CENDI PRINCIPALS AND ALTERNATES MEETING
Defense Technical Information Center
Ft. Belvoir, VA
May 6, 2008
Final Minutes
Strategic Thinking and High Performance Organizations
Developing e-Science in Research Libraries
The Role of Human Capital in High Performance Organizations
DTIC Showcase
Welcome
Mr. Ryan, CENDI Chair, opened the meeting at 9:10 am. He mentioned several CENDI membership changes. He welcomed Tina Gheen, the new head librarian, as the alternate for NSF. Elliot Siegel will become the principal for NLM, with Betsy Humphreys as the alternate. Cindi Etkins, Sr. Program Planning Specialist at GPO, will replace Ric Davis as the principal. Roberta Schoen, Director of Operations, will be the new alternate for DTIC.
“Developing e-Science in Research Libraries” Neil Rambo
Visiting Program Officer for Library Support for Research and e-Science Association of Research Libraries (ARL) / University of Washington
e-Science is a new research methodology which is fueled by network capabilities and the vast amounts of data. Many areas that have traditionally been data poor, such as the environmental sciences, are becoming more data rich. Computational science has improved and vast amounts of data are being collected, particularly from sensors.
The question is, what do we do with the data and how do we make sense of it? The issue is more about data access. Visualization and data mining tools can help to make sense of the data once you find it. In addition, there is a need to share the data; capabilities are expected to be available to research teams that may be interdisciplinary and international in scope.
Science is a team effort, multi-institutional and multi-disciplinary. All of these characteristics of e-science present challenges for libraries because they are institutionally based in terms of collection development, security, access, reference, etc.
In addition, e-science is not a single model across the sciences. The model differs based on the discipline, the kind of data, who has access, who can understand the data, and the cultural norms associated with the data and its use and re-use.
ARL’s Joint Task Force on e-Science operated from 2006-2007. The final report was published at the end of 2007. The task force was charged to develop an understanding of e-science in the research library community, to come up with approaches for data curation; to develop roles for libraries; to identify skills that are needed by the library workforce; and to identify opportunities and strategies for developing relationships with other organizations that are needed to move forward. They realized early on that e-science is an area where libraries will need deep partnerships.
Data curation is an area where libraries may be involved. It is defined as “preservation + access = re-use”. Libraries know how to preserve things; however, data curation involves re-use and the possibility of re-purposing in the future. Data curation borrows characteristics from archival practice, including selection/retention, access, long-term preservation, what does the data mean and to whom, and intellectual property rights management.
Emerging and new forms of publication are challenging. Examples include publications that link to underlying data, data that readers can mine and manipulate to verify or answer different questions than originally posed, and hybrid publications that are part journal and part database. Issues include licensing practices, IT requirements, and access. In the broader sense, these new types of publication change scholarly communication into a process rather than an end product. Outputs are used throughout the cycle; the process is more fluid, less defined, and more complex. Libraries may find themselves involved earlier in the life cycle rather than just at the end.
It was noted that with the change in publications, publishers are critical to the discussions and eventual partnerships, but there is concern about what the publishers are willing to undertake. There are currently many different models for preservation, including societies, the publishers, and the national libraries. There is no consensus as to where the responsibility should be. It is likely to differ by community. How can research libraries play a role and when is it appropriate? Ms. Adler mentioned that the lack of a clear model is one of the reasons why institutional repositories are being developed.
Supporting virtual organizations is another challenge for libraries. NSF has made this a priority funding area. Content and tools are needed for collaborative work. In general, licensing doesn’t work for multi-institutional services or deal neatly with diverse environments. Authentication methods don’t work. However, digital libraries can provide models for the development and support of domain repositories that mimic some of these multi-service functions. A core set of functions may need to be identified.
Government information and funding agency data policies are also lagging behind. There are initiatives such as the NIH public access law; open access movement, which is moving from journals into open data; and the Science Commons licensing models, which are being watched closely. Many agencies have policies that data should be broadly accessible; however, the details remain to be worked out. Support needs to be provided to help local faculty make their data available and properly assign the right access limitations to it.
Libraries have strengths in many of these areas, including a deep understanding of and experience in related policies, such as open access; and technologies, such as institutional and domain repositories; and integration and interoperability of tools such as link resolvers, federated search and metadata standards. Libraries also understand archival practices and policies from business and technical perspectives, and in terms of defining resources of interest and developing reference collections for an entire domain. Libraries are accustomed to a tiered curation model. Libraries understand preservation, metadata, information life cycles, and the importance of assuring access and usability better than any other component in the research organizations.
In the process of analyzing the e-science environment, the Task Group looked to other national and regional models. Countries such as the United Kingdom, Australia and Japan, and the European Union are definitely ahead of the US in addressing these issues. Of particular note are the Joint Information Systems Committee and the Digital Curation Centre in the UK. They have the benefit of a national model, whereas the US is very fragmented. The task force tried to base its thinking on these models, where appropriate.
The task force recommendations call for development of a structure and process for moving the ARL agenda forward; establishing a knowledgeable community; identifying the need for a skilled workforce, including a process to help think through what a skilled workforce requires; and advancing appropriate policies.
The task force was disbanded, but there is a new group that will work with Mr. Rambo over the next year to advance the ARL e-science agenda. The first meeting will take place in June. “E-science and libraries” is clearly a priority area across the three ARL portfolios and ARL is also interested in how it can work with CENDI organizations on issues related to e-science.
Discussion
DTIC has recently been invited to participate in DoD discussions about large data sets.
NLM is beginning discussions with publishers about data preservation, but publishers have been slow to become involved. NLM has a project with Elsevier to address interactive journals.
Neal Kaske described the World Database for Ocean Information which has terabytes of data cataloged as collections. Videos are available with links to the data files. In this case, the library is part of one of the data centers.
DOE/OSTI has several large data centers. However, when OSTI recently did an analysis of where data is located, it identified 200 separate entities that collect and preserve numeric data within DOE. Some sites have multiple data sets. OSTI is cataloging these and listing them on the OSTI web site. The next activity will be to make the DOE data sets more searchable.
Science.gov provides some access to data centers and data sets via the web catalog. However, they are not well organized and there is no access across them. ARL indicated that it would like to have a single place to go. Perhaps a model would be the National Center for Educational Statistics, which provides access to data as well as tools.
The report from the Interagency Working Group on Digital Data is due out soon. The focus of the report is to cover all these issues and to move activities forward. The IWGDD would be a natural connection for ARL and will provide a way for agencies to work together.
“The Role of Human Capital in High Performance Organizations” Tom Ward, Principal Associate
Commonwealth Center for High Performance Organizations (HPO)
The role of human capital is to identify skill needs and to perform gap analysis between these needs and the current status of skills in an organization. Interest in human capital within the government began with David Walker while he was at the Government Accounting Office (GAO). His thinking about human capital being the most important thing was brought from Price/Waterhouse/Coopers. However, as Walker’s ideas began to be implemented through the Office of Management and Budget (OMB) and the Office of Personnel Management (OPM), they became diluted.
John Pickering of the Commonwealth Center for HPO developed the Center’s approach after following up with people who had participated in the Federal Executive Institute. The Center found that the participants were frustrated when trying to implement the management ideas they had learned. The Center began to drill down and found that only through cohorts and teams of intact groups could change be effected.
Through these investigations, the Center developed its seminar on Building HPOs. The seminar itself takes participants through hundreds of questions around six change levers (leadership, vision, values, strategies, structures, and systems) and seven key diagnostic questions: What is “High Performance” for us? How would we know if we were High Performance? According to whom are we High Performance? Why be High Performance in the first place? Are we doing the right “what”? How good are we at it? And, how are we treating each other and our customers? It is very difficult to internalize it all let alone present the depth of the seminar in this briefing.
Mr. Ward presented some key concepts and tools for assessing human capital. He distributed relevant handouts and offered to provide electronic copies of some of his resources to the Secretariat for distribution.
Organizational Leadership Philosophy is organized around several models. The craft model, which existed before the Industrial Revolution, was essentially a sole proprietorship. The owner did the design and execution and made all decisions. This is the essence of the leader’s role. Management involves getting supplies and resources. The task level is where the actual work gets done. With the Industrial Revolution, these levels were divided up. This is essentially the existing paradigm, including that in government, even though there are aspects of the craft model in all work.
Management theory often uses Maslow’s Needs Hierarchy. The basic needs are often referred to in current theory as hygiene factors. However, when hygiene factors become important, the situation has already deteriorated. Surveys of a broad segment of workers found that the best work days were associated with growth and having direct control over their work. While management must cover hygiene needs, people are always motivated by growth.
Rensis Likert’s Four Organizational Philosophies is a definitive study on organizational efficiency. It defines four systems, Exploitive Autocratic, Benevolent Autocratic, Consultative, and Participative, and plots them against characteristics such as employee motivation, teamwork, communication, decision making, goal setting, employee attitudes, and output results. Exploitative Autocratic views people as lazy, needing to be closely supervised, and motivated by fear. Benevolent Autocratic, which is very paternalistic, adds status as an employee motivator but organizational decisions are still made without consultation. Consultative believes in empowerment, and Participative is very egalitarian. The formal government structure is focused around System 2 (Benevolent Autocratic) thinking. This is the way personnel regulations were written and government is slow to move from System 2 toward System 3. System 4 doesn’t work well in government because it is generally too egalitarian and holds management accountable. It is important to understand the philosophy of a given organization. Marvin Weisbord has identified System 0, called Laissez Faire, which is a mixture of the other systems, and results in only poor to good output and, generally, mixed results on other characteristics.
KEYS, published in the Harvard Business Review, identified a 10 point scale. It is possible with this tool to benchmark your organization by type.
Buckingham speaks of competency alignment. He has developed an instrument to self diagnose strengths. The idea is to mentor a person’s strengths rather than focus on overcoming weaknesses. Mr. Ward also believes that emotional intelligence surveys can be helpful. Emotional intelligence can be developed and it has proven to be as much as six times more important 20 years out of school than someone’s intelligence quotient (IQ). Interpersonal effectiveness is at the center of the ability to work in teams.
When looking to the human capital needs for the future, it is necessary to look to 2020 instead of just five years out. At the Environmental Protection Agency in 1999, three hundred fifty people were interviewed through scenario building. Four scenario groups were established and the people were asked what competencies would be needed in the future. They then looked at the current skill set and developed a gap analysis. From this assessment, a workforce development program was created. It is important to involve all the stakeholders in the development of such a program. All stakeholders must be aware of the issues and have a clear desire to fix the gaps. It is necessary to start with knowledge and an understanding of the barrier points.
Technology obviously plays a part in human capital requirements. Mr. Ward recommended Fast Company, a magazine that describes Silicon Valley scenarios in each issue. Web 2.0 will have a big impact on human capital decisions in the future, since social networking tends to flatten the organization and to invite more participation.
A clear issue when doing workforce development in government is the government budgeting process. Another area to consider is the need to incentivize retirement.
A high performance organization can be viewed in many different ways. Who says the organization is high performing? A mission/niche focus would address how you interact with customers. A tactical focus would deal with operations and planning.
Customer Interface Task Force – Judy Berger
The Customer Interface Task Force was created to develop requirements for information delivery to defined audiences, to identify and implement solutions meeting these requirements, to present potential development to the Chief Information Officer (CIO) and the Change Control Board for approval, and to implement approved solutions. The Task Force has utilized staff from across the DTIC enterprise.
The initial input to the process came from customer surveys, focus groups, evaluation of usage statistics, usability reports, etc. The approach includes “embedded” DTIC Community Managers who help to identify the information of interest and value to the community. An initial set of communities was created around specific scientific and technical (S&T) program topics, such as sensors, weapons systems, and modeling and simulation. The role of the Community Manager is to gather input based on conference attendance, identification of key people in the community, assessment of information resources, etc., and then to identify content for community pages and develop an outreach plan for the targeted community.
DTIC Online was then designed based on the usability requirements, the content from the Community Managers, and content from the current DTIC pages and linked sites. The strategy is to develop the sites in three phases. The first will deal with the Public Site. The second will involve access to controlled sites in private STINET. The third involves the SIPRNET, with classified STINET and the Total Electronic Migration System (TEMS).
Ms. Berger showed more detail about the usability testing. A demonstration was given at the Special Libraries Association (SLA) annual meeting and a usability test was conducted at the DTIC Conference with 30 participants. She presented the demographics and the results. She also showed the two designs and discussed the comments and preferences expressed by the test participants.
The usability testing showed that users scan the pages first, followed by the use of the search feature and then the DTIC A-Z browse function. The major comments focused on the small font size, confusing drop-down menus, and an intimidating advanced search page. DTIC added a feature at the top of the page to change the font size. They are also rewriting all the text to allow for clicking to Read More. Text was removed from graphics to make it more readable.
The Change Control Board will meet soon to make the final design decisions. The new site will be available in late spring or early summer.
DTIC’s Information Sharing Operations (ISO) with Combat Librarians – Helen Sherman
DTIC’s first COCOM (Combatant Command) Conference, held in October 2007, was attended by Combatant Commanders from all commands. DTIC and the Information Analysis Centers (IACs) gave presentations. The COCOM Workshop increased communication with Combatant Commands, provided DTIC with an opportunity to learn COCOM requirements, increased the COCOM’s awareness of DTIC products and services, and provided opportunities for DTIC personnel to attend future COCOM meetings and exercises.
A Memorandum of Understanding (MOU) was signed in March 2008 with the Networks and Information Integration of the Information Policy and Integration Directorate. Funding was received to allow DTIC personnel to enhance the information sharing process by acting as combat librarians out in the field.
DTIC participated in the Pacific Command (PACOM) Exercise held in Hawaii earlier this year. Researchers prepared searches on Private and Classified STINETs in advance. The DTIC combat librarian participated in the Science & Technology Cell, consisting of scientists, researchers and commanders from the participating agencies, for the two-week exercise. He also presented training on DTIC resources. DTIC and the IACs provided 24/7 “reachback” support for the librarian.
The Northern Command (NORTHCOM) Exercise in Colorado focused on terrorism and homeland security. Participants were interested in older reports from the 1960s through the 1990s. The librarian created an annotated bibliography using DTIC, IAC and Web resources. Again, “reachback” support was provided. The combat librarian manned the NORTHCOM Analysis Desk, an expansion of an existing Interagency Coordination Analysis Desk jointly manned by subject matter experts and analysts.
DTIC has fine-tuned the role of the combat librarian. The librarian follows the daily battle rhythm of the exercise, attends the commander’s update briefings, receives tasks and assignments, refers queries for “reachback” support as needed, discusses research with the assigned group, prepares charts, power points and briefings in the required formats, and writes reports for DTIC on the Information Sharing Operations (ISO) activities. These exercises demonstrated the power of the DTIC databases as the librarian was able to successfully answer hundreds of questions. The “reachback” plan worked.
The approach has received very positive comments and it has resulted in requests for DTIC training. DTIC now has an MOU with PACOM for annual involvement.
This approach provided valuable insight into the warfighter culture for DTIC product development. Indecision and delay in the flow of information causes casualties. Warfighters want searchable terms that reflect their language. The research varies and reflects the mission of each COCOM. The warfighters want analysis and reports in their preferred formats. For example, Quad charts are required to present brief interpretation of information on new products and technologies, while power points are used to present briefings to leaders.
Based on what it has learned, DTIC is increasing its outreach in several ways, such as developing COCOM communities on the web, analyzing specific databases for COM needs, presenting briefings, exhibiting at conferences, sending email offering assistance, and promoting the DTIC librarian as an information broker. A big promotional effort will be undertaken at the Program Executive Office Conference in Huntsville in the fall of 2008 involving Program Executive Offices from DoD, Army, Air Force, Navy and the Marine Corp. It will be an opportunity to inform the customer base about DTIC and IAC services.
Ms. Sherman also suggested that CENDI members save April 6-8, 2009, the dates for the next DTIC Conference.
Internet Search Engines – The DTIC Experience – Helen Sherman
DTIC has provided access to its collection via an OAI-PMH file for several years. OAI facilitates data mining of DoD science and research information. It also provides worldwide access to DTIC’s publicly available resources.
Google drives the majority of DTIC’s referrals and full-text document downloads. More than 880,000 unlimited, unclassified citations were disseminated through Google, Yahoo and MSN, through OCLC’s WorldCat, universities and other federal agencies, and OAI partners such as OAIster and Science Commons. DTIC’s metadata can be produced in COSATI, Dublin Core, MARC and HTML.
Downloads have been increasing. There were 57.4 million citation downloads and 8.4 million full text downloads in calendar year 2007. They do not have specific agreements with search engine companies but they use site maps and OAI. This results in approximately 1.2 million harvests per week and just under 240,000 full-text downloads per week. The plan is to research other protocols as they evolve.
The morning program concluded at approximately 12:00 Noon.
