-
Notifications
You must be signed in to change notification settings - Fork 2
LOD People Charter
The aim of the LOD People Activity is to survey practice and collect guidance in the handling of interoperable Linked Open Data relating to historical person datasets. Rather than replacing existing solutions around shared person data formats, we recommend lightweight, minimal standards for interchange data, compatible with Pelagios registries, APIs and annotation tools. Needs and solutions will arise from consultation with communities of practice and stakeholders, in close collaboration with other PN activities. Results and discussions are disseminated openly throughout via online platforms, forums and repositories.
The very large number of prosopographical databases and personal name datasets is indicative of how fundamental consistent information about people is to all periods of historical research, from ancient to early modern. People indeed sit alongside places and time periods as primitives of historical data. Searchable databases (and the even more numerous indexed print catalogues) are individually valuable, but the divergence of data models and formats makes cross-project sharing and search impossible.
Large scale data-driven research on ancient and historical persons would be enabled by consistent and common data infrastructure. Several projects have attempted to disambiguate or federate historical person-data in LOD, using formats such as SNAP, Wikidata Ontology, CIDOC-CRM or IPIF, but none of these have achieved the traction of the Pelagios or Linked Places formats for place.
The LOD People activity proposes to address the clear hunger for recommendations and standards in this area, and the interdependence of data about people, place, time, sources, events and descriptors, by engaging existing projects and stakeholders—many of which have been engaged with the Pelagios Network for years already—in discussion around these key themes. Our modest goals are to propose: a survey and overview of existing projects and suggested solutions to federation of historical person data; minimal standards as a starting point for sharing structured person data; one or more interchange formats, registries or APIs; a community of practitioners and researchers to keep the conversation as broad and inclusive as possible.
The LOD People activity does not aim for a specific solution to fit all possible needs. It is understood that any data model provides means for expressing relationships between people as well as between people and other entities. Many of these relationships will be mediated through events; generally also through sources. Some data, relating to people from recent times, including data on relationships may be withheld from publication in open repositories for ethical or legal concerns. We do not propose to replace the existing approaches and ontologies (many of which we will not know about until a better survey of the field).
Rather, a preliminary objective is the development of a simple interchange format (with similar goals to the Linked Places Format) to facilitate interoperability, both by allowing transfer/merging of complete resources, and by defining a “Linked Data Fragment” format which can be used in distributed database queries.
Further objectives may emerge from discussions and comparisons of practice, but the scope will be carefully limited to align with Pelagios Network priorities.
The Activity will begin with a series of meetings and asynchronous activities (to cater for the wide range of timezones represented in the community). The first concrete goal will be the survey of needs and existing solutions, to be followed by discussion of the feasibility of aligning and proposing interchange between these and other approaches. Further actions, including the development and documentation of concrete LOD recommendations and best practices, may arise from these discussions, and will be further specified once the activity is more mature.
The Activity will also arrange more intensive activities as part of Linked Pasts symposia and other short-term or in-person events. Results will be disseminated as early and as frequently as possible, via the activity wiki and repository (https://github.com/DigiClass/LOD-People) and occasional more formal publications (on which see “Communication” below).
The principal participants in this activity will be academics and heritage professionals who work with data on historical persons in a digital or quantitative context. Many existing projects will also contribute to activity priorities as part of their ongoing research, as resources permit. Projects and communities that we are most eager to engage with include:
- SNAP:DRGN
- Connecting Late Antiquities
- Trismegistos
- Lexicon of Greek Personal Names
- Prosopography of the Byzantine World
- MANTO
- Chinese Biographical Database Project
- Historical Network Research Community
- ToposText
- Wikidata
- ORCID
This activity will involve liaison and collaboration with many of the PN activities, in particular:
- Annotation: As with Pelagios itself, a key motivation for most person LOD projects is the development of a virtual authority of historical people for annotation of primary and secondary sources. Preliminary work on a person-annotation fork of the Recogito tool was started several years ago (see https://github.com/DigiClass/ProsopoCogito), and discussions for taking this forward would be valuable.
- Gazetteers: The importance of shared formats and lessons between place datasets and person datasets will be key to the development of interchange recommendations. As mentioned above, person data inevitably includes place references.
- Registry: Prosopographies and other person datasets are arguably even more decentralised and in need of federation than gazetteers, so registry services will be absolutely essential to the success of work in this activity.
- Time: Date (in its various forms) represents one of key disambiguations for all person data. Conversely, people and prosopographical sources are often key to identifying time periods, dating events, relationships, and even political identities of place,
- Visualisation: A complex and challenging research question will be how to visualise person data, both spatially and otherwise. New axes of visualisation may need to be explored, such as networks and network analysis, genealogies and other relationship types, or graphing and tabulating person characteristics.
The primary venue for communication of this activity will be the “Ancient People” mailing list (for which all LOD and other digital approaches to historical person data is on-topic, despite the more limited group title): https://groups.google.com/g/ancient-people
Reports, agendas and other documents and data will be shared via the LOD People wiki and repository: https://github.com/DigiClass/LOD-People/wiki
Periodic reports and requests for feedback will also be disseminated via the PN channels and meetings, at LP symposia, and other real-time events or conferences.
Occasional stable recommendations will be issued and disseminated more widely, perhaps via external sites such as Zenodo or institutional repos, which can assign a DOI. More formal authored publications are also possible eventually, but not currently planned.
There should be no financial implications of this activity in the near term; all communication will be carried out using cloud and conferencing technologies. Attendance at LP and other conferences would be the responsibility of individual participants. New funded projects may arise from some of the collaborations facilitated by this activity, but any financial considerations of such work will be the responsibility of scholars and research departments involved.
All work directly produced by or in association with this activity, whether recommendations, documentation or software, will be released under open licenses and as Linked Open Data. Many other tools and data involved are licensed by individual projects and therefore decentralised; the activity will respect licensing rules and authorship attribution of all such collaborative and shared work.
This activity will focus almost entirely on datasets of (and data fields relating to) historical persons. We caution anyone working with data on living or recently living persons to further consider the ethical and legal implications of collecting and sharing this data (privacy, data protection, freedom of information, reputation, right to removal, etc.). This would include open access data such as Wikidata records relating to living people.