Learning From Implementation

As a data management tool, DINAA has the potential to transform the way we think about and conduct basic archaeological research, data and heritage management, and public education in the United States and beyond. The integration of site file data at continental scales in an open and readily accessible informational infrastructure allows, for the first time, the exploration of the North American archaeological record across multiple temporal periods and geographic regions. The utility of such a resource was directly seen in the 2017 PLOS ONE paper discussed previously, that showed the effect of sea-level rise on known archaeological sites and properties listed as eligible for inclusion on the National Register of Historic Places.¹⁶ For the first time, the entire site database from a substantial portion of the Gulf and Atlantic coasts of the United States was examined, showing the scale of the problem, and making recommendations for management and mitigation (Figure 9.4).

However, DINAA’s greatest value for museums, libraries and the public centers on Linked Open Data (LOD) applications. Open Context, like other LOD systems, emphasizes the use of stable Web Uniform Resource Identifiers (URIs, i.e., stable URLs that serve as universally unique “primary key” identifiers) to identify concepts and other entities so they can be easily and precisely referenced and related across different data collections on the Web. DINAA uses Open Context and the EZID service to mint persistent URIs for each site files record. In archaeology and historical geography, the “site” is a key organizational entity. Minting stable Web URIs and offering rich temporal, geographic, and cultural metadata (also available in machine- readable JSON-LD format) about sites will therefore create significant LOD resources essential for broadly integrating museum, library, and scientific datasets.

DINAA cross-references diverse museum, library, and archival resources. However, using DINAA involves several challenges for organizations that may have limited technical support and staffing. In our experience, efforts to reuse data offer some of the best ways to discover problems in data and data services.¹⁹ To build experience needed to guide uses of DINAA, we have undertaken the following activities: ^[1]

published online journals and other data sources. An initial test found numerous references to archaeological sites in back issues of American Antiquity. In 2019, we began asking state information managers for lists of sites by county in their state, identified by their Smithsonian trinomial codes, so we can begin indexing items even in the absence of specific site information from their office. Ten states responded positively to this request, with the result that the DINAA team now has well over one million site numbers, of which 881,243 are already indexed.²⁰

• 2. Linking to external collections: We established cross-referenced linkages between DINAA and the Federal Register, the primary source of US government regulatory determinations. The Federal Register references archaeological site records, and DINAA provides a powerful index illustrating the geographic and chronological scope of US government cultural heritage management. Similarly, in a collaboration with the Phoebe A. Flearst Museum of Anthropology, we indexed some 5,000 sites in California available in the Hearst’s public online collection. DINAA can now more easily cross-reference with additional Hearst records because of improvements in the Hearst’s collections management system resulting from this project. Furthermore, because the Hearst Museum’s collections are documented in CollectionSpace, an open source museum information system that is used by several institutions that also curate significant North American collections (most notably the San Diego Museum of Man), extending CollectionSpace to use DINAA, as well as documenting cost- effective implementation methods, will help future institutional partnerships. While the technologies for connecting DINAA to other resources now function well, the greater challenge of determining what resources should be linked remains (see following).
• 3. DINAA’s Linked Open Data approach through Open Context cross-references distributed collections on the Web, enables users to find and access relevant data in other online datasets using site numbers as the common identifier. To date, these linkages include:

a. Links to and from tDAR: DINAA cross references site records with tDAR metadata records. Open Context uses this information to interface with tDAR’s API to display links to tDAR-archived reports and data relevant to site records in DINAA. Additionally, tDAR recently enhanced its spatial metadata records to include DINAA site file record URIs. This marks a major development in interoperability between American archaeological information systems.

b. Eastern Woodlands Household Archaeology Database Project (EWHADP):²¹

Andrew White, a researcher investigating household structures for Woodland period sites in the Midwest and South, has incorporated DINAA identifiers in his online database. In doing so, his datasets are precisely related to DINAA data, and the DINAA search interface can be used to discover data about ancient households compiled and curated through his research efforts.

c. VertNet/GBIF:²² VertNet is a major contributor to the Global Biodiversity Information Facility (GBIF). VertNet has started to use DINAA identifiers and data as spatial metadata for zooarchaeological specimens. This means DINAA now helps support research and information management in a much broader world of bioinformatics systems.

d. Pelagios: Pelagios aggregates gazetteer data and annotations that link cultural heritage content to gazetteers. In order to more broadly disseminated DINAA data and annotation, Open Context implemented Pelagios-recommended Linked Open Data standards so that DINAA data are now discoverable via the Peripleo-Pelagios network.²’ This broadens the community of researchers and software developers working with DINAA data and annotations.

As discussed, DINAA primarily serves as an index and finding aid that helps make resources scattered across the web easier to discover. But even in this role, where DINAA relies upon partner institutions to properly and ethically curate collections, DINAA runs into ethical challenges. For example, at the workshop in August 2019, DINAA showed examples of linking site records to resources in the Hearst Museum’s online collection. An Indigenous archaeologist workshop participant followed links from DINAA to the Hearst’s online collection and saw materials inappropriate for public display. After raising this problem with the museum, the Hearst quickly took down the inappropriate materials. This incident highlights some issues. DINAA helped facilitate identification of materials that should not be public, leading to the improved curation of the Hearst collection. At the same time, however, this incident raises important questions about DINAA’s role in linking to outside resources. What resources should DINAA point toward? DINAA can and should link to public online resources maintained by Tribal Nations, but many tribal communities do not have much of an online presence. How can DINAA help highlight Indigenous voices documenting their own heritage while avoiding inappropriate or even abusive resources? Furthermore, DINAA has a continental scope and scale. Most programs that can serve as a model for good ethical practices in working in partnership with Indigenous communities occur on a local scale. How can such models for partnership work to responsively meet the needs of several hundred diverse sovereign Tribal Nations across all of North America?

• [1] Expansion through text-mining: As a way of expanding DINAA’s indexingcapabilities, and adding site information from areas where we do not yet havesite file data, we developed text mining software to find references to sites reportedusing a Smithsonian/River Basin Survey trinomial site numbering format in