Back to May 2002 CRN Table of Contents

[Published originally in the May 2002 edition of Computing Research News, p. 7.]

By James J. Hickman

One of the largest hurdles a researcher faces in starting a new field is the creation of a source of funding in amounts large enough to allow a critical mass of researchers to push the field at a rapid pace. These groups tend to be a small community at first, one that requires input from new scientists in the primary area and, especially in emerging multidisciplinary fields, scientists from other fields who can be attracted by targeted investment.

One of the best ways to push new ideas at a funding agency is to bring in one of the leaders in this field to set up a new program(s). However, active researchers often cannot devote two years or so on a full-time basis because it would set their careers back a significant amount. It is also very difficult to maintain research programs on a part-time basis with a full-time IPA (Intergovernmental Personnel Action--an easy way to go from a nonprofit institution to work for the government for a limited time). I would like to suggest to researchers in emerging areas that an expert in a particular field can make a very large difference, even if contributing as little as one day a week, in an official capacity at the National Science Foundation.

For the past 18 months, I have been working in a part-time capacity at the National Science Foundation in the CISE directorate. I am writing this article from the perspective of an expert in the new field of biocomputation who spends one day a week at NSF to help set up a program in this area. By 'biological computation' I mean the process of studying how biological systems process and store information to serve as models or components for the next generation of algorithms, software, and hardware in computer and information science. I think this will be a critical area where advances in understanding will be the foundation on which next-generation artificial intelligence systems can be built.

When I started this IPA at NSF in June 2000, I was also setting up my labs in the Department of Bioengineering at Clemson University, a position I had accepted in May 2000. Juggling these two responsibilities for the past 18 months has been an interesting task. It raises an important point one needs to think about when considering a part-time IPA, especially if you are not local to Washington, but distance is certainly not an insurmountable obstacle.

For me, the primary driving force for helping to start this program at NSF, besides the basic intellectual challenge, was the scarcity of programs dedicated to biocomputation research at any government agency. There have been attempts to start biological computation programs over the past 10 years, such as the ill-fated Ultra-Scale Computing Program at DARPA and some small efforts such as the DOE's engineering focus within basic energy sciences. But there have been few large-scale, dedicated efforts. I felt it was necessary for someone to step up and try to get this area going, especially from a basic-research standpoint. Creating this program at NSF in CISE seemed like the perfect opportunity to address all of these issues, and I thought this was the right time and place to attempt this.

I was brought in to the Experimental and Investigative Activities Division of CISE by Rick Adrion and Ruzena Bajcsy because they saw biocomputation as an area CISE needed to develop from a computer science standpoint, not just a biological perspective. It was great working with visionaries like Ruzena who are not afraid to undertake non-traditional efforts and to continue to support them in the face of resistance.

To begin this program we first ran two workshops. One was a joint programmatic effort by Frederica Darema, Rick Adrion, and myself held at NSF in September 2000, which was highly successful. We brought in a small group of leading researchers and proponents of this area for a one-day workshop. These individuals represented a breadth of experience, from neuroscience to computer science as well as biomedical and electrical engineering and chemistry. This workshop laid the groundwork for the development of the program, and the report can be found on the EIA website.

A second workshop was held over a two-day period in January 2001 at Clemson, South Carolina to further explore the latest developments in biocomputation. More than 75 researchers with a wide variety of interests attended this event, including some international participants. Most importantly, approximately 15 attendees were from federal agencies such as NSF, NIH, DARPA, and DOE. This workshop report will be posted soon on the EIA website.

The program announcement entitled "Biological Information Technology," or BITS, was released in March 2001, and more than 40 applicants submitted proposals. Because not all of the awards have been finalized, I am unable to provide details of the types of submissions or numbers of awards. I can say, however, that a wide range of projects, from hardcore biology to hardcore computer science, have been supported, and everyone agrees the award process has been highly successful. BITS will be an ongoing program in EIA/CISE, with an annual due date of the first week in February. Details of the program can be found on the EIA website or on the general NSF program announcement website.

If you are considering a position as a part-time program manager or consultant in a government agency like NSF, you should be aware that you will probably be paired with, or be expected to work with, a full-time program manager in that division. Make sure before you accept the position that this is a person you can work with! If not, this could turn out to be a very difficult arrangement and much harder and less rewarding than you would hope for. You have to realize that you are low person on the totem pole, and it is really important to have a good working relationship with agency staff.

Another point that a part-time IPA needs to be aware of is that someone coming into NSF with this arrangement must be prepared, if successful, for others to be interested in your program, both from a positive and negative standpoint. In my experience, most people have been positive, supportive, and helpful. It is necessary to work for accommodation when negative or difficult situations arise; it may also be necessary to accept the fact that in certain cases this may be impossible. Like all highly visible situations, programs in new areas can attract very competitive individuals. Some will view your success as having a negative effect, either on their existing programs or on programs they may want to start. I saw some of this, but I believe that once the program got started and was proving successful we were able to navigate the occasional rough waters.

In my case, I believe the two workshops and the program announcement had sufficiently raised the awareness of biological computation in other areas of NSF, and in other federal agencies, to inspire similar efforts or complementary programs. This was an added benefit to my efforts to provide resources to research in this area.

Another difficulty you may encounter is that often NSF staff and full-time IPAs who come there are not at the cutting edge of a new field. Be prepared to be called self-serving if you have the occasion to talk about or present your own work. Presenting one's own work normally does not happen at NSF, but it may be necessary if you find that your perspective on a given topic is unique.

There are many positive aspects from my tour at NSF that anyone doing a similar service could also expect. I have met a number of wonderful people, both inside and outside the NSF. I also have a much better idea of how the system works and how one can make a large positive difference even in a limited capacity if the situation is right. There are certainly some things that were more difficult than they had to be, but all in all I am happy a successful program was created in an area that desperately needed support.

James J. Hickman is the scientific advisor to the EIA director in CISE on biocomputation, as well as the Hunter Chair of Biomaterials in the Department of Bioengineering at Clemson University. He is also an adjunct associate professor of Chemistry at the George Washington University.