6
Technology Transfer Pathways for Digital Products
Preceding chapters have described policies, regulations, and laws that govern federal laboratories and their role in the development of digital products. The purpose of this chapter is to describe the pathways by which the research, inventions, and data produced by federal labs are commercialized and disseminated to the broader marketplace. The chapter also considers the importance of individual and organizational factors in advancing technology transfer.
Federal labs show considerable heterogeneity in their approaches to technology transfer. The labs have different missions and norms, and the research and technology development efforts across labs, even those within the same federal agency, are seldom coordinated. Moreover, different types of digital products are not always amenable to the same commercialization and dissemination pathways. Thus, the pathway for commercializing or disseminating a particular digital product often depends on both the lab and the type of product.
TECHNOLOGY TRANSFER OFFICES AT THE FEDERAL LABORATORIES
As discussed in Chapter 2, the Stevenson-Wydler Act, 15 U.S.C. § 3710(b), requires each federal laboratory to establish an internal technology transfer office (TTO, also known as an office of research and technology applications [ORTA]) and to support that office with “sufficient” funding. In addition, each lab having 200 or more full-time equivalent (FTE) scientific, engineering, and related technical positions must staff its TTO with at least one FTE position.
In 2017, the National Institute of Standards and Technology (NIST) surveyed federal TTOs to gather data on their budgets, staffing, and resources (Gingrich, 2018). NIST reported that these TTOs receive their funding through a variety of channels: for fiscal year 2016, the funding source for 14 percent of responding TTOs was a specific line item within an agency budget, for 26 percent was overhead, and for 25 percent was a superior office within an agency. Royalty
payments alone funded only a small number of responding lab TTOs (2 percent were funded by royalty payments alone), although one-sixth of the responding TTOs reported being funded through a combination of royalty payments and overhead or appropriated funds). TTOs with budgets of more than $5 million had an average of 33.67 FTEs dedicated to technology transfer functions, while those with a budget of less than $1 million had an average of 2.44 FTEs.
The committee also heard from representatives of a number of federal agencies and labs regarding their TTO operations and other technology transfer activities to gain a better understanding of the institutional frameworks and resources for technology transfer within the labs. The committee heard from representatives of four Department of Energy (DOE) labs, one National Science Foundation (NSF) lab, one National Aeronautics and Space Administration (NASA) lab, and one Department of Defense (DOD) lab whose annual TTO budgets ranged from $500,000 to $10 million, with staffing of between 1 and 42 FTEs. Although these labs do not represent the universe of federal labs, they do represent a wide range of internal technology transfer capacity, and they provided both qualitative and quantitative measures of their technology transfer functions and practices.
Three of the labs whose representatives spoke to the committee indicated that more than half of their TTO personnel had science/engineering/medical training, while the staff of two TTOs consisted of individuals with training in business. The staffs of the larger TTOs (19 or more FTEs) included legal personnel, who made up 26 percent of the largest TTO’s staff. The staffs of all the TTOs included administrative personnel, generally representing 15 percent or less of the TTO’s FTEs but reaching 27 percent in one case. One lab was a federally funded research and development center (FFRDC) administered by a university. This lab maintained its own small TTO (4 FTEs), but also relied on the university’s TTO (40 FTEs) for legal and administrative support.
Although the committee spoke only to representatives of a limited number of labs, it does appear that the TTOs’ reporting structures vary. In four of the seven labs, the TTO director reports to a senior administrative official at the lab, and in the other three, the TTO director reports to a senior research, science, or technology official. These observations may shed light on the perceived role of the TTO within a lab—whether it is viewed primarily as an administrative office or as part of the research enterprise. It is worth noting that, according to preliminary results of a qualitative study, scientists at federal labs are less likely than their counterparts at universities to “bypass” the TTO (Choi et al., 2020).
Decisions about the dissemination and commercialization of digital products are not always centralized within federal labs. While decisions regarding more explicitly legal forms of technology transfer (e.g., patents, license agreements, cooperative research and development agreements [CRADAs]) are made by a lab’s TTO personnel in consultation with its management, decisions about less formal dissemination of technology and knowledge (e.g., data release, scientific publications, open-source software) may be made by the lab’s research management, individual research units, or individual investigators. Most studies
have focused on formal technology transfer mechanisms, in part because of the greater ease of obtaining data on patenting and CRADAs relative to less formal mechanisms. However, all of these mechanisms are important means of transferring knowledge, know-how, and scientific expertise from the federal labs, and each is discussed in greater detail below.
Finally, although most federal labs operate independently with respect to technology transfer, some agencies have taken steps to coordinate activity among the labs they oversee. In 2007, for example, the national laboratory directors of the 17 DOE labs established the National Laboratory Directors’ Council (NLDC) to encourage collaboration and support on issues of common interest to the labs. In May 2015, the NLDC established a new Working Group of National Laboratory Technology Transfer Executives (NLTT) to advise the laboratory directors on issues and opportunities in technology transition, innovation, and commercialization. The NLTT serves as an interface with DOE Headquarters on department-wide issues and opportunities for improving the transition of technologies from the lab into commercial practice.1 In 2015, DOE established an Office of Technology Transition to coordinate the department’s technology transfer activities. The agency also established a Technology Transfer Working Group (TTWG) to improve technology transfer activities, enhance existing processes, and promote consistency of processes across DOE field elements and labs, as required in legislation (42 U.S.C. § 16391). Among other things, the TTWG has developed materials intended to assist labs with technology transfer activities.2 Despite these initiatives, however, no overarching organization oversees the technology commercialization activities of all of the federal labs or any subset of labs across agency lines.
TECHNOLOGY TRANSFER AND DISSEMINATION PATHWAYS
As with the following chapter on measures of technology transfer and commercialization, it is useful to distinguish among knowledge, invention, and innovation. Knowledge is an input into invention and is reflected in, for example, publications or data. Inventions are the novel tangible or virtual artifacts of knowledge and can be documented, for example, by patents. Inventions are an input into innovations, which are new products, processes, or services introduced to the market (i.e., commercialized). In this section, technology transfer and dissemination pathways for four types of federal laboratory outputs are described: (1) knowledge, (2) data and databases, (3) software, and (4) patentable inventions. An additional pathway—cooperative research arrangements—is also described. Data on the measurement of inputs to innovation (such as publications), inventions, and some of the other technology transfer pathways are described more fully in the next chapter.
___________________
1 Email from Richard Rankin, Director, Innovation and Partnerships Office, Lawrence Livermore National Laboratory and then President of the NLTT, dated December 11, 2019.
2 See https://www.energy.gov/technologytransitions/technology-transfer-working-group-ttwg.
Knowledge
Knowledge from the laboratories, including new discoveries, insights, inventions, and ways of doing things, may serve as inputs into the research and development (R&D) activities of firms that ultimately lead to new digital products introduced to the market. Such knowledge may move both through formal pathways, such as publications, and through such informal channels as conference presentations and professional networks, which several studies have shown to be a potentially important means of transferring knowledge from the lab to the private sector, especially in the software domain (Cohen and Lemley, 2001). Survey data have led some researchers to conclude that publications, conferences, and informal interactions are more important than licenses or cooperative ventures as channels for accessing research from government labs and universities (Cohen et al., 2002).
Researchers at federal labs regularly publish their work in scientific and technical journals. The 2013 Office of Science and Technology Policy (OSTP) memorandum “Promoting Access to Publications Arising from Federally Funded Research” mandates that federal agencies ensure that publications arising from their research activities be made available as broadly as possible on an open-access basis. Since 2008, the National Institutes of Health (NIH) has required that publications arising from all the research it funds, both extramural and intramural, be deposited in its PubMed Central repository within 12 months of publication.3 As of this writing, PubMed Central contains more than 1 million peer-reviewed articles that are publicly available at no charge. Other agencies, including DOE and DOD, are also implementing measures to make the results of their research publicly available. For example, DOE’s Public Access Gateway for Energy and Science (DOE PAGES) is a discovery tool that makes peer-reviewed scientific publications resulting from DOE research publicly accessible within 12 months of publication.4
Today, most open-access publication policies are self-executing, requiring little intervention from a lab’s TTO or management. In many fields, the scientific publishing industry has largely internalized federal open-access policies, and has adjusted publication agreements to accommodate both the lack of federal copyright in publications by federal employees and the need to make articles publicly available within a designated time period following publication (Contreras, 2013). In fact, in most cases, publishers themselves submit required preprint versions of the articles they will publish to federal open-access databases such as PubMed Central so as to retain control over the release process.
___________________
Data and Databases
Federal laboratories generate vast quantities of observational, experimental, and computational data in fields ranging from meteorology and oceanography, to radio astronomy and particle physics, to epidemiology and population health. Numerous federal policies require the public release and availability of federally created data, and many federal labs have released large quantities of nonclassified data to the public. In its information gathering, the committee heard from no labs that had ever licensed data on a commercial basis.
Some federal data release programs were in place long before current federal open-data policies were enacted. For example, NASA has made its earth science data fully open since 1994, sharing data from satellites and other instruments as soon as they become available. Likewise, since 1992, NIH and its individual institutes have adopted policies requiring the sharing and public release of data arising from genomic and other biomedical research. NIH has led the creation, maintenance, and growth of a genomic data commons that has become a key resource for the global biomedical research community, both public and private (Contreras and Knoppers, 2018).
Federal labs make data publicly available today through multiple online channels. In many cases, labs have created web-based portals through which data can be accessed, searched, downloaded, and used. A large amount of federal data is freely available at data.gov. In some cases, such as NIH’s Database of Genotypes and Phenotypes (dbGaP), data are more tightly controlled, and access requests must be approved to ensure that appropriate precautions are taken with respect to individually identifiable information.
That said, large datasets often require substantial maintenance, updating, quality control, annotation, and other associated services. Agencies such as NIH spend upwards of $100 million annually on the curation and hosting of their many datasets (Contreras, 2017; Contreras and Reichman, 2015). Other agencies spend less, and private actors wishing to utilize some publicly accessible government data may need to invest resources to make the data useful in particular commercial contexts.
Input provided to the committee by lab representatives suggests that decisions about the release and curation of federal data are generally not made by a lab’s TTO. Rather, those decisions, as well as decisions about the resources committed to putting data in usable form, appear to be made by scientific and technical staff with managerial responsibility for the respective projects. In addition, some larger labs, such as the National Renewable Energy Laboratory (NREL), have developed software applications to facilitate the release of datasets by research groups throughout the lab.5
___________________
5 See data.nrel.gov.
Software
A wide variety of software is developed at the federal laboratories, from highly specialized scientific programs and instrumentation systems; to data analysis algorithms and data sharing platforms; to programming languages and simulations; to artificial intelligence, statistical models, and machine learning tools; to more consumer-focused desktop and mobile applications. Given this diversity of software types, it is not surprising that the means by which individual lab-developed software programs are disseminated are highly situation dependent.
The principal decision regarding the dissemination of lab-developed software is whether it should be released on an open-source software (OSS) basis or licensed to one or more private-sector firms on a commercial basis.6 The federal government has adopted an open-source policy that encourages federal agencies to utilize OSS channels for the release of software, and large quantities of lab-developed software have accordingly been distributed under open-source licenses or contribution agreements. Yet concerns have been raised, as discussed in previous chapters, that the financial incentives accompanying exclusive rights may be necessary or desirable to promote the most effective commercialization of some software programs. Labs must thus decide which dissemination route to take with respect to any given software program.7 (See Box 6-1).
Input provided to the committee by representatives of some of the labs suggests that different labs take different approaches to the decision about whether to release software on an OSS or commercial basis. First, there appears to be no uniform decision maker responsible for this determination. While representatives of several labs stated that new software programs must be reported using a system operated by the lab TTO, the mode of software release can be determined by the individual software developer, the relevant research group or group leader, or a higher-ranking research or administrative official at the lab. Only one lab reported that a representative of the lab’s TTO is also involved in making this decision. In most cases, external project funders or sponsors, whether private or governmental, are also consulted regarding the means by which software is to be disseminated, and in some cases, they are largely responsible for this decision.
___________________
6 While the lack of federal copyright in software, discussed in Chapter 5, inhibits labs from commercially licensing software that is wholly developed by federal employees, large quantities of software are produced either by employees of government-owned, contractor-operated (GOCO) labs or by federal employees in conjunction with private-sector collaborators, both resulting in software copyrights that can be licensed commercially.
7 It is important to note that while some software-based inventions may be subject to patent applications and issued patents (see Chapters 4 and 7), most federal lab–developed software is disseminated via software licensing agreements (including OSS licenses).
Federal lab representatives who addressed the committee described seven different factors their labs consider when deciding how to release software developed at the lab:8
- Programmatic goals—What are the specific goals of the overall program under which the software was developed?
- Future development plans—Does the lab plan to develop and maintain the software in the future? If no further development is planned, it may be important to find a new team—whether an OSS community or commercial partner—to further develop and maintain the software.
- Purpose of the software—Is the software broadly applicable or limited to a specialized application? If broad usage by the public is envisioned, OSS release may be most appropriate; if use in specialized equipment produced by a particular vendor is envisioned, a commercial license may be most appropriate.
- Commercial market/application space—For software with limited or no commercial market, releasing it as open source can be most beneficial and achieve the widest use since there is no direct cost for acquisition.
- Third-party dependencies—Some federal lab–developed software may incorporate third-party code, and any release of that software must comply with contractual and other restrictions on that third-party code. Such restrictions may arise with both OSS code licensed under “copyleft” and similar licenses9 and proprietary software subject to the terms of a commercial license.
- Development environment—Software is developed at federal labs using both such common programming languages as C++, Python, and Java and more specialized development environments. The commercialization pathway for lab-developed software should take into account the overlap between the software’s development environment and the language preferences of the target industrial sector.
- Maturity of the code—Many software development projects conducted at federal labs are intended to produce only a proof of concept or prototype functionality. These projects are not intended to result in finished products that embody such design principles as security, resilience, and usability. Projects of this nature may require
___________________
8 Robert Leland, NREL presentation at open session of committee meeting on December 5, 2019; open session of committee meeting on January 30, 2020 with representatives of DOE’s TTWG; and presentations by Mary Monson and Robert Westervelt, Sandia National Laboratories, at open session of committee meeting on March 2, 2020.
9 For example, the GNU General Public License (GPL) requires that any software program constituting a derivative work of software licensed under the GPL must itself be released under the GPL.
- substantial additional investment to be transitioned to usable commercial products. Thus it may be useful to grant exclusive rights in such early-stage code to a private firm that is willing to expend those resources. On the other hand, it may be difficult to find a firm willing to invest the necessary resources to commercialize early-stage code, in which case releasing the code on an OSS basis may be the only practical dissemination mechanism.
Patentable Inventions
Unlike decisions concerning publications, data, and software, those concerning the filing of patent applications and the licensing of patentable inventions at the federal laboratories are handled almost exclusively by lab TTO personnel. All labs have a formalized process through which researchers disclose inventive concepts to the TTO using “invention disclosure” or “technical advance” forms. Officials at the TTO, sometimes in conjunction with legal counsel and science/engineering staff, evaluate these disclosures to determine which inventions merit patent protection. Representatives of the seven labs providing input to the committee reported that 45–90 percent of invention disclosures result in filed patent applications (although they did not differentiate digital products from other inventions).
As noted in Chapter 4, some federal agencies have developed policies regarding the patenting of lab developments. NIH, for example, prepared best practices for patenting and licensing genomic inventions (NIH, 2005).
Once a patent application has been filed, a lab’s TTO typically seeks out commercialization partners and licensees for the invention, although the lab’s technical personnel are often consulted. Inventions may be licensed on an exclusive or nonexclusive basis. The decision as to which of these licensing routes will be taken depends on a number of factors, including the invention’s commercial potential, the number of users that might benefit from it, and the investment required to convert it into a commercial product or application. Some agencies, such as NIH, have expressed preferences for nonexclusive licensing of inventions that have broad applicability or could be utilized as research tools, reserving exclusive licensing for inventions requiring significant investment before they are commercially viable (NIH, 2005). And as discussed in Chapter 4, certain statutory public interest requirements are imposed on government-owned, government-operated (GOGO) federal labs that wish to grant exclusive licenses for lab-owned patents. In Chapter 4, the committee recommends that these requirements be extended to government-owned, contractor-operated (GOCO) labs as well.
Recently, some federal agencies have experimented with new models for identifying potential partners to commercialize federal lab inventions. In 2014–2016, for example, NIH tried using a series of challenges to move early-stage NIH
technologies to the market.10 The winning entries received rights to the technology for a limited period of time.
In some cases, federal labs have determined that the greatest social benefit may arise from making patentable inventions available without charge or formal licensing. In the 1980s, for example, NIH patented the DNA sequence of the HEXA gene associated with Tay-Sachs disease, but chose not to enforce the patent against those who used it in diagnostic tests (Colaianni et al., 2010). More recently, Sandia National Laboratories and the Jet Propulsion Laboratory (JPL) have made certain patents freely available in the fight against COVID-19 pursuant to the Open COVID Pledge.11
Cooperative Research Arrangements
An additional important pathway for commercialization of federally developed digital products is the use of cooperative research arrangements, which include formal CRADAs, joint ventures, and other research or development arrangements between federal laboratories and both universities and industry partners. Indeed, researchers have suggested that cooperative research arrangements such as CRADAs are “the single most important channel” for private firms to acquire the underlying inventions12 that lead to commercialized innovations (Arora et al., 2016). As discussed in more detail in Chapter 7, in 2016 there were more than 11,600 active CRADAs across the federal lab system (NIST, 2019a). Under these arrangements, lab partners provide personnel, research funds, or in-kind contributions, while the lab provides facilities, equipment, or intellectual property. Information generated under CRADAs may be protected from public disclosure for up to 5 years, and the intellectual property that results from the joint research generally belongs to the private firm.
Researchers have shown that lab technologies are often far from commercialization, and thus require a considerable amount of codevelopment through cooperative research arrangements before being ready for the market (Ham and Mowery, 1995; Choudhry and Ponzio, 2020). There is also evidence that, relative to other mechanisms, CRADAs lead to higher levels of patenting for both the federal labs and their industrial lab partners, most likely because of the more intensive collaboration that occurs under CRADAs (Adams et al., 2003).13
___________________
10 These challenges include the Breast Cancer Startup Challenge, Nanotechnology Startup in Cancer, and Neurostartup run by the Center for Advancing Innovation in partnership with the National Cancer Institute at NIH. See, e.g., Neurostartupchallenge.org.
11 See opencovidplege.org.
12 Inventions are to be distinguished from the knowledge inputs into invention, for which, as noted above, the most pervasive pathways from public research institutions to industry are reported to be publications, public meetings, informal information exchange, and consulting (Cohen et al., 2002).
13 Similar to CRADAs, NASA’s agreements, called Space Act Agreements (SAAs), were authorized by Congress under the National Aeronautics and Space Act (51 U.S.C. § 20113[e]).
TECHNOLOGY TRANSFER/COMMERCIALIZATION ECOSYSTEM ENABLERS
R&D activities are part of a broader innovation ecosystem. Federal agencies have implemented numerous programs to promote entrepreneurship at the federal laboratories, to enhance public awareness of available lab expertise and technology, to transition technologies from lab to market, to build partnerships with local communities, and to promote economic growth.
For example, Sandia and Oak Ridge National Laboratory have established science and technology parks, and many DOD labs have partner intermediaries, such as Techlink, located at Montana State University, which serves as a partnership intermediary for technology transfer. DOE has established an internal I-Corps program that provides entrepreneurial services and training to researchers within the department’s federal labs. This program pairs researchers with industry mentors to foster a culture of market awareness within the lab and encourage lab employees to undertake entrepreneurial activities. DOE also recently initiated a Technology Commercialization Fund to leverage R&D funding in the applied energy programs to assist in the commercialization of promising energy technologies developed at DOE labs.
In addition, a number of agencies, including DOE, NIH, and DOD, offer postdoctoral fellowships to train, encourage, network, and mentor innovators through such programs as Cyclotron Road at Lawrence Berkeley National Laboratory.14 And the Department of Health and Human Services (HHS) has established an Entrepreneurs-in-Residence program to help identify, evaluate, and support the development of startups utilizing technology developed or funded by the agency.15
The intent and spirit of these programs are to be applauded, although the committee heard little empirical evidence regarding their success in enhancing technology transfer from the labs. Moreover, most of these programs were introduced relatively recently, so it is difficult to determine their success at this time.
INCENTIVES FOR SCIENTISTS AND ENGINEERS TO ENGAGE IN TECHNOLOGY TRANSFER AT FEDERAL LABORATORIES
The literature on university technology transfer identifies several individual and organizational factors that are also likely to influence technology transfer at federal laboratories: (1) individual financial and nonfinancial incentives; (2) social networks; (3) organizational structure, culture, and support for technology transfer; (4) relations between scientists and the TTO/administration and other aspects of organizational justice; (5) identity as a
___________________
14 Ilan Gur, Activate, Presentation to the Committee, December 6, 2019.
15 See https://www.hhs.gov/cto/initiatives/entrepreneurs-in-residence/index.html.
researcher or entrepreneur; (6) role conflict; (7) work–life balance; and (8) championing and leadership. The main focus below is on the first of these factors.
Financial incentives for researchers are reflected in a lab’s “royalty distribution formula,” which stipulates the fraction of revenue from a licensing transaction that is allocated to the researcher(s) who developed the licensed technology. Studies of university researchers suggest that universities allocating a higher percentage of royalty payments to faculty members garner greater licensing revenues (Lach and Schankerman, 2004; Link and Siegel, 2005). Another study found that attractive financial incentives for faculty help a university attract more productive, commercially oriented researchers (Jensen et al., 2003). And financial incentives and bonuses for TTO employees have been found to increase TTO licensing revenue (Belenzon and Schankerman, 2009).
In considering the role of financial incentives in the context of federal labs, it is of course important to be mindful of several institutional differences between the labs and universities. One example is wage differentials between university and federal lab scientists. Another important difference is that university scientists may be required to secure grants to support some or all of their salary, a requirement generally not faced by their federal lab counterparts. Still another is that federal rules regarding conflicts of interest may constrain both lab researchers and labs themselves from participating in the financial gains of their licensees and from forming startup companies based on licensed technology, as do many university researchers and universities. Finally, federal lab scientists have important roles and priorities other than commercialization, including work furthering the primary governmental missions of the individual labs and agencies, much of which holds no potential for commercialization.
In addition, intellectual property rights that facilitate commercialization can challenge scientific disclosure norms. As identified in the university context, university researchers traditionally have published and presented their scientific findings as soon as possible, in accordance with communal norms promoting the prompt and open sharing of data. Since passage of the Bayh-Dole Act, compliance with patent novelty rules generally has necessitated that university TTOs and restrictive terms in industry sponsorship agreements encourage or require researchers to delay publishing and presenting their work until a patent application has been filed, and sometimes even longer than that (Bagley, 2006). Thus, the unforgiving nature of the patent novelty rules may hamper early public disclosure and even dictate the pace, form, and scope of discourse and disclosure. These factors may explain in part recent findings that federal lab scientists are not highly incentivized to engage in technology transfer and may even experience cognitive dissonance in pursuing entrepreneurial activities (Choi et al., 2020).
Nonfinancial rewards may also be important to researchers. Some universities have started to adapt promotion and tenure and remuneration systems
for academics so that commercialization activities are valued.16 The desire to have social impact, peer recognition, and career advancement may be an important motivator for researchers as well (Cohen et al., 2020). Indeed, the committee heard evidence that such nonfinanical rewards may be important to researchers at federal labs, and that it is important for efforts to increase technology transfer and commercialization to take into account the relevant support needs, such as allowing conference attendance and participation.
Research on “star” university life science researchers and biotech startups shows that social networks may also play a role in technology transfer (Zucker and Darby, 2001; Powell and Owen-Smith, 1998). A recent study found that championing by department chairs/principal investigators (PIs)/center directors is a powerful enabler of technology transfer at both federal labs and universities (Choi et al., 2020).
Sandia was one of the first labs to establish a sabbatical program for lab employees, officially called the Entrepreneurial Separation to Transfer Technology program, to enable them to pursue business ideas in the community. In operation for more than 25 years, this program allows staff to leave the lab with a guaranteed job if they return within 2 years. Since 1994, the program has helped 162 employees—74 who started new companies and 88 who expanded companies—bring business ideas into their communities, primarily in New Mexico. Other federal labs offer similar sabbatical opportunities.
One important but understudied issue relating to the dissemination of federal lab technology is the role played by postdoctoral fellows and other junior members of the research team. Programs to support technology transfer as a career path for postdocs, such as the Technology Transfer Ambassadors Program at the National Cancer Institute, may help transfer technology out of the lab.17 However, a recent study found that postdocs at federal labs would like to engage in technology transfer but are not encouraged to do so by some PIs (Choi et al., 2020).
FINDINGS AND RECOMMENDATIONS
Finding 6-1: Although the approaches taken by federal laboratories to the dissemination and commercialization of digital publications and data are generally consistent, the approaches taken with respect to software vary across labs.
Finding 6-2: Studies of individual and organizational factors in university technology transfer have yielded insights on the importance
___________________
16 Based on a survey of North American institutions, Stevens and colleagues (2011) report that 16 major universities in the United States and Canada considered patents and commercialization in tenure and promotion decisions.
17https://techtransfer.cancer.gov/aboutttc/ambassadors#:~:text=The%20NCI%20Technology%20Transfer%20Ambassadors,development%2C%20commercialization%2C%20and%20entrepreneurship.
of incentives (both financial and nonfinancial), championing, and other managerial practices in stimulating technology commercialization and entrepreneurship, but there have been few such studies of managerial practices in the federal laboratory context.
Recommendation 6-1: The Federal Interagency Working Group on Technology Transfer should develop a set of written best practices for federal laboratories to use in determining dissemination pathways for lab-developed software.
Recommendation 6-2: An appropriate federal agency should conduct a study of the potential impact of different incentive and organizational factors on the motivation of federal laboratory researchers to engage in technology transfer and commercialization and the success of such efforts. Federal labs should use the results of this study when considering changes to their incentive structure and organizational practices.
