Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix B Minority Opinion This report has a wealth of good material in it, but I feel that ~ must write a minority opinion on one main issue, the committee's recommendation to create the NSIR Federation. I think that the exact functions of the NSIR Federation are still not clear enough to immediately form it, especially since mechanisms to coordinate data activities already exist. A group such as the NSTR Federation would not be a good method to set the hardware standards that are used in data systems (networks, tapes, etch. The coordinated part of data directory efforts can be built around present interagency work. It is reasonable that NARA should request lists of datasets intended for long-term archival, but most of the process of evaluating datasets needs to be kept close to the working level. The discussion of standardization in the report should not be interpreted to mean that all agencies and archives should be forced to adopt certain standards and rework their data holdings into a common form and format. There are other concerns for which an analysis of the issues could be useful, but I believe that the NSIR Federation requires a better description of tasks and more debate before such a new body is established. Otherwise we may have more coordination, more systems, more cost, and less data. Consider the important task of developing information about data. Information about datasets is needed in at least two or three levels of detail. At the highest level of information, the Master Directory methods that are in place for the GCDIS can be adopted (or even simplified more) to describe the datasets. This interagency Directory Interchange Format (DIF) is used nationally and internationally. We need to keep it simple enough so that people will submit the information. Some agency-level catalog efforts for datasets have existed since about 1968, and became more serious in the late 1970s. We should build on the GCDIS catalog efforts, and certainly not invent more complicated systems. Other data information efforts are needed, but they will be based on a bottom-up flow of ideas, on workshops, and the like. Each data system does not have to do exactly the same thing, but they must be easy to use. It is not clear that a formal NSIR Federation is needed to coordinate this. How does the NSTR Federation relate to other data coordinating mechanisms? The Interagency Working Group on Data Management for Global Change (IWGDMGC) meets regularly to help coordi- nate data issues across many "global change" disciplines, which include air, water, ice, rocks, soils, and some biology. It seems to me that the IWGDMGC and the proposed NSIR Federation are mainly trying to do the same thing. They cover much of the same turf in terms of disciplines. They both want information about data, access to data, and data that will exist for more than 20 years. If we create separate organizations doing roughly the same thing, then it becomes even less likely that key agency 66
Appendix B 67 people will attend the meetings. NARA asked the committee to consider how to deal with all the observational and laboratory physical sciences. The argument is made that we need an NSIR Federation because the IWGDMGC does not include some disciplines that were included in this study. However, NASA has control of most of the data for planetary sciences and astronomy so that this area may not be very hard to coordinate, except that data from ground-based telescopes should be included. This leaves the laboratory sciences, which can be handled as a special case. Can the TWGDMGC be characterized as only agencies talking to agencies? No, there is a long- standing NRC panel, the Committee on Geophysical and Environmental Data, that has been asked to oversee its work, and that group has sponsored periodic national data forums. Does this mean that it is perfect? No, but it is not convincing to me that a roughly parallel coordination effort by an NSIR Federation would be necessary. Some coordinating mechanisms besides the TWGDMGC will be needed to achieve the goal of making sure that long-term digital archives do exist, are adequately described, and can be used. NARA could hold periodic advisory panel meetings or workshops to talk about concerns and possible solutions. Similar issue-oriented meetings have been sponsored by other agencies and should continue. In interagency planning, the agencies should remember that some good data activities outside of agencies are funded by the agencies, but are probably not adequately represented by typical agency planning. This could be an argument for an NSTR Federation, but the problem could be handled in other ways. The idea of an NSIR Federation that is nimble, non-bureaucratic, and small is attractive, and it could even be a counter-weight to the agencies when that is needed. But we still have to ask: What would the NSIR Federation really do? Why would not it be just another coordinating office? Why would the agencies want to support it? I believe that the NRC staff for this study has been very able and conscientious in helping to pull together this report. The report underwent significant change, but r was unable to fully support the committee's majority position regarding the proposed NSTR Federation. Nevertheless, I think that some divergence in viewpoints can help the sponsors and other readers to evaluate the best course of action. Roy Jenne