Interviewee: Don Walker Job: Technology Transfer Officer Organisation: Office for Technology and Trademark Licensing Interviewer: Thomas Haigh Date: Wednesday 21st September 1994 Location: Harvard University Revision: 1
The Office for Technology and Trademark Licensing (OTTL) is staffed by three technology transfer professionals, a working director, and a half-dozen support staff. Walker has been assigned the technical fields of software and certain applied sciences. One of his colleagues handles chemistry and molecular biology and yet another colleague handles the biological sciences. "We all do our [marketing] tasks quite differently-due simply to the difference in technology and the diverse nature of the markets being served." In the broad field of physics, for example, when reviewing a new report of an invention and thinking about how he would market the technology, no single "road map" comes to mind. By contrast, he thinks that a colleague evaluating a discovery in, say, the molecular pharmacology field, may well have in mind a "short list" of industry companies that can be approached with licensing in mind. Inventors report their technologies "pretty much on a voluntary basis" and sometimes the office needs to engage in "outreach" to make sure that faculty are aware of their services.
Another role for OTTL is the licensing of Harvard insignia for use on various products, including coffee mugs, folders and clothing. (Boston was flooded with Harvard t-shirts during my visit.) Revenues from trademark licensing can fluctuate sharply with some universities, often tracking the success or failure of the college's athletic teams. If a national manufacturers can be convinced to produce authorized university-emblazoned sportswear, sizable royalties-around $25 million for the University of Michigan-can be reaped by the institution. Harvard takes in about $2.5 million yearly, and whilst this is not significant in light of Harvard's investment income from a nearly $6 billion endowment, "the trademark [royalties] are important to us, among other things, because they are used to fund student scholarships."
The university's technology licensing and insignia licensing functions have been grouped into a single office because they both involve the negotiation of intellectual property (IP) agreements. The overall role of the office is to serve as a centre of expertise in IP for the university. (A noticeable exception to this rule is the Harvard Medical School, which has its own Office of Technology Licensing and Industry-Sponsored Research (OTL). Even so, OTL's licensing activities are reviewed and approved by OTTL.). The functions of this latter office parallels that of OTTL, bringing in substantial license fees and royalties. As the name implies, this office handles sponsored research and technology licensing in an integrated fashion-not surprisingly, the world-class Medical School is responsible for a great many of Harvard's total invention disclosures each year.
Harvard does not have a central Industrial Liaison Office (ILO) programme such as that found at the Massachusetts Institute of Technology (MIT). Walker's opinion is that this is so because Harvard has not stressed the engineering sciences (the base of MIT's ILO constituency) for a number of years.
This university also has an Office for Sponsored Research (OSR) which handles all incoming grants-whether from foundations, government agencies or industry. They work closely with OTTL, which must review and approve the intellectual property terms in proposed grants or research agreements with industry. OTTL will, on request from OSR, negotiate with industry on the IP terms in these agreements.
Harvard's Development Office-tasked with the responsibility of seeking donations from alumni and other benefactors-is located in the same building, and the same storey, as OTTL. Yet contact between the two offices on marketing matters is rare. This may not be surprising since a supposed key marketing tool of the two entities (respective computerized data bases of companies, individual contacts, etc.) may be incompatible for each other's purposes.
Walker believes that most aspects of technology transfer, as currently practised at Harvard, are similar those employed by other institutions, "although each university believes there are better ways to deal with the growing administrative burden attending technology transfer." One favorite, and apparently common, "solution" is the design of the computerized database mentioned earlier, one aspect of which would be the linking of potential licensees to newly reported licensable technology. This system, in addition, would also facilitate the necessary government reporting, patent expense tracking and royalty disbursement chores. Stanford University has developed a similar program. However, its effectiveness-if employed by other universities-might not be optimal.
One area where policy differs between institutions is the formation of start-up (often called spin-off) companies where university educators and researchers may have a close financial interest in the new firm. What is Harvard's current policy and experience there? There has been moderate activity with respect to helping launch start-ups, but Harvard's experience is certainly not as great as that of other universities, and nowhere close to MIT's. "Financial risk issues aside, there's always the spectre of a real or perceived conflict of interest (COI) issue which arises in start-ups, where the involved faculty member's ongoing university research is the sole basis of the company's existence." Walker believes the best way to minimise the potential for COI is to market the new technology to a firm or venture capital group in which the inventor is a disinterested financial party. Failing that, one might license the invention to a firm in which the university researcher's financial interest is "not substantial." Just what this latter term means is subject to a great range of interpretations. There are, of course, problems whatever course is taken. If a technology is successfully licensed to an existing company but the inventor is not prepared to champion the technology and work closely with the company, either for personal or professional reasons, then the discovery may not flourish. "My co-workers in this office, and the tech transfer people at the Medical School, are better qualified to comment on this COI-management question. Thankfully, in my stay here, I haven't had to deal much with it."
While OTTL does its part to monitor university conflict of interest situations involving technology licensing, the best watchdogs, believes Walker, are the various senior faculty and peer committees which review and act upon questionable professional faculty conduct. Walker feels that such bodies "often don't have enough information to consider a case until the fire is well under way."
Faculty consulting is another area where COI situations could arise, but OTTL is "not really in the picture on such matters. The faculty are expected to be self-monitoring. If asked, we can provide a sample consulting agreement." When a formal technology transfer agreement has been concluded, "it's almost inevitable that the inventor will become a consultant to a licensee-and that's fine as long as the independence of their academic research is not compromised." Faculty are not allowed to use university facilities or time for their consultancy, or to steer the academic pursuits of their students towards the objectives of the industrial licensee.
At Harvard, software is licensed strictly "as is" with no support, maintenance or bug-fixes. It is quite likely that a company will seek consultancy from its author at some point.
One Harvard software program, out of many created at the university, has been granted a U.S. patent.
"Seeking patent protection for [scientific] software is rare at Harvard. I may have influenced that policy: I'm convinced that, with few exceptions, copyright protection is both timely and adequate, given the dynamics of the market. The sole patent I mentioned before has yet to be licensed. I'm sure that's a coincidence, but the fact isn't lost on me when I consider how to quickly make the technology available to the public."
The exclusion of support and maintenance from software licenses is one area in which Walker "will not yield." Improvements will be considered if they fall naturally within the direction of on-going research in the area and the research interests of those involved, but "we cannot be a development house." (At the time of the interview, Walker described a situation wherein a licensee was arguing to reduce royalty payments on the grounds that support from Harvard had not been forthcoming. Aside from the fact that the language of the agreement clearly stipulated the "as is" condition of the program, Walker's counterpoint was that the fee and royalties the licensee had paid were perhaps only a tenth of what it would have cost them to develop the software on their own.)
The legal status of software which is derived from licensed material is problematic-the U.S. copyright laws suggest that the right to create derivatives remain with the original copyright holder and would have to be specifically granted to the licensee if they are to own the results of its commercialisation.
"That's a very hard issue to resolve... I am still looking for ways to measure the relative intellectual input to derivative works... simple examination of the code is clearly inadequate.... It's hard to know who "owns" what portion of a program after it's gone through multiple revisions."
Walker describes the threatened erosion of the U.S. Bayh-Dole Act as an ongoing concern of his office and of its director, Joyce Brinton. She is president-elect (1995) of the professional association which serves university technology transfer professionals, serves on several other national-level collegiate committees, and has helped draft material for presentation to congressional staffs. Brinton is "quite vocal" in discussing with government the new programmes for technology transfer which
"curiously, leave the intellectual property ownership with the industrial partner rather than the university. That is difficult for us to grasp because the rules with which the university operates are different from those of industry, particularly the issue of confidentiality. Universities need, and must reserve, the right to publish freely."
Another current issue for Harvard is "the University's acceptance of equity in licensee companies in lieu of (or in addition to) license fees and royalties .... In the past, the conservative nature of Harvard led to policies which avoided that kind of involvement for fear of compromising its academic principles for financial gain." Derek Bok, Harvard's former president, felt this very strongly-but under new university leadership "we're moving towards a point where greater university ownership is possible" with regard to start-up licensees.
Inter-university agreements can be another problematic area. "Frequently, institutional partners which jointly own a piece of intellectual property cannot agree among themselves how best to license the technology, pay the patent prosecution costs, and share royalties. And, with the movement of faculty inventors from one institution to another, the ownership and royalty sharing rules become quite complex." Policies laid down by different federal agencies or foundations which have funded the research can also add to the mess. Differentiation of the ownership of software derivative works, according to Walker, can be particularly vexing.
As previously mentioned, sponsored research agreements are handled by a separate office, OSR, although they work closely with OTTL on intellectual property clauses. Often, such agreements include some kind of priority access or option to license the results of the research; few companies are willing to make outright grants without some condition of having a look at the researcher's results, "but there are no guarantees ... that by funding the research they automatically become licensees." There are some companies that hope to, right from the beginning,
"establish licensing terms as conditions which are buried in the grant document; [Harvard] won't participate in that ... it is inappropriate to lock in and make captive the research activity of a university laboratory. The faculty could then become indentured researchers for the companies supplying the money ... the influence can become quite pervasive."
On a project lasting several years a company which saw work moving in a direction unlikely to benefit them commercially might seek to redirect it-occasionally by withholding money-but usually in more subtle ways.
What the company would get, in addition to a right of first option, would probably be a chance to read papers produced from the research before their competitors-though not before they are sent for publication.
"Many companies, in their proposals for industrial research, will require the researcher to submit the paper to the sponsor before they submit it for publication. That, of course, is not acceptable-not in my cases. Maybe others in the University would find that acceptable, but I don't."
One of the other benefits which companies see from such an agreement is the increased prestige gained from "affiliation with the university and with a particular researcher." Given the high profile of many Harvard faculty, this can be a significant incentive. Interestingly, the Japanese "take a long term view of it: they will often impress you with their long-range plans, as well as the size of their pocket books."
Academic networking is a valuable source of contacts for potential agreements, and Harvard makes its alumni "part of the family." "But, I haven't found having an alumni champion inside the company for a technology license of very significant importance. It could be that the CEO is, in fact, an alumni and is more approachable, but business decisions are made on very rational grounds." However, industry can tend to be more tuned to the "cult of personality" and therefore, a world-renowned principal investigator is a major asset when seeking research funding. Even so, a degree of "vision" is still required on the industrial side.
Firms are prohibited from capitalising on the Harvard name when promoting products based on licensed technology. For example, Harvard reserves the right to inspect material mentioning Harvard which is to be used in a national magazine. On the other hand, due recognition of their contribution is also expected-in the case of software this might be an acknowledgment message on the title screen.
Harvard benefits from the cluster of high technology companies in the Boston areas attracted by, or spun-off from the many universities, particularly MIT. "Within a radius of twenty or thirty miles of this campus there are literally hundreds of firms that owe their existence to the universities in this area .... We have a huge talent pool of consultants available." Although MIT's departments, as a whole, can tend to overshadow Harvard's in many applied science areas, not least in terms of size, many Harvard faculty in these areas are very strong. This can mean more of a concentration in particular areas. Harvard's policy is to bring in the best available faculty, although this may mean that many bright, junior Harvard faculty may have to look elsewhere for tenure.
Walker agrees that many areas of computing are inherently applied, and can benefit from industrial contacts.
"You need a lot of feedback. In licensing you try and find the industry leader, because you feel that their direction is the most `real world' as well as most likely to result in revenues which lead to high royalties .... One of the benefits of licensing is that the faculty and the graduate students working on it get to go beyond the theoretical-they never loose sight of what it is that they are doing, they are doing basic research for educational purposes and they're going to write their theses about it-but all the more valuable that the thesis will be widely read and not just by academics, but by those who are in the field. It certainly doesn't hurt in applying for work to have on your resume that you have written something which is fairly widely accepted in the industry in which you are trying to find a job."
How important are the revenues raised from licensing? Walker believes the main purpose of technology transfer at Harvard is to disseminate knowledge and ensure that university discoveries are promptly placed into public use.
"That's a very real, as well as idealistic, goal for this office. Although nobody [in the university's central administration] looks at our small group with a totally 'green eyeshade' viewpoint, they have been known to ask us to project what our royalty revenues will be next year.... It's my observation that universities themselves quietly note their relative technology transfer 'rank' among peer institutions. It's natural to compare which school had the most [invention disclosures], or had the highest royalty income-to-staffing ratios, and so forth. However, technology transfer professionals readily admit such comparisons are specious: The broad licensing of a single `hot technology' can make even the most inept office look productive. But, as a group, we need the 'comfort' of quantitative reckoning."
Will the reorganisations of large corporations such as IBM and DEC alter the technology transfer landscape?
"I think that most of the best work done in applied computer science today has not been done in universities at all-it's being done by the large commercial organisations with well staffed research groups [such as] IBM, Digital, and Silicon Graphics .... The reason they've been able to do this is that their profits were, or are, large enough to support this. Furthermore, they are driven by competitiveness and anticipation of market changes. Universities seeking knowledge don't have the 'commercial hunger' seen in for-profit companies."
Walker has seen "the pendulum swing three or four cycles over the last twenty-five years" with regard to industrial attitudes towards research. Policies change and new approaches are tried. At one time aerospace companies such as Raytheon Company and TRW used to sharpen the wits of its research people by posing the same design problem to two groups, one significantly larger (and better funded) than the other, to gain the benefit of different approaches. (It is reported that Apple Computer has used this same internal competitive style.)
One trend on the university side has been towards the re-labeling and restructuring of group activities into "centres" and "institutes." For example, Walker mentioned Harvard's Center for Textile and Apparel Research. These initiatives represent a re-focusing of university talent to allow its application to national economic needs-while steadfastly maintaining the university's academic research goals.
How does Walker see the situation developing in the future? First, reduced federal funding of research at institutions of higher education forces universities to consider alternate sources of revenues. Walker believes that economic necessity may cause some universities (but, certainly not Harvard) to re-examine their "freedom to publish" policies with the possibility of imposing limited confidentiality on results of industrially-sponsoredresearch.
Second, there is a real possibility that, under certain circumstances, universities may loose permission to take title to certain future federally-funded research discoveries. Walker concludes,
"In recent years the creation of a new `advanced technology program' by federal agencies effectively counter key provisions of the [Bayh-Dole Act]. There is also another 'technology reinvestment program' which, while not precluding university ownership of intellectual property, nonetheless places control of the research in the hands of the for-profit partner, and places the university in a disadvantageous negotiating position. And, there is draft legislation being considered by the U.S. congress that would create a government-operated corporation to centralise and control all federally-related technology transfer-including, perhaps, university inventions."