Lessons learned from a student licensing to a startup

The purpose of this page is to relay my experience so that others can hopefully have a smoother time through the process. I've organized this by the stages that I went through, so if there's some stage that you are particularly interested in, you may find it faster just to skip to that section.

The whole tale

Background

My quals project chiefly involved the creation of a new algorithm, writing it in code, and then running experiments on it. The only people involved in it were me and my advisor. As with most quals projects and research, it's tricky to say exactly who made what contribution.... other than the fact that I wrote the code. ;)

Around the time that I was finishing up my quals project, my advisor started a new company. I fully believe that he, and all of the people in this story, were acting in good faith and attempting to do the right thing. So keep in mind that what I tell you here is what happens when people are trying to do the right thing (both for me and for them). I blame no one for anything that may have happened; everyone was trying to do the right thing, but sometimes what the right thing was was unclear.

When my advisor started up his company, since he had had discussions with some students that helped to formulate ideas that would relate to the company, he wanted to give those students some stake in the company as a reward. He said that the idea was that in order to make this happen, the company would probably license my code, even if they didn't use it. This was incredibly generous.

Disclosing the invention

What this stage was

In order to license anything outside of the university, it must be disclosed to the Office of Intellectual Property and Technology Transfer (OTL). There are a number of reasons for this, but the summary is that OTL is required to make sure that we are sticking by federal and state guidelines regarding IP. The form of this disclosure is the Invention disclosure form, which requires you to fill out a number of different pieces of information, including who was on the invention and what funding was involved.

What happened

I think almost every area of mine was filled out incorrectly. I got the form during one meeting, and because things in startups move quickly, my advisor wanted it turned in the same day. First, because my advisor was trying to license primarily to make things good for his students and he had two students with related, though completely disjoint, work, he decided that the best thing to do would be to have both of us sign on the same invention disclosure form. The second thing that happened was that in the funding box we were required to check whether there was federal funding involved. I asked my advisor, and he said no, he didn't think so, so I checked no and sent it off. If there is federal funding then OTL is required to pursue rights that it is not otherwise. So it is very important to make sure that this box is filled out correctly. Don't settle for an "I don't think so".

How this complicated things

So that's what went wrong with the invention disclosure form. What was the result of this? The form went off to the people at OTL, and they started processing the paperwork under the assumption that the paperwork had been filled out correctly. Legal things like this take a lot of time (both from OTL and whatever company you are working with), so OTL went down the road pursuing whether the university had any responsibility to own part of the research for some time. By the time that they decided that something might be awry and that they should talk to the students, I believe that about a month had passed. In this time the company had started negotiating with both OTL and us students. But because this piece of paper had been filled out incorrectly, we needed to start over again. In this case it was particularly tricky because while there had been no federal funding in my case (since I had an NSF), the other student was on federal funding, and lumping the two inventions together complicated things. Eventually the correct information was filled out and sent back to OTL, and the process continued, but the time and aggravation could have been avoided. At the end of this, the forms were filled out again and it was ruled that I owned my work, both the code (which is normal regardless of your funding) and the intellectual property for the algorithm.

What I should have done

The first thing that I should have done is that I should have resisted the idea of us putting our work on the same piece of paper. Do not list your work with disjoint work when filling out a disclosure form. Obviously, at the time I had no idea how this was going to complicate life, and I thought that since my advisor was telling me it was the right thing to do, it was probably the right thing to do. The key here is that even though I was correct in trusting my advisor's intentions, my advisor is here as a researcher, not an expert in technology transfer. I should have talked to someone else. Similarly, I should have made sure that someone was clear on the meaning of the federal funding issue before signing the paper. Be sure that the federal funding information is correct.

In particular, the people that I should have talked to were the members of the members of the department's commercialization oversight committee (who are listed on the commercialization web site). These people are there to make sure that the students, faculty and staff have someone to talk to about these issues, and I should have used them as a resource. It would also have been appropriate for me to talk to the people at OTL (who I believe both have the best interests of the students at heart and knowledge in the area), but at the time I had been told that OTL was slow and that it was best to avoid them. But the fact of the matter is that by talking to them earlier, the whole process would have been much faster since we would have avoided the loop of learning that the forms had been filled out incorrectly.

Negotiating with the company by myself

What this stage was

If it is ruled that you do not own your work, then you have no choice but to turn over the negotiations to OTL; they are obligated to negotiate for the rights for the university (and by extension, you). If you do own your work, then you have the choice of assigning your rights to the university and having OTL deal with it (and thus take some of the profits, should there be any), or licensing it yourself. The company brings in its lawyers and you deal with them directly (or hire a lawyer to do so on your behalf.

What happened

Initially, since the whole idea was just to give equity to the students, I was told that the company wished to license my code, since that is the easiest way for this to occur. I was nervous about dealing with the company, so I went and talked to the commercialization oversight committee and the people at OTL. Based on these conversations, I came to the conclusion that since the company was only interested in the code, and I felt qualified to make decisions about what was going to happen to the code, I decided not to involve OTL and to license the code myself. I met with the CEO of the company, and we began to negotiate what the license would be. This part was fine.

After the first round of discussions, it was decided that the company was not interested in the code; they wanted to license the algorithm. What this means is that instead of just using the code they were interested in being the sole user of the idea and perhaps filing a patent. Because my advisor was involved in the work, the university was obligated to pursue patent rights. Even though my advisor did not want any compensation (and in fact asked if he could be taken off of it if it would make things easier), the university was obligated to pursue the patent rights. However, patent/intellectual property laws say that the company is able to negotiate with each individual separately, and because the first round of negotiations had been pretty reasonable, I decided to continue to negotiate on my own and not turn things over to OTL.

After another while had passed, I was presented with a license agreement for the algorithm that seemed to me (note that I have no legal training) to assign most of my rights to the company in all perpetuity. Should I ever work on this area again, the company would own it. I would not have a right to let anyone (including myself) use the code. I would not have the right to publish without their permission. I immediately talked with the commercialization oversight committee, the department chair, and OTL. All of them advised me that I should either hire a lawyer or to turn things over to OTL. The people at OTL were particularly helpful in explaining what the terms meant and my options were without requiring that I give them any rights, and I advise anyone even close to this situation to talk to them. Since my primary function in grad school is to be a grad student, and this was already eating up more of my time than I was happy with, I chose to assign the rights to OTL.

What I should have done

I did pretty well initially on this stage; I talked to people and got a lot of input. Make sure that you talk to knowledgeable people so that you are well informed at every stage. Had it been the case that all that they wanted was the code, I probably could have negotiated the license myself. At the point when they mentioned that they were interested in negotiating for the algorithm, however, I probably should have talked to people again, and at least have been better informed as to what was happening. In hindsight, it would have been much better for me to turn the process over to OTL at that point. My final decision at this stage (to turn it over to OTL), given that I valued being a grad student more than licensing the technology was, for me, absolutely the right choice.

The rest of the story

There is nothing left that happened to me, but some of it may be instructive to you, so I'll include it.

Several things occurred because there were wrinkles in the laws according to how a faculty member can license an invention to his or her own startup. Hopefully this has now been dealt with and it will not affect you. Nonetheless, this continued to delay things.

The next wrinkle appeared because I had previously discussed my work at a workshop. Once you have disclosed the work in a public forum, you have a year to patent it, or your US patent rights expire; I believe the international patent rights expire as soon as you present it unless you have already made moves to patent it. Because things had dragged on, if there was to be anything to patent, OTL needed to file a provisional patent. I got to go through a round of discussion with lawyers over the patent. This was both informative and annoying.

Last I heard (6 months ago) OTL and the company were arguing over details and who had caused delays, who started on the complicated and arguably undesired process to patenting, etc. I have kept out of it, and this was made much easier because I had kept all relevant mail and had kept good notes of who had said what. Keeping track of these details made life much, much better.

The whole process has now been ongoing for approximately 17 months, and there is still no resolution. I do not expect there ever to be a resolution on this, and frankly that doesn't bother me; I have learned a number of things, and that's what grad school is all about.