Linking Process and Content in a Distributed Spatial Production System

KUCERA, Henry (hkucera@ibm.net), Mercator Systems, Ltd., 2936 Phyllis Street, Victoria, V8V 4L8, B.C, Canada; and LAFOND, Pierre (plafond@holonics.ca)), Holonics Data Management Group, Ltd., 200 Montcalm, Suite, 105, Hull, J8Y 3B5, Q.C., Canada

Key Words: metacontent, distributed computing, multi-resolution, repository

Advances in technology have so improved the efficiency of data collection that organizations find it difficult to keep up with the growing flood of data. Many organizations continue to prepare the data for single narrowly focused applications in which only the local creators of particular databases may be very knowledgeable about their characteristics and legacy. While local analysts may use a particular data set with confidence, consideration is seldom given to recording the history of the database in a form that can be conveyed to remote users. As a potential user of this data, how do I know if these data are good? Can I easily find out where it came from? How was it transformed and cleansed before being loaded in the database? What other activities are currently going on in the geographic area that I am working in? These are questions that many people may be interested in, and should be easily answered by simply querying an existing database, using standard query tools, over a local-area network, or the Internet.

Remote users who have had no part in collecting or storing the data originally, and are separated from the data collection activity by time and space, may not be experts in a given subject area that the data represents. Remote information analysts who wish to query and/or process the data are dependent on information about the assumptions made prior to collecting the data, the sampling procedure, potential sources of bias, inconsistencies and/or error. Lack of knowledge of these factors can limit the manner in which particular data can be manipulated and interpreted. Because of this, a wealth of information collected at enormous expense remains locked in databases and cannot be mined to extract additional knowledge.

This paper presents a process management approach that transforms information to knowledge through the concept of "Metacontent Management." The approach combines standard methods for planning and tracking of spatial data acquisition, validation and production activities, with cutting edge tools for managing complex information. The paper discusses the steps we have taken in ongoing projects to move beyond the primitive use of "metadata," and demonstrates how process management tools can be used to facilitate collaboration between people involved in all phases of the spatial data management life cycle. This approach ensures quality and repeatability of the compilation process and streamlines analysis and synthesis of the resulting information.