Category Archives: Linked Open Data

Working to define “factoid prosopography”

The Department of Digital Humanities (DDH) (previously the Centre for Computing in the Humanities) at King’s College London first developed the idea of factoid prosopography back in about 1995. It grew out of a need then to think about how to do prosopography in a structured data (database) environment.  Since these early days, CCH/DDH has been involved in 6 prosopographical projects (latest finished in 2016) that consciously took up the factoid approach.  Furthermore, the idea of a factoid prosopography has generated interest from historians around the world who wish to do their own projects independently from DDH.

Although papers have been published on the idea of the factoid in prosopography (see Bradley and Short 2005 and Pasin and Bradley 2015), they provide only a rather high level overview of what the structural implications for a factoid prosopography might be.  Hence, this new site, developed by me, at:

http://factoid-dighum.kcl.ac.uk/

entitled Factoids: A site that introduces Factoid Prosopography.

Why have I created this site? I was not the originator of the idea of the factoid (this was developed by Gordon Gallacher and Dion Smyth in 1995, before I was at KCL).  However, I am the only one at DDH to have been involved in work (ranging from junior developer in the early days, to senior developer and then co-investigator more recently) in all six of the prosopographical projects that took a factoid approach and that involved DDH.  Based on this sense of continuity alone, it seemed natural for me to be the right person to describe the factoid approach as we have developed it at DDH.

The Factoids site contains three sections:

  1. a brief document that describes what I think factoid prosopography is all about,
  2. links to the various projects that CCH/DDH (and in particular, I) have been involved in that describe themselves as factoid prosopographies (along with two prosopographies done with the participation of CCH/DDH and myself that do not!), and
  3. a first attempt at a formal ontology (called the “Factoid Prosopography Ontology”: FPO) that is meant to capture what seems to me to be some of the important formalisms that have underpinned the CCH/DDH factoid prosopographies.

Item 3 deserves a brief explanation. This formal ontology is expressed as an RDFS/OWL ontology (and is, in fact, distributed through GITHub).  Why is this useful? Well, as one of the earliest thinkers about computer ontologies put it, an ontology is “an explicit specification of a conceptualization” (Gruber 1995, p. 908), and through it, as Noy and McGuinnes say, allows one:

  • To share common understanding of the structure of information among people or software agents
  • To enable reuse of domain knowledge
  • To make domain assumptions explicit
  • To separate domain knowledge from operational knowledge
  • To analyze domain knowledge (Noy and McGuinness, undated)

Prosopography data from different projects, by its very nature, is likely to gradually link together.  A structured data prosopography of Anglo-Saxon England is likely to connect through common persons and places with prosopographies for, say, Scandinavia that explore material of the same time period.  The more consistent the structure that they share, the more straightforward and stronger the connection that can be made between them.  As Gruber says, “[w]e use common ontologies to describe ontological commitments for a set of agents so that they can communicate about a domain of discourse without necessarily operating on a globally shared theory.  We say that an agent commits to an ontology if its observable actions are consistent with the definitions in the ontology.” (p 908)

By proposing FPO in the languages of the Semantic Web, RDFS and OWL, I was able to think of FPO as formulation of a formal core for factoid prosopography, a core that could be naturally expanded using the range of techniques that RDFS and OWL enable to meet the differing needs of various projects that implement it.

I think of FPO as still being rather preliminary, and as the title of this blog suggests, FPO should be most definitely thought of as work in progress. Indeed, for this reason I have assigned it a version number of 0.2 to it! Furthermore, although other projects can choose to commit to the view of factoid prosopography that FPO represents, DDH’s view of what formal structure enables and constitutes factoid prosopography, as presented in FPO, needn’t be the only possible view, of course.  Others are most definitely free to take up some part of the idea of factoid to suit the needs of their own project and yet implement a quite different approach to modelling their prosopography.  However, I think it fair to say that DDH has had perhaps the longest, and the most, experience with working successfully on the factoid approach, and for that reason alone it is worthwhile presenting, in some detail, what DDH’s views on these matters might be.  This is what the new Factoids site and the FPO prototype ontology are aiming to achieve.

References

Bradley, John and Harold Short (2005). “Texts into databases: the Evolving Field of New-style Prosopography” in Literary and Linguistic Computing Vol. 20 Suppl. 1:3-24.

Gruber, Thomas R. (1995). “Toward principles for the design of ontologies used for knowledge sharing?” In International Journal of Human-Computer Studies. Vol. 43, No. 5–6, November 1995, pp. 907-928.

Noy, Natalya F. and Deborah L. McGuinness (undated). “Ontology Development 101: A Guide to Creating Your First Ontology”. Online http://www.ksl.stanford.edu/people/dlm/papers/ontology101/ontology101-noy-mcguinness.html

Pasin, Michele and John Bradley (2015). “Factoid-based prosopography and computer ontologies: Towards an integrated approach”. In Digital Scholarship in the Humanities. Vol 30 No. 1, pp. 86-97.  published online June 29, 2013 doi:10.1093/llc/fqt037.

EpiDoc Workshop, London, April 20-24, 2015

We invite applications for a 5-day training workshop on digital editing of epigraphic and papyrological texts, to be held in the Institute of Classical Studies, University of London, April 20-24, 2015. The workshop will be taught by Gabriel Bodard (KCL), Simona Stoyanova (Leipzig) and Charlotte Tupman (KCL). There will be no charge for the workshop, but participants should arrange their own travel and accommodation.

EpiDoc (epidoc.sf.net) is a community of practice and guidance for using TEI XML for the encoding of inscriptions, papyri and other ancient texts. It has been used to publish digital projects including Inscriptions of Aphrodisias, Vindolanda Tablets Online, Duke Databank of Documentary Papyri and Digital Corpus of Literary Papyri, and is also being used by Perseus Digital Library and EAGLE Europeana Project. The workshop will introduce participants to the basics of XML markup and give hands-on experience of tagging textual features and object descriptions in TEI, identifying and linking to external person and place authorities, and use of the tags-free Papyrological Editor (papyri.info/editor).

No technical skills are required, but a working knowledge of Greek or Latin, epigraphy or papyrology, and the Leiden Conventions will be assumed. The workshop is open to participants of all levels, from graduate students to professors and professionals.

To apply for a place on this workshop please email simona.stoyanova@informatik.uni-leipzig.de with a brief description of your reason for interest and summarising your relevant background and experience, by Friday February 27th, 2015.

Linking Ancient People, Places, Objects and Texts

Linking Ancient People, Places, Objects and Texts
a round table discussion
Gabriel Bodard (KCL), Daniel Pett (British Museum), Humphrey Southall (Portsmouth), Charlotte Tupman (KCL); with response by Eleanor Robson (UCL)

18:00, Tuesday, December 2nd, 2014
Anatomy Museum, Strand Building 6th Floor
(http://www.kcl.ac.uk/campuslife/campuses/download/KBLevel6forweb.pdf)
King’s College London, Strand London WC2R 2LS

As classicists and ancient historians have become increasingly reliant on large online research tools over recent years, it has become ever more imperative to find ways of integrating those tools. Linked Open Data (LOD) has the potential to leverage both the connectivity, accessibility and universal standards of the Web, and the power, structure and semantics of relational data. This potential is being used by several scholars and projects in the area of ancient world and historical studies. The SNAP:DRGN project (snapdrgn.net) is using LOD to bring together many technically varied databases and authorities lists of ancient persons into a single virtual authority file; the Pleiades gazetteer and service projects such as Pelagios and PastPlace are creating open vocabularies for historical places and networks of references to them. Museums and other heritage institutions are at the forefront of work to encode semantic archaeological and material culture data, and projects such as Sharing Ancient Wisdoms (ancientwisdoms.ac.uk) and the Homer Multitext (homermultitext.org) are developing citation protocols and an ontology for relating texts with variants, translations and influences.

The panel will introduce some of these key projects and concepts, and then the audience will be invited to participate in open discussion of the issues and potentials of Linked Ancient World Data.

DH2014: ontology for 3D visualisation poster

DH_poster_valeria_LR

The poster was displayed at the first poster session of DH2014 on July the 10th. It gave me the chance to meet many scholars interested in my research and to discuss the project with them. I received very good feedback, and started useful dialogues that might turn into future collaborations with University of Mainz, UCLA and EPFL.

This project was presented as a paper at the last CAA conference in Paris.

 

SNAP:DRGN consultation workshop

Last week we held the first workshop of the SNAP:DRGN (Standards for Networking Ancient Prosopographies: Data and Relations in Greco-roman Names) project, here in King’s College London.

As announced in our press release the SNAP:DRGN project aims to recommend standards for linking together basic identities and information about the entities in various person-databases relating to the ancient world, with a view to facilitating the production of a federated network including millions of ancient person-records, compatible with the Linked Ancient World Data graph. At this workshop (see Workshop slides and recap) we presented our preliminary proposals, data models and ontology for feedback to a representative group of scholars from both the classical prosopography/onomaastics and Linked Open Data communities. We also spoke to several people with large datasets that might be contributed to the SNAP graph.

It was decided that SNAP:DRGN will attempt to address recommendations to five key use-cases of networked prosopographical data:

  1. Putting prosopographical data online, including stable URIs and openly-accessible data and metadata in standard formats (not defined by us).
  2. Contibuting a summary of said data, including identifiers for all persons and a simplifed subset of core identifying information about each entity, to the SNAP graph to that it can be built upon and referred to by other projects.
  3. Annotating SNAP entities to establish alignment and identify co-references between related datasets.
  4. Marking-up online documents to identify personal names within them to persons identified in the SNAP graph and its constituent databases.
  5. Adding relationships between persons, both within and between databases: person X is the daughter of person Y; person A in one database was killed in battle by person B in another database.

The SNAP:DRGN project will continue to work on the “Cookbook“, the summary of recommendations and examples for these five use-cases, over the coming months, in the run-up to adding several new datasets to the graph. We are also experimenting in a modest way with tool and implementations for working with the vast graph of ancient persons created: named entity recognition (NER) workflows for finding new personal names in texts; co-reference resolution for finding overlap and links between datasets; search and browse tools and APIs. This work will be reported on the SNAP:DRGN blog, and in conferences and seminars throughout the year.