Data Model and API

Once operational, MDMP will need to be able to gather and collect various kinds of metadata, including information about published texts or unpublished curriculum authored by women; metadata that helps locate archival ephemera or uncirculated work by, for or about women pedagogues; citational references to their work and/or immaterial evidence of their administrative activities; and most importantly, user-contributed information–such as information generated from individual researchers’ notes or tagged from archival finding aids–reflecting the various ways that their pedagogical activities have moved through our intellectual consciousness, or the reasons why they have become the focus of our research.

In order to do all of these things, MDMP’s Application Programming Interface (API) must be able to accommodate various metadata formats–from linked open data (LOD), to spreadsheets, to individual records created from contribution forms–and to draw equally from them when making visualizations and building ontologies. In its early phase, MDMP is supported by a data model that privileges a cultural-heritage orientation toward information. It identifies intellectual migration as something that occurs through a series of fluctuating (non-static) relationships between physical agents–such as documents and their references–and  conceptual activities–such as researchers’ motives and queries–all operating within the same domain.

Data Capture

In order to accept a wide range of contributions from participants, MDMP will populate in two ways:

      1. Manual entry via a web-based form for small amounts of information about the circulation of a particular pedagogue, curriculum, or text, including the researcher’s motivations for finding it; and
      2. Batch ingestion of bibliographies and lists (Zotero, Endnote, CSV/Excel spreadsheets, or Word documents) for greater amounts of information about the long-term status of a particular institution, curriculum, pedagogue, or text. Word documents will be manually extracted and placed into a notes section, or uploaded and linked through a notes section.

MDMP is also exploring the ability to connect to other open data sources, such as OCLC’s ArchiveGrid.

At the present time, a set of experimental metadata dictates the parameters of MDMP and determines the kinds of “events” it can show; however, the tool is only as useful as its contributors and users. If you would like to contribute metadata to help shape its parameters as we finish its construction and before we launch, please feel free to do so using our Contribution Form. We welcome and rely on your contributions. 

Data and Metadata Standards

Because MDMP seeks to visualize relationships and encounters between data and their metadata, it will use open non-proprietary data storage formats such as MySQL, and JSON (JavaScript Object Notation) to make various data sets immediately available. Complete representations of mapped ecologies (places where users can see the distribution and location of textual encounters, both historically and contemporarily), ontologies (a preliminary organizational system that organizes and affiliates scholars, humans and location), and user contributions will be published via an API that supplies JSON to other platforms and projects. Furthermore, MDMP will use open-source tools and modules, such as Leaflet, the jQuery data-visualization library D3, and minimal PHP code, to manage its internal application data representations.

Metadata Sharing

Because MDMP aims to construct user-based ecologies of women’s intellectual work, descriptive metadata will be voluntarily created as part of the process of contributing records via the user submission form. MDMP is committed to open access and will be licensed under the Creative Commons BY-NC-SA 3.0 License, allowing the metadata from individual text records to be exported and used non-commercially by other sources. In practical terms, this means that–by contributing to MDMP–users may voluntarily contribute information about their institutions, professional affiliations, scholarly interests, and motivations. When offered, this metadata becomes a part of the data record and thus a part of the historical ecologies that other researchers will query or search.

Metadata Integration

MDMP will rely on the kinds of data schemas and taxonomy structures that are already featured in mapping APIs such as TileMill and REST, but it will integrate what are commonly understood to be separate mapping and mining functions, to generate user-based relationships between locations and topics. We are especially interested in investigating alternative landscapes and (non-map-based) visualizations.