China In Time And Space: An Outline Proposal William Lavely July 15, 1994 This document sketches a re-organized CITAS project. The basic thrust is as follows: (1) CITAS should become more sinocentric, by which I mean, more of the project activities should involve Chinese organizations, and more, if not most, of CITAS activities should take place in China. (2) CITAS should seek affiliation with the Committee on Scholarly Communication with China, and shift a large part of its administration to the CSCC offices in Washington and Beijing. (3) Relations with CIESIN should be strengthened, with CIESIN providing more technical support and training for Chinese collaborating organizations; (4) CITAS should sponsor a set of decentralized research projects that would utilize and develop important data sets. These research projects would be international collaborations in global change, historical, social science, and GIS fields. CITAS Goals The original impetus for CITAS was a desire to assimilate and organize vast quantities of Chinese county-level data for the benefit of China scholars. From that relatively straightforward beginning, CITAS has become a multifarious enterprise. The initial goals embodied what I would call an integrative view of CITAS: the gathering of disparate data and their integration in a GIS framework. CIESIN, as a data network, brought to CITAS a broader view, as a clearinghouse that would locate, gain access to, and document disparate data sets for use by a wide spectrum of scholars. From the beginning, we have pursued these two visions simultaneously. On the one hand, we have emphasized the use of GIS technology, and have set out to construct a general-purpose 1:1M national GIS of China emphasizing county attribute data. On the other, we have begun projects to locate data and produce "metadata" on Chinese society, with the mild proviso that our interest is in digital, georeferenced data. CITAS is potentially a very cost-effective way to provide important new data for global change and social science research. Setting aside the well- rehearsed case that Chinese data are unique in their historical depth and coverage, one could argue that the integrative technology of GIS will permit the reclamation of data previously collected but now lost to the scholarly community. A tremendous investment in Chinese data has already been made. Hundreds of historians, in China and abroad, have spent their lives gathering data on grain prices, weather, transport, and population. Millions of man-hours have been spent in compiling local statistics, mapping land use patterns, and tracing changing river courses and coastlines. And billions of dollars have been spent on remote sensing data. At the moment, these data largely reside in a forbidding and inaccessible wilderness. CITAS can reclaim these resources by providing the organizing intelligence behind a decentralized effort to standardize, document, integrate, and share Chinese data. This would involve a range of activities including sponsored research, cataloging and documentation, GIS development and the creation of specific data products, development of protocols for data sharing, training seminars, and conferences. Collaboration with Chinese Organizations Recognizing from the start that the source of Chinese data is in China, a priority task in the first year was to learn about Chinese GIS and environmental research, and to make contacts with Chinese organizations. An account of the major organizations contacted is contained in a separate memo. This is an auspicious moment to establish relationships with Chinese organizations because (1) collaborations with foreign entities are now mandated by the Eighth Five Year Plan; (2) regulations governing the classification and distribution of data are undergoing liberalization; (3) the GIS field in China has already produced important data resources and technical advances, yet the field is still young enough that standards are still being set and technical cooperation with foreign partners is sought after. The organizations contacted expressed great interest in cooperation with CITAS without exception. Let me mention, somewhat redundantly, some of the kinds of data that are available: The Bureau of Surveying is planning to release a version of the 1:1M Fundamental GIS of China's Territory, essentially, a "turn key" GIS in ARC/INFO format, containing data that underlies the LUMC and virtually all 1:1M map work in Chinese research organizations. Digital, county-level data exist in great abundance, from censuses, yearbooks, and other sources unavailable abroad in any form. Even when available abroad, it does not make economic sense to have foreign scholars engaged in data entry projects. Chinese institutes are very well set up to do data entry, and they can do it for a fraction of the cost that it can be done in the US. Important national coverage environmental and climatological data already exist, much of it in readily assimilable GIS formats at scales ranging from 1:4M up to 1:1M. Examples are the land use data of the Institute of Geography and the vegetation data of the Institute of Botany. But there are also scores of regional data sets at much finer resolution, focused on specific ecosystems such as grasslands, plains, forests, and river basins. Important historical data sets exist, some in digital form. Examples are the tabular data on historical climate and disasters in the Institute of Geography, the base maps of the Historical Atlas of China from the Fudan Institute of Historical Geography, and the remotely sensed data on historical river courses, deltas, and coastlines from the Institute of Remote Sensing Applications. Given that there are extraordinary data resources in China, how do we establish relationships with data originators? It will not be easy. In my view, CITAS needs to shape itself into an organization that can engage in substantive research with Chinese researchers, provide technical assistance to Chinese institutions, and foster collaboration among Chinese researchers and between Chinese and foreign research organizations. In other words, we need to establish complex, ongoing links with Chinese organizations. Such links would serve CITAS goals, and they are what Chinese organizations want. This is a broader vision of CITAS activities than we had a year ago. When we first discussed the idea of agreements with Chinese organizations last year, CIESIN suggested a model based on simple reciprocity. Simply put, they give us their data, and we'll give them ours. Some of us were frankly skeptical that this approach would work. A different model of exchange, perhaps more in keeping with the times, would be a pure commercial transaction. If CITAS needs a data set, we would discuss price and attempt to purchase it. Although each of these models may be feasible in some circumstances, most of the organizations I held discussions with are looking for a much more complex relationship that in many cases can be described as a research collaboration. The reason for this is no doubt in some part due to the strictures of government policy. Research organizations are not supposed to sell the data that were created at public expense (although some do). However, research collaborations are encouraged, and the creation or modification of data in the course of the collaboration can, in effect, establish foreign proprietary rights. In other words, to avoid the appearance of a commercial transaction, one establishes an exchange in the form of a research collaboration. However, the desire for substantive collaboration is not merely a legalistic ploy. What the institutes generally want from a collaboration is technology: training opportunities, equipment, software. The goal of collaboration is to help them to keep up with the cutting edge of the field. From the point of view of CITAS, substantive research ties represent an opportunity to leverage our activities. CITAS should be doing in collaboration with Chinese organizations what we have been attempting to do with a very small infrastructure, namely, to locate and document data sets, create new data sets, integrate them, where appropriate, in a GIS framework, set standards for coding and place names, and find technical solutions to GIS problems such as change over time. The fact is, these are big jobs. Even if CITAS were far larger, we could not do them all. But the infrastructure for these tasks exists in China. The CAS institutes have hundreds of technical personnel involved in GIS work. Through engagement with these organizations, CITAS can tap into a tremendous pool of manpower and expertise that can be oriented to projects of mutual interest. Research collaboration implies considerable investments of time and expertise. CITAS will need to organize scholars and channel resources into strategic collaborations. There are a number of ways that this could be done. CITAS needs to establish a network of scholars and consultants who would be invited to participate in specific collaborations. These collaborative teams would in turn be linked together via electronic networks and periodic meetings of broader expert groups. The first, prototype collaboration may be established later this year. We intend to collaborate with the Chinese Academy of Surveying and Mapping to produce a GIS of county boundaries and place names for the period 1949- 1994. CITAS will establish a working group which will meet with Chinese counterparts to define problems and set tasks. We can assemble this working group from current CITAS staff and associates, but if we had to set up five or ten such collaborations, we would not have the manpower to do it. CITAS Organization and Administration CITAS organizational structure is still nebulous. Although CITAS originated as an ad hoc committee appointed by the Joint Committee on Chinese Studies, CITAS exists as an informal association of China scholars (the Policy and Planning Committee), and an equally informal panel of advisors (the CITAS Board). Its only formal identity is as a research project at the University of Washington. It has been clear for some time that CITAS needs a more formal structure and broader national and international institutional affiliations, but now these are urgent matters. Institutional Affiliation. In the past we have discussed options for the institutionalization of CITAS. There is broad agreement that an expanded CITAS cannot function merely as a university research project, and there is also agreement that incorporation as an independent non-profit organization is impractical. Most discussion has centered around organizing CITAS under an existing scholarly organization such as the ACLS. CITAS needs a prestigious parent organization with links to the China field, links to the scientific community involved with global change research, and with links to Chinese organizations, and preferably, one that has a pre-existing administrative structure that can coordinate scholarly activity in the US and China. One such organization exists, the Committee on Scholarly Communication with China. The CSCC is organized under the ACLS, the SSRC, and the National Academy of Sciences. It is the only organization that cuts across the natural sciences, the social sciences, and the Chinese studies field. In light of the fact that CITAS wishes to develop close ties with several laboratories and institutes in the Chinese Academy of Sciences, CSCC affiliation with NAS is particularly useful. CSCC is set up to coordinate and administer interdisciplinary research projects involving Chinese research organizations. Another advantage of CSCC as a CITAS administrative center is the CSCC Beijing office. I have argued that a large part of CITAS operations should be involved with fostering research collaborations between Chinese organizations and the international scholarly community. If CITAS activities focus on relations with Chinese organizations, it implies that the center of gravity of CITAS activities shifts from North America to China. If CITAS sponsors research activities, data acquisition, technical assistance, training, or committee meetings involving Chinese organizations, efficiency would dictate that many liaison, technical, and management tasks be located in China as well. No matter how large or well-funded an organization that CITAS might become, it would be extremely difficult to sustain a high level of regular contact with Chinese organizations if these activities were located in the main in an American institution. Moreover, many Chinese organizations are working on similar, complementary, or allied projects, or facing common technical problems. County-level attribute data, for example, is a common denominator of much Chinese GIS work. Historical reconstructions similarly encounter a host of common problems. CITAS thus should not merely be seeking bilateral ties, but a set of interlocking associations with and among Chinese organizations. In other words, CITAS should seek a seat at the table where Chinese partners work on common technical and substantive issues. In this way, CITAS could be a catalyst for research on standardization of GIS, data sharing, networks, and related issues. These interactions would logically take place in Beijing. Such meetings would require CITAS representation and advance work. If CITAS is to provide technical assistance or consulting, these activities would best be coordinated locally. And research done in China would best be monitored locally, as would, to some extent, research budgets. And at least one or two international conferences could be expected to be mounted in China. All of these activities would require a CITAS presence in China. The CSCC Beijing office exists now. It has close ties to CAS (CAS has been the host of the CSCC Bejing office since 1985) and is already involved in activities of this kind. The CSCC currently has a two-person staff and costs about $200,000 per year to run at present. The office is currently funded by the Ford , Luce and MacArthur foundations, among others. CITAS business could perhaps take up a quarter to a half of the office's efforts. It is possible to envision the expansion of the office to include a resident CITAS post-doctoral fellow or other personnel. Based on my conversations with James Feinerman, CSCC Director, and with Keith Clemenger, CSCC Beijing representative, CSCC seems to be quite interested in a linkage with CITAS. The role of CSCC is changing, and CSCC is looking for a large, interdisciplinary, and environmentally- oriented project, which is what CITAS is. Although the argument for a CITAS affiliation with CSCC seems compelling, it is less clear what form the affiliation should take. It would be desirable for CITAS to maintain a formal identity and governing structure separate from CSCC. Perhaps CITAS could become a committee under the ACLS, SSRC and NAS, formally co-equal with CSCC, but administered within CSCC. This would make it easier for CITAS to assume the status of an international organization while it would reduce administrative overhead. A distributed research network. CITAS needs to develop a research component. This does not mean hiring a group of researchers. Rather, CITAS should foster research on key projects that would result in technical progress on GIS problems or which would utilize, improve, and expand data sets in key areas. In some cases, for example, research on a specific GIS problem, CITAS might be the sole sponsor of a project. But for the most part, I would envisage that CITAS would provide additional support to ongoing projects to assist on areas of direct interest to CITAS. CITAS could, for example, sponsor a component of an ongoing project in the global change area that would involve both Chinese and foreign researchers and which would produce new data and documentation. Such a research component is needed for several reasons: (1) Specialized expertise is needed in the construction and documentation of data sets. This is particularly true in the case of historical research. (2) Priorities for data acquisition should be driven by substantive research needs. Research projects would allow experts to decide which data are important. (3) International research collaborations involve considerable demands on time and facilities. A decentralized set of research projects would spread this overhead cost across many institutions. (4) With the completion of projects, CITAS would be associated with specific research advances. Relations with CIESIN CIESIN has been the only funder of CITAS, but from the beginning CIESIN has been more than just a source of funds. It is fair to say that CIESIN has been a partner that has provided technical services and advice, assumed responsibility for data distribution and related administrative services, and, most importantly, broadened our initial horizons and guided us towards a more ambitious conception of what CITAS should do. Despite some minor ups and downs in CITAS-CIESIN relations, it has overall been a close and positive relationship. As we look forward to the day when CITAS will be funded from other sources, we need to consider the future of relations with CIESIN. In my view, in the future CITAS will need to establish closer relations with CIESIN and rely even more heavily on CIESIN for technical services. CITAS, in turn, may take on a role of negotiating and implementing expanding relations between CIESIN and Chinese organizations. Data distribution. To the extent possible, CITAS wants to remain a technical organization that is not involved with customer services, thus CIESIN's role as a center for distribution of CITAS data serves us well. This aspect of CITAS-CIESIN relations should remain unchanged. Technical assistance to Chinese organizations. As mentioned elsewhere in this memo, exchanges with Chinese organizations will require substantial financial resources, but the actual exchanges will rarely take place as purely cash transactions. What Chinese organizations generally want from an exchange with CITAS is technical assistance, particularly on advanced communications technologies, GIS, and information technologies. Links to the Internet are widely sought by Chinese research organizations. Training opportunities are desired. Interest is expressed in information software such as CIESIN's software that permits on-line queries of the US Census PUMS. Desire for equipment and software are often mentioned, as are data exchanges. While CITAS is in no position to provide these services, CIESIN is, and in fact is already involved in transferring these technologies to China. Exchanges with Chinese organizations will require a source of technical expertise, much of which, although not all, could be provided by CIESIN. This might include opportunities for training at CIESIN in Saginaw; CIESIN consultants visiting Chinese organizations; and transfer of equipment and software related to communication and information systems. These inputs will be expensive, and it is difficult to anticipate in advance just what will be needed. It is possible to see CIESIN as an integral partner in our dealings with Chinese organizations, just as it is possible to envision CITAS as acting on CIESIN's behalf in China. Issues of copyright. A fundamental aim of CITAS is to distribute data worldwide, via CIESIN, to scholars and non-commercial users at nominal cost. To do this, it is necessary to obtain the rights to distribute data from the originating organization. This is a far more complicated problem than it initially appeared (the Griffith problem with the Land Use Map of China is a case in point). We can anticipate the following problems: (1) Rights to data in China are ambiguous, especially with respect to data sets that are created by multi-institute projects; (2) There may be an unwillingness in Chinese organizations to cede rights to distribution worldwide. In the case of the LUMC data, the Geography Institute would like to make separate bi-lateral arrangements with Griffith for Australian rights. (3) There is limited expertise in CITAS and CIESIN on the legal implications of data distribution. Even now, the legal status of the tabular data sets being created by Professors Dernberger and Skinner are not well understood. (4) Negotiating and finalizing agreements on data sharing is likely to be a protracted process requiring more legal expertise and backup than is currently available to the CITAS project. In anticipation of these problems, it would be well for CITAS to take the following measures: (1) CITAS should, where possible, seek agreements with high-level organizations and organizations that are the originators of major data sets. An example of a high-level organization would be CERN, which is a high level data distribution node for environmental data. Examples of data originators would be The Bureau of Surveying, purveyor of the Fundamental GIS of China, and the Chinese Academy of Social Sciences, co-owner of the rights to the Historical Atlas of China. (2) CITAS must assume a firm identity as an international, rather than a US organization. This will help us head off attempts to conceive data sharing agreements as a purely Sino-American matter. (3) CITAS needs a source of legal advice on intellectual property matters, preferably including issues of technical innovation in software. (4) CITAS will probably need legal counsel to get directly involved in the negotiating process with Chinese organizations. Up to now we have assumed that CIESIN would provide this expertise. However, our proposal should budget for legal services, unless it can be assured that CIESIN has budgeted for them. 8