Open Menu Close Menu

Product Development | Feature

Taking Homegrown Products to Market

Purdue University made the decision to pursue a commercial partnership to take Signals, its student-retention program, to market. Here's why.

Purdue University (IN) has partnered with SunGard Higher Education to commercialize Signals, a student-retention application developed on campus in 2005. The product won a Campus Technology Innovators award in 2011. Campus Technology spoke with John Campbell, associate vice president for academic technologies and Signals' chief architect, about Purdue's decision to choose a commercial partnership over an open source approach.

Campus Technology: Why did Purdue decide to commercialize Signals in partnership with SunGard?

John Campbell: As a campus develops tools or applications, it has many things to consider. In the case of Signals, we were content with the progress we were making with the project on the Purdue campus. Then, about three years ago, NBC News did a story about Signals. All of a sudden we had many schools contacting us, asking if we could give Signals to them or if we could help them develop something similar. We did not have enough resources to support all these schools, nor was it something we were interested in pursuing. The code is written very specifically for our institution, pulling data from specific applications. So we were faced with pursuing options that could make Signals applicable to other campuses for predictive modeling.

CT: Did you consider making it available under an open source license?

Campbell: First, you have to ask yourself if your project is a viable open source effort. Is there a high level of interest among institutions? Is there a development community available? Does the project have a long life, so the community can develop over time? When we initially developed Signals, analytics was still in its early stages. At that point, I felt it would be a stretch to get enough people together in the open source community to create viable code. The decision might be different if it were made today. Sakai is starting to build an open source analytics component that will probably be ready in a few years and the nature of the decision will change.

Means of Production
Universities are wrestling with the possibilities and pitfalls of making homegrown IT products available beyond their campuses. In "Means of Production," writer David Raths examines the benefits of the two major options: open source or a commercialized venture.

CT: What was attractive about the idea of working with SunGard?

Campbell: A university has to realize its limitations. Most institutions do not have unlimited funds to support adding new innovations all the time. SunGard was prepared to be a full partner and develop the system to accommodate multiple student information systems and course management systems. SunGard is also well positioned to stay on top of those systems as they change. There's no way we would have the resources to do that.

CT: What were some of the issues that had to be negotiated as part of the commercialization agreement?

Campbell: When you work with an outside entity, you define the box in which the collaborative work will take place, and then you find how things might work inside that box. For Signals, we created a partnership that allows SunGard to develop a product for many types of institutions. We also retained our ability to continue our research and efforts based on Signals. If we find new models, SunGard has the right of first refusal to new features/approaches. We worked with our Office of Technology Transfer to define the nature of our collaboration. Because I was the inventor on this project, I was not involved in the negotiations with SunGard. The lead role for our department was our CIO, Gerry McCartney, who is fabulous at developing collaborative partnerships.

CT: Does Purdue have other products that went open source?

Campbell: We have HUBzero, a platform for scientific virtual organizations to share data and run simulations. One example involving nanotechnology is nanohub.org, which has 200,000 members and is completely open source. We are seeing growth in the use of these hubs.

CT: Is Purdue working on anything else that has potential for commercialization?

Campbell: Right now we are working on five teaching and learning applications and talking to different groups about commercializing them or putting them in a consortium model (see Purdue University Studio).

CT: Do you have any advice for universities looking to develop open source software for wider distribution?

Campbell: Be aware of what other technology you will be integrating with. For instance, don't get hung up by requiring others to license a scientific library to make it work, because that limits your options. Be aware of your own institutional policy for releasing an open source product. Universities are becoming more realistic, I think, than they were 15 or 20 years ago, when they thought everything might be the next Netscape. They realize everything developed on campus is not going to produce significant revenue. But they also realize that innovation, development, and support cost money. Each institution will need to weigh the benefits and the risks. We have a fairly succinct process at Purdue involving Office of Technology Commercialization approval if you want to take something open source. You explain how it is used, how you see it in a larger context, and which Creative Commons license you would use. The office examines all the factors and makes the best decision.

comments powered by Disqus