A QUESTION-DRIVEN STUDY OF CONTEMPORARY OR/MS PRACTICE

October 20, 1995
Arthur Geoffrion


TO:   INFORMS Board
CC:   E. Duncan, M. DeMelim, G. Freestone, T. Spencer
FROM: A. Geoffrion for the Roundtable
SUBJ: Joint Board-Roundtable Meeting to Discuss 
      "A Question-Driven Study of Contemporary OR/MS Practice"
DATE: October 20, 1995


   What follows is 13 pages of notes describing the initiative to
be discussed at the joint meeting of the Board and the Roundtable
in New Orleans on Monday afternoon: "A Question-Driven Study of
Contemporary OR/MS Practice".

   This was a team effort over the last three months by many
people, including an ad hoc Pre-Steering Committee comprising Tom
Baker, Burnell Brown, Arthur Geoffrion, Sid Hess, Mark Lembersky,
Randy Robinson, Leon Schwartz, and Tom Spencer.  Now in its
fourth draft, earlier drafts have been commented upon by the
Roundtable Executive Committee, Al Blumstein, John Birge, Robert
Fildes, George Freestone, Terry Harrison, John Llewellyn, John
Ranyard, and Mike Rothkopf.  All deserve thanks for their
contributions.

   These notes, which are being distributed today to all
Roundtable members as well, have been changed in numerous but
mostly evolutionary ways from the third draft of October 15 which
a few of you received.  Perhaps the most significant change, in
response to advice from those who just completed a major study of
OR practice sponsored by the Operational Research Society,
relates to funding.

   You will be asked at the meeting to help improve the Draft
Project Plan Guidelines below, to authorize and fund the project,
and to appoint an official Steering Committee.

   Just one more thing: the project as proposed may be overly
ambitious for the human resources realistically available to
coordinate it.  Please think about how to cut it down to more
manageable proportions (e.g., by dropping many of the key
questions), and whether a pilot study is needed.

   Thank you very much for your thoughtful consideration.


There are different interpretations of the current status of OR/MS when one looks at what is happening to industrial groups, to academic groups and their curricular role, and to INFORMS membership trends. There are causes for alarm and causes for optimism. However one interprets these trends and developments, there should be general agreement that the key to improving the future of OR/MS in general is to improve the prosperity of OR/MS in practice. How to improve the prosperity of OR/MS in practice? By seriously studying and trying to understand the contemporary practice scene much better than we do, and then by using this understanding to strengthen how OR/MS is practiced, to strengthen INFORMS' services for practitioners, and to strengthen the academic programs that train future practitioners and those who will develop applicable theory and methods. The Roundtable proposes to undertake, in partnership with the INFORMS Board, a major project that aims to travel down the road just indicated. This initiative will be the main topic for the Board/Roundtable meeting scheduled for Monday afternoon in New Orleans.
10/20 DRAFT PROJECT PLAN GUIDELINES PROJECT TITLE "A Question-Driven Study of Contemporary OR/MS Practice" PURPOSE To study and better understand the present state and major trends of OR/MS practice, including employment, and therefrom to obtain facts and insights of immediate value to: (a) practitioners as they strive to survive and succeed in the face of rapid change and organizational challenges, (b) the INFORMS Board as it strives to position INFORMS to attract many more practitioner members, and (c) educators as they strive to attract and properly train tomorrow's practitioners, and to update today's practitioners. Do this in a way that is driven by carefully chosen key questions. Do this with a carefully "segmented" view of the practice landscape that recognizes its lack of homogeneity, comprising as it does diverse kinds of practitioners. Do this comprehensively, but at the lowest possible cost through an electronically coordinated network of project volunteers. 5-PHASE PROJECT STRUCTURE 1. Compose a list of key practice-focused questions, good answers to which would strategically and tactically inform practitioners, the INFORMS Board, and educators. Design a segmented view of the world of practice. 2. Marshal already existing data and knowledge as it pertains to the key questions. 3. Broadly collect data and views on the key questions. Emphasize interviews and open discussions with practitioners and their clients over traditional surveys, taking care to preserve anonymity when necessary for candid response. Summarization and synthesis activities should not lose any primary data. 4. Place most primary data and all project results in an on-line knowledge base for INFORMS members to access and analyze as they wish. Moreover, make it easy for practitioners, at their own initiative while on-line, to add further contributions and comments to this knowledge base. Practice Online and INFORMS Online will be the primary sites. On-line materials should begin appearing by early in 1996. 5. Distill and organize what is learned from #2 and #3 for action-oriented consumption by the INFORMS Board, practitioners, and educators. KEY QUESTIONS Most key questions are phrased in terms of "OR/MS practice segments". This notion, which seems prerequisite to sufficiently discriminating conclusions, will be explained shortly. The litmus test for each key question is that it should be of great interest to one or more of the INFORMS Board, practitioners, and those who educate practitioners. The final list of questions will drive the entire study, and probably should be many fewer than the possibilities listed here. Employment What is the recent history of OR/MS practitioner employment (incl. job openings) during the last 5 years in each OR/MS practice segment? Compile data from cooperating organizations, including information on career ladders. What major employment-influencing dynamics and trends have been at work during the last decade? (To see dynamics and trends clearly, it is best to look at a fairly long time period.) BLS Clarification Are the Bureau of Labor Statistics estimates for OR/MS employment valid? What adjustments are advisable? With approximate adjustments applied, what employment picture emerges for OR/MS and what is the "market share" of INFORMS' membership? Career Paths What are the typical career paths of practitioners in each OR/MS practice segment? Compile individual histories, including their perceptions of de facto career ladders (as distinguished from official ones). Do not overlook people who leave OR/MS. OR/MS Opportunities What have been the greatest opportunities for each OR/MS practice segment during the last decade, and why? Compile examples of where these opportunities have been exploited successfully. OR/MS Threats What have been the greatest threats to each OR/MS practice segment during the last decade, and why? Compile examples of where these threats have caused damage. Exemplars What are the most successful groups and individuals in each OR/MS practice segment? Compile interview materials aimed at explaining why, and look for the main segment-specific success factors, if any, beyond the ones commonly cited. ("Success" might be measured by rate of growth, documented benefit/cost ratios, and/or degree of self-determination.) Obituaries What OR/MS groups were disbanded or decentralized in each OR/MS practice segment? Who was fired? What OR/MS projects failed? Compile interview materials aimed at explaining why, and look for the main segment-specific risk factors responsible for the unfortunate outcomes. Awareness How aware of the potential contributions of OR/MS are prospective sponsors and users in each of the practice segments? What measures can improve awareness, and what are their rough cost/benefit ratios? Compile hard data, opinions, and evidence concerning the effectiveness of past awareness-building measures. IT's Role What role does the rapidly developing area of Information Technology play in each OR/MS practice segment? How is it changing the conduct of OR/MS? How have OR/MS groups fared when associated organizationally with IS or some other IT-related group? What are the main opportunities for INFORMS to cooperate with SIM and ACM, and would this help? Compile specific examples and views. Internet's Role What value has the Internet yielded organizationally and to OR/MS professionals in each OR/MS practice segment? Compile specific examples. What are the most promising areas for expanded Internet use, and why? Non-INFORMS Practitioners Who are the non-INFORMS practitioners in each OR/MS practice segment? Compile their Email, fax, and postal addresses, and job descriptions. Usefulness of INFORMS Services For each type of service that INFORMS currently provides, how truly valuable is it to each OR/MS practice segment, and why? (Structure this question around a simple taxonomy of INFORMS services/ activities.) How could the value of each service be improved? Compile individual views. Desired INFORMS Services What new INFORMS services would be of greatest value to each OR/MS practice segment, and why? (Partially structure this question around a simple but open-ended taxonomy of possible new services, including electronic services.) Compile individual views. Choosing a Professional Association How do practitioners in each OR/MS practice segment decide which professional association to join? How could INFORMS compete more successfully in each OR/MS practice segment? Compile individual views and specific choices. Choosing Meetings How do practitioners in each OR/MS practice segment decide which professional meetings to attend? (Well-run meetings can produce large profits.) How could INFORMS meetings compete more successfully in each OR/MS practice segment? Compile individual views and historical choices. Educating Future Practitioners Undergraduate, masters, and Ph.D. education in OR/MS: which degrees, kinds of courses, and pedagogical styles are most useful for each OR/MS practice segment, and why? Compile individual experiences and views. (Take pains to make this information available to students.) Updating Current Practitioners What are the greatest continuing education and seminar needs for each OR/MS practice segment, and why? (Don't overlook topics that fall outside OR/MS and its supporting disciplines, like communication and organizational intervention skills.) Compile individual views. SEGMENTATION The OR/MS practitioner community needs to be "segmented" throughout the project, starting with data gathering, because the answers to many key questions may vary substantially across segments. A segment is defined for present purposes as a discrete-valued attribute associated with a practitioner or client. N such attributes would create a N-dimensional segmentation. One obviously important dimension is to distinguish between practitioners who are part of an organized group with OR/MS responsibility and those who are not (so-called isolated or dispersed practitioners). This dichotomy could be refined to distinguish those who practice OR/MS part-time (their primary job being teaching or some other non-OR/MS responsibility) and those who practice OR/MS as their primary job. Another important dimension is economic sector. The partition could be rough -- e.g., consulting services, non-consulting services, goods-producing, government/public sector, and military -- or somewhat finer. In addition, the population could be segmented by functional responsibility, by favored OR/MS technique, or by "strand of practice" in the sense of Corbett et al. (1995). Information providers should be classified as to whether they are currently an INFORMS member, a lapsed former member of ORSA or TIMS, or never a member of one of these associations. This should permit discerning any important differences between members and non-members, and is pertinent to several key questions. Whatever the segmentation design turns out to be, time will automatically be an additional dimension at the yearly level of granularity. This attribute, to be attached to all observations (not the year of recording, but rather the year to which the observation pertains), will permit summaries and analyses in which time is significant. The segmentation design should take into consideration which particular segmentation dimensions are likely to be pertinent to each key question. MULTIPLE DISPERSED PROJECT TEAMS The Network Age makes it possible, through the Internet and other networks, for even a widely dispersed team to work effectively toward its objective. It also provides the enabling technologies for coordinating multiple subteams. The project will exploit these possibilities extensively, relying on a Steering Committee for overall coordination of multiple Project Teams. A team can be as small as one person, and will be expected to function nearly autonomously once accepted into the project. Each Project Team will focus on some subset of the Segment x Key Question matrix. This matrix has many cells. Some Project Teams may elect to work on an entire row (i.e., to study broadly a single segment) or on an entire column (i.e., to study broadly a single question). Other Project Teams may prefer to work on a single cell or some set of cells. All possibilities are welcome so long as the people in each Project Team are qualified and motivated. Duplicate coverage of cells will not be discouraged, but the Steering Committee will coordinate such situations. It will also seek to promote as much consistency as possible among Project Team activities (e.g., their interviewing techniques). The Project Teams will be formed from volunteers during the first few months of 1996. The Steering Committee will help volunteers with similar interests find each other and will perform inter-team coordination after formation. In both tasks, Email reflectors are likely to play an important role. Every non-academic Roundtable representative will be invited to volunteer to coordinate an all-key-question canvas of as many OR/MS workers in their organization as possible. Thus, ideally, there will be more than 40 Project Teams from Roundtable member organizations. CPMS members will be invited to volunteer to coordinate similar canvases of their organizations, although more sharply focused canvases also will be welcome. The INFORMS membership at large will be invited to volunteer teams with any segment/key-question focus. Participation by academic departments as well as companies will be most welcome; we hope that many will decide that this is the right time to conduct a class project or alumni-tracking study of their own that dovetails with this project. ACORD (Association of Chairs of OR Departments in universities) is the perfect body to orchestrate team formation in academia, and President John Birge has placed this topic on its Fall Meeting Agenda for discussion. Non-INFORMS volunteers will be welcome also. For example, we hope that some neighboring professional associations can be induced to conduct a study of their own that dovetails with this one. Since the project relies so heavily on volunteers, it is important to know what incentives will attract them. Partial list: - the satisfaction of doing important pro bono work for the profession at a time of great need - early access to project materials and results - visibility in the OR/MS community - public credits given by INFORMS and the Roundtable (consider how to credit particularly good project work) - a good excuse to get in touch with people with whom there is good reason to establish or improve contact (e.g., most universities find it valuable to improve contacts with local industry and professional alumni) - publication opportunities. The last item could be a particularly potent motivator for academic volunteers. To take full advantage of it, the project needs the support of the editors of Interfaces, OR/MS Today, ITORMS (the new Interactive Transactions on Operations Research and Management Science) and as many other INFORMS publications as possible. Editors should be asked to invite project-based submissions under special reviewing guidelines designed to encourage and expedite publication of project results. John Llewellyn and Mike Rothkopf already have stated their intention to cooperate. Editorial cooperation means much more to the project than merely providing outlets for worthy work, important as that is: it gives access to editorial staffs and reviewers whose business is quality control and clear exposition. This should greatly relieve the Steering Committee of much of the work that would otherwise inevitably fall to it. Expert advice should be sought on coordination and net-based survey methods, and on appropriate tools to support them (e.g., contact management and text database software). No more Project Teams should be accepted than can reasonably be coordinated by the Steering Committee using the best available procedures and tools. INTERNET The Internet is this project's critical enabling technology. Without it, a project of this magnitude would be utterly impractical to coordinate or execute even with the able assistance of editorial staffs and reviewers (as just noted). Consequently, the Internet must be exploited aggressively to reduce the project's labor requirements and costs as much as possible, to shorten its schedule, and to facilitate dialog-style interaction with practitioners and their clients on the key questions (it remains to be seen how effective this last use can be by comparison with telephone and face-to-face interaction). In addition to familiar Email and file transfer capabilities like ftp, the Internet offers INFORMS Online (INFORMS' official web server), Practice Online (a moderated Email reflector/file server and also a web server for the practitioner community co-sponsored by Interfaces, the Roundtable, and CPMS), ORCS-L (an Email reflector focusing on the OR/Computer Science interface), sci.op-research (an Internet USENET newsgroup catering to OR/MS) and several other news groups, WORMS (an OR web server in Australia), etc. Here are some examples of Internet use: The Internet can put the Project Plan into wide circulation long before it appears in OR/MS Today: simply announce it through the various Internet channels just mentioned. OR/MS Today should still carry the full Project Plan along with an invitation to readers to offer suggestions (e.g., data sources that might otherwise be overlooked), and the December issue should carry a prominent notice about the project and where to find the full Project Plan on-line (closing date Nov. 6). Email can be used as a speedy no-cost substitute for postal mail surveys. Since respondents can just hit Reply and fill in the blanks, the response rate may be fairly high. Web pages with embedded forms should also be effective (this raises the intriguing prospect of practitioners being able to contribute data to the project without anyone having to contact them!). Email reflectors, USENET news groups, and also discussion forums on Aol and CompuServe can be used to discuss the key questions and air Project Teams' findings and conclusions as they emerge. It can be expected that discussants will offer personal experiences, information about their organizations, cogent criticisms, alternative viewpoints, and constructive suggestions. Of course, Project Teams will have to provide the discussion leaders. Practice Online (sketched above) should play a leading role for both discussion and the project's on-line knowledge base. Terry Harrison and Mike Rothkopf are fully cooperative. The on-line knowledge base to be created by this project will be its centerpiece, structured and designed so that Project Teams can flesh it out with little or no interaction with the Steering Committee, and even so that individuals can contribute to it whenever they wish. This is a crucial technology for making a project of this magnitude manageable. Moreover, this knowledge base can remain alive and growing even after project termination. COST The seed cost range is $10,000 - $20,000 to cover out-of-pocket expenses only: fax/telephone, postage, duplication/ printing, clerical/student assistance, supplies, travel, etc.. Although it would be desirable to make provision for modest honoraria for Project Team professional services, the scarcity of funds implies a project policy that all such services will have to be pro bono. Sample numbers: 300 hours of clerical/student time at $15/hr, 400 toll-call interviews @ $10/ea, postage for one mailing to 1000 non-INFORMS people, 1000 faxes at $2/each, and $2,000 for duplication/supplies comes to $13,000 even before travel (which will, for the most part, have to piggyback on regular personal and professional travel). The Roundtable will be asked to guarantee half of the seed cost by voluntary member donations (cash or services), backed by its general treasury. The INFORMS Board will be asked guarantee the other half of the seed cost, some of it possibly in the form of Business Office clerical assistance between peak load times. Failure of either INFORMS or the Roundtable to shoulder half the seed cost probably would result in a withdrawal of this proposal. An INFORMS-wide appeal for contributions will be made once seed funding is secure. Proceeds will be used to supplement the seed budget for out-of-pocket expenses and possibly also for specialized coordination services. The appeal will include academic departments, which stand to gain much from a successful project. ACORD (mentioned earlier) will be asked to orchestrate university contributions. SCHEDULE 8/95 Form a Pre-Steering Committee to draft Project Plan Guidelines and plan the joint Roundtable/INFORMS Board meeting in New Orleans. Members (alphabetical): Tom Baker, Burnell Brown, Art Geoffrion, Sid Hess, Mark Lembersky, Randy Robinson, Leon Schwartz, and Tom Spencer. All agreed to serve. 10/95 New Orleans INFORMS Conference The Pre-Steering Committee leads the joint Roundtable/INFORMS Board meeting. Meeting purpose: 1) improve the 10/20/95 Draft Project Plan Guidelines, including a possible major reduction of scope; 2) authorize the proposed project as modified and guarantee funding; and 3) appoint an official Steering Committee to finalize the Project Plan and supervise the project. Steering Committee to be composed mostly of representatives from both INFORMS and the Roundtable, with the weight heavily toward the Roundtable because the Board already has much to do and the Roundtable has focused traditionally on practitioners. Suggested Members: Begin with the Pre-Steering Committee minus the outgoing VP-Practice, with most people given the option to withdraw because of the project's considerable work requirements. (Every Steering Committee member will need to be actively involved.) However, INFORMS' incoming VP-Practice and Randy Robinson need to stay because of their key roles. Add Terry Harrison, who runs Practice Online and is a strong supporter of this initiative. The INFORMS Board is invited to consider assigning one or two pertinent Directors-at-Large to the Steering Committee, preferably the ones charged with membership and outreach responsibility. ACORD should be invited to contribute a member. Consider adding others on the basis of ability and willingness to contribute. 12/95 The Steering Committee releases the final Project Plan with official approval from both INFORMS and the Roundtable, and begins seeking and screening volunteer Project Teams. Release is through OR/MS Today and all pertinent Internet outlets (Email reflectors, news groups, and web sites). Screening and approval will occur continuously through 4/96. Accepted Project Teams will be encouraged to commence work immediately, and to make their project data available in the on-line knowledge base as early as possible. The Steering Committee commences to coordinate across (but not within) Project Teams, disburse available funds, and monitor Project Plan execution. Great reliance is placed on Project Team autonomy, on the on-line knowledge base, and on editorial staffs and reviewers to keep the Committee's workload reasonable. 5/96 Washington INFORMS Conference As many members of the Project Teams (by now fully selected) as possible meet with the Steering Committee and with each other to coordinate and refine the Project Plan. Commonality of framework and consistency of approach will be pursued, but not to the point that these become obstacles. Take advantage of this occasion to do data gathering among conference attendees and among non-Roundtable organizations in the Washington area. 8/96 Project Teams all make written progress reports to the Steering Committee, which offers constructive feedback and suggests mid-course Project Plan corrections as needed. Teams submit all available project data in electronic form suitable for on-line access, a task that has been optional until this point. Notwithstanding this milestone, the Steering Committee continues to coordinate the Project Teams thereafter. 11/96 Atlanta INFORMS Conference Project Teams present to the Steering Committee their respective preliminary findings, typically segment-specific, and action recommendations for practitioners, INFORMS Board, and the Roundtable; outline of results submitted 3-4 weeks prior, so that feedback/ discussion/action items can be prepared. Steering Committee then presents distilled preliminary findings and recommendations to the Roundtable and the INFORMS Board, with assistance by invitation only from the Project Teams. INFORMS Board and the Roundtable take whatever actions they deem appropriate. 3/97 Project Teams finalize their respective findings and recommendations, typically segment-specific, and submit these to Steering Committee together with all as-yet-unsubmitted project data in electronic form suitable for on-line access. 5/97 San Diego INFORMS Conference Steering Committee, with assistance by invitation only from the Project Teams, presents final project findings and action recommendations to a special open session at this Conference, with time set aside for audience feedback and discussion. Steering Committee works with INFORMS officers and Roundtable Executive Committee on action items. INFORMS Board and Roundtable take whatever actions they deem appropriate. Later Project Teams are encouraged to make versions of their respective results available for publication in OR/MS Today, Interfaces, ITORMS, and other suitable outlets. Steering Committee will seek to facilitate such activities via prior arrangements with the editors. SOME PARTICULARS Coordination with New INFORMS Efforts Increasing the number of its members has recently become one of INFORMS' most urgent priorities. By far the easiest way to do this is to seek new practitioner members, since only a tiny fraction of all OR/MS practitioners (perhaps on the order of one tenth) so far has elected to join INFORMS. The Marketing Strategy Committee and Membership Committee want to do a survey. Very likely the next Marketing Director also will want to do surveys. Diligent coordination can produce a great deal of synergy between such efforts and this project. Definition of Practitioner The project needs to define the term "practitioner". One possibility: anyone knowledgeable about OR/MS who derives a significant part of their livelihood outside of academia by building or using models. Another possibility would be to let people self-identify. Communications As mentioned, great reliance will be placed on Email reflectors, and to some extent also on the web and discussion groups on USENET and the popular on-line services, for data acquisition and key-question dialog with practitioners and their clients. This has the advantage of being essentially free, speedy, easy, and of leaving a written record that can be edited, copied, searched, and made available on-line for anyone to access. However, there are practitioners and clients who are not yet on the Internet, although we suspect that Aol, CompuServe, and Prodigy have by now brought Internet access to most professionals who don't have access at work. It will therefore be necessary to use fax and phone to contact these people when personal interviews are not practical. This will cost money. A project 800 number for half a year or so might produce savings. Communications subsidies should be sought from phone companies (e.g., AT&T, GTE, and Pacific Bell are Roundtable members). The phone should not be viewed as an undesirable substitute for Internet communication. In fact, for some purposes, a good case can be made for exactly the opposite viewpoint; the personal touch and greater interactivity can elicit comments and details that would not otherwise emerge. Project Teams will need to share their experiences in this regard early in the project. There is an experimental Internet service that sends faxes for free through participating sites. The basic idea is to use the Internet for long haul, and then to make final delivery via a local or nearly-local call. Data Sources Phase 2 of the project requires marshalling already existing data and knowledge pertaining to the key questions. This work should begin very early in the project. Existing data sources include: Commerce Dept., including the famous BLS estimates concerning the population of "Operations Research Analysts" Archives of practitioners' messages to OR-related Email reflectors and USENET news groups (usually these can be identified easily from their originating Email addresses), subscription rosters for Email reflectors, and access logs for OR-related gopher sites and Web pages; this is also a good way to find practitioners who are on-line Recruiting firms with OR/MS-related practices The practitioner membership history of ORSA and TIMS as reflected in past membership directories (request machine- readable versions); look for non-INFORMS members in INFORMS journal subscription lists Past ORSA and TIMS membership surveys (at least 9 in the last decade, with more in the offing); cull the practitioner responses Published literature and data collected by past studies of OR/MS practice; the most recent such study is the important project sponsored by the U.K.'s Operational Research Society, which briefed the Roundtable earlier this year: see Ranyard et al. (1995) and the comprehensive literature review by Fildes and Ranyard (1995) INFORMS people knowledgeable about aspects of the practice community; for example, Doug Samuelson is reported to be familiar with how the BLS develops its statistics, and Arnold Reisman has studied extensively how OR/MS publications and academic curricula are frequently misaligned with the methods practitioners actually use Edelman competition papers and archives Personal contacts with former Roundtable representatives at companies which are no longer Roundtable members (usually due to internal troubles) University-initiated surveys like the one now being done by AT&T (which has a group that is expert in doing surveys and focus groups) for New Jersey Institute of Technology; alumni surveys done by academic OR/MS programs (perhaps this project will stimulate new ones that probe some of the project's key questions) Studies done by neighboring professional associations Archives of the Roundtable's 1986 "Problems and Opportunities" report to TIMS Council (also a possible source of additional key questions) Enclosures in OR/MS Today and in the registration packets at the 5/96 Washington INFORMS Conference would be an economical way to obtain data from (mostly) inside the INFORMS community; if permission can be obtained, enclosures in magazines of related professional associations would be an economical way to obtain data outside the INFORMS orbit. Can such an ambitious project succeed? Can the OR/MS practitioner community really be mapped by the coordinated efforts of many small teams? Surely this task is far simpler in most respects than mapping the human genome. I leave you with the following food for thought: "The Human Genome Project has often been compared with World War II's Manhattan Project in its ambition and scope. Yet, unlike the Manhattan Project ... the Human Genome Project is almost completely decentralized. The funding agencies have parceled out tiny pieces of the project to scores of research teams scattered around the world. In spite of this decentralization, which you might naturally suspect would be a barrier to effective collaboration and lead to fragmented, duplicated efforts, the Human Genome Project is ahead of schedule and under budget. How is this possible? The data each research team generates is electronically submitted over the Internet to one of many specialized databases ... the data is then linked to related data in the same and other databases, and it's made available almost immediately to the entire genome research community via Web servers that are universally accessible on the Internet." (Duncan 1995)
REFERENCES Corbett, C., W. Overmeer and L. Van Wassenhove, "Strands of Practice in OR," Working Paper 95/44/TM, 5 May 1995, INSEAD, Fountainebleau, France. Duncan, R., "Publishing Databases on the World-Wide Web," PC Magazine, August 1995, pp. 403-412. Fildes, R. and J. Ranyard, "Success and Survival of OR Groups: A Review," Dept. of Management Science, Lancaster Univ., 1995. Ranyard, J., W. Crymble and R. Fildes, "Factors Influencing the Success of OR Groups - A Progress Report (June 1995)," to appear in ORS Newsletter. Sent to all Roundtable members in mid-September. Final report in transit. =================================================================