really an easy read, but it seems that, by default, the University wants in on anything you do in their facilities. They make this condition for students, though:

"Students are exempt from this policy for cases in which University facilities are not used and where the invention or discovery is not made in the course of the students studies at the University."

So since it's sort of "in the course of the student's studies", I might have to go after the following conditions in the policy:

" If, however, the inventor or another institution believes that the circumstances surrounding the invention, including such factors as support provided by other than the University, place where discovery was made, or lack of relevance to the regular work of the member of the faculty or staff, warrant another distribution, the inventor or the institution may request the Office of Technology Management, in consultation with the Technology Transfer Committee, to review the circumstances. After review, and upon consultation with the Office of General Counsel and the Senior Vice Chancellors, the Office of Technology Management will make recommendations to the Chancellor on the distribution of proceeds.

      The University, as determined by the Technology Transfer Committee and the Office of Technology Management, may choose to waive all rights to file a patent on a particular invention or to pursue licensing of such invention, electing instead to grant the inventor permission to proceed on his or her own in whatever manner the inventor deemsappropriate."

Most of the language seems really targeted towards faculty and staff, however, so I think I can work around it. I'll try the person you suggested once I hear back from my department head.

Yeah. Unfortunatly, the OTM website is apparently down: http://www.otm.pitt.edu/

As luck would have it, my cofounder here used to work for them... he says that you're probably fine, but you should go to the office and double check. Or you can email Dan Bates at dbates@provost.pitt.edu, who's apparently a reasonable dude there.

For those curious, the response from my department head (who I worked with to set up the independent study):

" This is a complicated question.

Effectively, the University owns everything (all IP) that we do while we are here, faculty, staff, and students alike. However, they have very limited capability to effectively copyright, patent, and market everything that is created. If you were to go through the process of doing this through the University, they would return a portion of any income that they acquire. In return, they cover the legal, patent, and marketing costs to get the product to market, or to find some company that would want to market it. Given the number of things that are created in a university, the process of selecting which items to pursue is long and involved.

My suggestion is that you proceed on your own. If you are hugely successful, then you would deal with the university at that time. "

So it seems like most people here are right on. If I can't easily get any more information, I'll probably just go for it. I was just trying to see where I stand so I don't get myself in a ridiculous situation later where I say "Why didn't I think of this earlier?".

> In return, they cover the legal, patent, and marketing costs to get the product to market, or to find some company that would want to market it. Given the number of things that are created in a university, the process of selecting which items to pursue is long and involved.

Translation: the university is where your IP goes to die. The professor in this article (sorry, couldn't find a complete one anywhere) http://www.highbeam.com/doc/1G1-13985948.html is a family friend, and his invention (violin bow made for <$10 in parts that beats >$1000 bows in blind tests) is 100% legit, but last I knew the University of South Florida was just sitting on it. He was still experimenting and building bows for himself and friends, but there was nothing he could do to actually get it to market.

My recommendation: do NOT do it under these circumstances. See, if they own a portion of the IP and you're successful, they'll probably come hunt you down in the future.

So steer clear of this situation because they are not giving you the necessary clarity you need at this stage. Develop it at home on your own computer without using any of the uni's resources and do something else for your project.

The project will probably still kick ass and you'll own the full IP of your company.

Frankly, that sounds like a bullshit policy, and certainly doesn't apply to all universities. Students are paying for access to the resources of the university and shouldn't be giving up IP. UMich has a much more reasonable policy here: http://www.techtransfer.umich.edu/resources/policies.php

........

The University will not generally claim ownership of Intellectual Property created by students. (A “student” is a person enrolled in University courses for credit except when that person is an Employee.) However, the University does claim ownership of Intellectual Property created by students in their capacity as Employees. Such students shall be considered to be Employees for the purposes of this Policy. Students and others may, if agreeable to the student and OTT, assign their Intellectual Property rights to the University in consideration for being treated as an Employee Inventor under this Policy.

........

Depends on the university as many have noted. Some universities have a technology transfer office or a commercialization director or something similar. I'd double check with them BEFORE you do anything for your project.

He's right. Find the Technology Transfer office today and tell them what you're working on. Likely they won't be interested and will be happy to give you something in writing that says they refuse all rights.

The University probably gets the copyright to your code.

There are several ways around that. One is to link to a GPL'd project. Now they own the copyright, but they have to keep the source code available for others to use for any purpose. Including your business. (If in doubt, you can always do something like cut-n-paste Emacs' malloc, link in libreadline, etc.)

Another option is to just rewrite the code. Copyright only covers verbatim redistribution of the literal program text, which is generally not too valuable. It is very easy to write a program a second time. Ideas and algorithms are not copyright-able, after all, nor is the experience in developing software you gained from writing it the first time. Those things are where the value is, and copyright does not protect them. (Software patents may, but the University will have to apply for one before that's even an issue. Then they will have to spend their own money to sue you, and then they will have to collect damages. Not going to happen unless you become the next Google, and if you do, you can afford to pay them off.

OTOH, software patents are almost dead. So this might not be worth worrying about.)

So anyway, to be extra safe, be open source, or just redo anything that you did with University resources. That is a lot faster than dealing with legal problems that may arise if you are successful.

(Then again, most startups fail pretty quickly, so perhaps it's not really worth your time to "care" this time around.)

When I enrolled at York, it was required that I sign a contract which laid out the IP situation. IIRC, this was fairly lenient in my favour, although this was for an undergraduate degree. I gather that the situation was different for PhD programmes. You have probably signed something like that too when you signed up, in among the endless paperwork. Dig it up and read it. Chances are you're safe as long as you don't use their equipment and facilities.

If there's any remaining doubt I'd probably try to solve the problem in advance by drafting a plain-English document which declares that the university has no rights to the works and get the advisor, the departmental head, and a relevant person who oversees research across the entire university to sign it. In fact, you probably have a dedicated person who oversees such matters at the university, so it may be best to take the draft to them directly and see if they have an equivalent standardised contract that they'll be more willing to sign than your draft.

Now that is an interesting question.

This advice may be so ill-informed that I'll put the disclaimer up front: I'm not a lawyer, let alone an IP lawyer. That said, if the university isn't paying you, and you don't use any university resources, and you don't assign them any rights, I don't see how they can assert any rights. Looking over your work and assigning it a grade doesn't make it theirs. Otherwise they would own everything.

Now, if you were an assistant professor using university offices and labs and a certain amount of public funding to conduct work that is nigh-indistinguishable from the work your side business is doing -- to the extent that many of your grad students can't tell from moment to moment if they're doing schoolwork or startup work -- that would be a more interesting question.

> and whatever I signed when I committed to coming to the University four years ago.

I think this "education contract" could state that the university has the rights to use your generated IP.

That University wouldn't happen to start with a D would it? I know my university definitly says that anything you do while working on your Senior Project is technically theirs and makes you sign a waiver agreeing to it. However, you situation seems to be a little different if your saying you have some sort of agreement to not use their property to create it.

My schools actually listed in my profile for the curious. But no, I haven't signed any kind of agreement saying that I won't use their property. The only agreement that would be standing currently is if I agreed in my acceptance letter to let them keep any IP I create as part of classwork, etc. That's something I need to look into... very soon.

I used to be a Manager at UNLV. I thought that Nevada was the only state that allowed IP to be retained by the individual instead of one's employer or school. In fact one of my staff created and patented 2 devices while working there.

Why not just do something else that would meet the university's criteria for a good mark. Then after you leave, then you do the project as you envisage it. Free from any complications, etc.

This isn't legal advice, but software written by undergrads and not funded by the University is usually owned by the student.

As a UK student in the last millennium we signed a form on matriculation that handed over copyright and patent rights to the Uni for any work done as part of our study there (or I expect using their facilities; but definitely the former).

I absolutely would bug the department about it.

If the issue is "copyright" the author retains all interest in the copyrighted work unless a specific form is signed; there is no other way to assign copyright.

You should be easily able to ask for and receive copies of anything you signed when you first came to University.

Personally I would not worry about it, just go ahead. Further don't bug faculty about it, it will only raise red flags.

The problem is that if the OP's startup is a success then the University will come after him for what they perceive as his abuse of their property. Also if he has external investors unless he has disclosed this problem to them they could well sue him personally - I've been threated by legal action by an unhappy VC and that was also deeply unpleasant.

If I was the OP I would walk away from his code - don't even take a copy of it. I would check with an IP lawyer what measures I would have to take to produce a "clean" copy that the University has no claim over and base whatever I did on that recommendation. At least if you do things that way you have some comeback against the lawyers.

I have no code written currently, and have only met with my adviser to have him sign off on his agreement to oversee the project. I'll probably go the route of getting the school to sign off on something saying, basically "I won't use your stuff, you don't have to do anything for me, just don't sue me later".

If they won't do that, I might have to change my project to use another web app-type idea that I consider less viable as a business but could teach me a lot about web development. I hope that's not the case, as currently I'd be earning 3 credits to build my dream project and if I can retain all IP I'd be golden.

It is a tricky one - a sensible organisation would wish you the best of luck and hope that your eventual success reflects well on them. However dealing with any large bureacracy means that finding someone who will sign off on such a thing can be tricky!

Yeah, agreed. I've found the contact information for the person in charge of our tech transfer office (called the Office of Technology Management... I assume they're basically the same thing). I'll wait to hear back from my department head and probably contact someone in OTM (whoever is appropriate) afterward.

This might be history in the making, as I potentially might have to pitch why my idea is NOT viable enough for them to care (inverse VC pitch, lol).