Computerizing the Moon Rock Collection

Curatorial Data Architecture - Present & Future

by Claire Dardano* and Jim Gooding**
*Lockheed Martin Engineering & Science Services Company
**NASA JSC

As important as the careful handling of lunar samples is the orderly preservation of information about the samples. While the technology for chipping and cutting Moon rocks has remained a safe, non-threatening constant since the 1970s, the technology available for handling the associated records of sample history has grown by amazing leaps. As a result, one of the most important and cost-effective improvements that can be made to lunar sample curation is the modernization of our data systems. The architecture that we conceived two years ago, which we are systematically implementing within our budget constraints, is shown in Fig. 1; our plan and progress toward that end is shown in Fig. 2. Our campaign has three principal goals: (1) Automate selected operations to allow more samples to be processed per unit time; (2) Accomplish direct entry of numerical data to eliminate transcription errors; (3) Capture all records in electronic formats that can be reliably preserved for posterity as well as rapidly served to remote users.

From 1969 to the present, primary data from sample processing have been recorded on paper (Fig. 2, Step 1). Sample weight, container, and descriptive information are handwritten on paper forms. In addition, black & white Polaroid photographs are taken at various stages of processing to document the subdivision of the samples, and color photographic film is used for new samples and special operations. The forms and the photographs are placed in folders called data packs. Selected information is then extracted from the handwritten forms and entered using customized programs into a text-only database which resides on a Micro Vax III computer (Fig.2, Step 2). The information in the curatorial databases has been available to the general public for several years via TCP/IP using TELNET or via modem (Fig. 2, Steps 3 and 4).

Beginning in the summer of 1994, two projects were undertaken to upgrade curatorial data operations. First, the team began a project to digitize and store the data packs for the lunar sample collection in an electronic document management system (Fig. 2, Step 6) using File Magic software. (See Lunar News, No. 58, First Steps Toward Lunar Sample History on "Multimedia", February 1995.) These data packs represent over 25 years of aging paper and photographs, and approximately 40% of this information has now been incorporated into the electronic document system. The second project was the establishment of a curatorial World Wide Web (WWW) site (Fig. 2, Steps 11 and 12). (See Lunar News, No. 58, Announcing a New World Wide Web Site!, February 1995.) This site is housed on a Windows NT Server running Microsoft's Internet Information Server. The software is capable of accessing and querying SQL (Structured Query Language) databases using a component known as the Internet Database Connector. The home page contains information about the mission, purpose, and future of Planetary Materials Curation, as well as pages for each of the four sample types - lunar samples, Antarctic meteorites, stratospheric dust, and space-exposed hardware. A TELNET link to the databases on the MicroVAX is also included.

Where do we go from here? First, we will move existing databases from the MicroVAX environment to a system based on client-server architecture (Fig. 2, Step 13). We will establish a database server which will utilize Microsoft's SQL Server software in conjunction with their Windows NT Server operating system. The classification, description, weight, and location information will be transferred from the VAX system to the SQL database.

Second, we will automate data-logging functions to eliminate intermediate paper records (Fig. 2, Steps 5 and 7). Numerical and text data will be typed into electronic Microsoft Word forms, and sample weights will be transferred directly from the balance to the form. Photographs will be taken using digital cameras, and the images integrated with the forms. The Word form will be imported in its native format into the File Magic document management system, and selected information will be transferred automatically to the SQL databases. The project to digitize the backlog of data pack material (Fig. 2, Step 6) will continue, and the Digital Data Pack Repository will ultimately consist of a combination of digitized images and Microsoft Word files containing both text and images. The curatorial team will access both the SQL database server and the Digital Data Pack Repository directly using applications running on the curatorial local area network. Public customers will access information using the WWW server, which will also act as a gateway to the SQL databases.

Plans are being developed to expand the Curator's web site to make the information in the sample catalogs available on line (Fig. 2, Step 14). The hardware used for scanning the reference data pack collection may be used in conjunction with Adobe's Acrobat software to scan previously published catalogs into files which will be stored as Adobe PDF (Portable Document Format) files. These files will be placed on the WWW server and made available for downloading. Users may read or print these files using Adobe's Acrobat Reader, which may be downloaded free of charge from the Adobe Web site.

Although the current discussion has mentioned lunar samples only, our true intention is that records for all planetary sample collections in our custody, including Antarctic meteorites and interplanetary dust particles, eventually will be handled by the same architecture that we are creating for the national collection of lunar samples. In fact, the paper data packs for many meteorites of exceptional interest already have been digitized and tests of automated sample weighing have been successful in the meteorite processing laboratory.

Both the magnitude of the project and the annual budgetary constraints that we face necessitate that our upgrades be spread over a period of a few years. Sample processing and allocations must proceed as we perform parallel activities to migrate existing data bases and introduce new automation equipment. Our aim for the year 1998 is that sample curation will have reduced by at least one-half its dependence on, and production of, paper for primary records and that electronic data packs will be available online for at least some of the high-interest samples.