Linked Pasts 11 programme

Uncategorized

Linked Pasts Symposium 11 (online), December 1-12, 2025

The Linked Pasts Symposium is a goal-oriented meeting at which you can build, plan or learn about the application of linked open data (LOD) to historical texts, events and data. The following activities are offered in 2025: to participate, sign up for any individual activities or events that interest you.

Programme

  1. Chronique en ligne/AG Online: First geodata hackathon (Catherine Bouras, Eleni Gkadolou). 2nd–4th December 2025.
  2. Documenting Epigraphic Data in Wikidata: From Ontology to Practice (Anna Clara Maniero Azzolini, Maxime Guénette, Emily Helm, Anne Hunnell Chen). 2nd–3rd December 2025.
  3. Enriching digital heritage with LLMs and Linked Open Data (Gethin Rees, Elton Barker, Sarah Middle, Anna-Maria Sichani, Mia Ridge). 9th–10th December 2025.
  4. Linking Knowledge Through Place: ISHI, WHG, and the Future of Gazetteer Collaboration (Ruth Mostern, Stephen Gadd, Alexandra Straub, Palak Vashist). 9th December 2025.
  5. Pelagios Network Annotation Working Group (Elton Barker, Anne Chen). 3rd December 2025.
  6. Pelagios Network People Activity: Sample Data Hackathon (Gabriel Bodard, Jun Ogawa). 1st–4th December 2025.
  7. Promoting Collaboration, Linking Communities: building Pelagios connections through the Digital Classicist Wiki (Tom Gheldof, Sarah Middle). 1st–8th December 2025.
  8. Reconstructing events in time and space (Yu Lee An, Ryan Shaw). 10th December 2025.
  9. SNAP:DRGN secondary recommendations: TEI XML and Wikidata (Gabriel Bodard, K. Faith Lawrence, Camillo Carlo Pellizzari di San Girolamo, Charlotte Tupman). 1st–4th December 2025.
  10. Wikidata/CIDOC property mapping: discovery and planning (Anne Chen, Kimiko Adler, Katherine Thornton, Florian Thiery, Daria Stefan, Maxime Guénette). 2nd–3rd December 2025.
  11. Writing SPARQL queries with GitHub Copilot (Duncan Hay). 12th December 2025.

Organising committee

  • Gabriel Bodard
  • Leif Isaksen
  • Sarah Middle
  • Jun Ogawa

La Chronique en ligne/AG Online: First geodata hackathon

Convenors: Catherine Bouras (EFA) and Eleni Gkadolou (BSA)

This activity aims at presenting the organisation of places in Archaeology in Greece Online, a combination of modern places, ancient places and research sites and offering an opportunity to participate in the collaborative curation of what seeks to become a community tool.

This workshop invitation, that can also fall under the PN Gazetteer Activity but is also fairly freestanding, concerns the geographic data of Archaeology in Greece Online/La Chronique des fouilles en ligne. This open access online database, which was first developed and released in 2009, collects and presents on a regular basis recent archaeological fieldwork reports to academic researchers and to the interested public. It has a powerful search engine that allows advanced research through regions, toponyms, keywords, periods, etc.

This activity intends to present the platform and offer an overview of the tools that are available to feed and enrich it as well as to cross-reference the data. Participants will learn directly from the curators of the database how the various entries evolve through a double system of geographic information, how it dialogues and interoperates with similar archaeological and geographical tools, placing it in a linked open data environment.

Because the platform was created in 2009, AG Online has a long experience in geolocating its reports and more importantly perhaps, in improving and enriching these geodata. Since it was first launched, there have indeed been significant updates in functionalities and in precision. Earlier entries have not been systematically updated to match this progress. The more AG Online interoperates and is linked to other similar tools, the more its resources need better curation. Participants will have the opportunity to contribute to a cutting-edge collaborative (community?) tool as one of the goals of this activity will seek to curate geodata that AG Online uses and creates in order for it to better align with Pleiades, Topos Text, the Archaeological Cadastre and so on. Here are some of the case studies that this activity may take on:

  1. Archaeology in Crete: untangling toponyms and organizing ancient sites resources.
  2. Archaeology in big cities: the example of Athens (granularity/micro-gazetteer)

The activity will be organized as an online “hackathon”, via zoom and will bring together participants who will have the opportunity to connect to the platform in smaller breakout groups and discuss and/or directly curate reference data/resources, after a presentation of the platform and its functionalities. We hope to start the activity on December 2nd and close by Thursday 4th of December at which point we can regroup and present progress and conclusive remarks. The result of this collaborative work will appear shortly after on AG Online and there will be a full report after the symposium.

Sign up information:

Documenting Epigraphic Data in Wikidata: From Ontology to Practice

Convenors: Anna Clara Maniero Azzolini (University of London), Maxime Guénette (Université de Montréal), Emily Helm (Yale University), Anne Hunnell Chen (Bard College)

The activity consists of two sessions of one and a half hours, on Tuesday 2nd December at 14:00-15:30 GMT and Wednesday 3rd December at 14:00-15:30 GMT.

The first session will serve as an introduction to Wikidata, presented as a shared environment for interdisciplinary development. The session will outline the database’s data structure and its underlying ontology, in order to clarify key concepts that will be revisited later. Once participants have gained a foundational understanding of Wikidata’s ontological framework, brief examples from the speakers’ projects will be provided, followed by a discussion on its application to epigraphic records — specifically, how the Wikidata model has been adapted for the documentation of ancient inscriptions. The IDEA (see data models for inscriptions and papyri) and Altinum projects will be showcased as illustrative case studies, offering a window into the laboratory experience that participants will engage with during the second session.

The second session will be a hands-on workshop, designed not only to allow participants to apply what they have learned but also to facilitate a final discussion on the challenges encountered. Participants will be asked to work on an inscription of their choice (or one proposed by the speakers) and to create a corresponding item in Wikidata, adding key properties. In the editing process, each speaker will coordinate a group of participants. At the conclusion of the editing activity, an initial draft of the complete epigraphic data modelling in Wikidata will be presented, providing participants with the opportunity to discuss it, offer suggestions, or pose questions.

Those wishing to edit must have created a Wikidata account at least 24 hours in advance. However, the session will also be open to participants interested solely in the discussion component, which is a necessary moment for the development of new ideas and also allows the speakers to collect impressions, reflections, and suggestions from the participants who will contribute verbally, via chat, or occasionally by generating online word clouds.

To join this activity: https://www.sas.ac.uk/digital-humanities-research-hub/events/documenting-epigraphic-data-wikidata-ontology-practice

Further information will be recorded at: https://www.wikidata.org/wiki/Wikidata:WikiProject_Epigraphy/Linked_Pasts_11

Enriching digital heritage with LLMs and Linked Open Data

Convenors: Gethin Rees (King’s College London), Elton Barker (The Open University), Sarah Middle (Archaeological Data Service), Anna-Maria Sichani (University of London), Mia Ridge (British Library)

Combining Large Language Models (LLMs) with Linked Open Data (LOD) offers great potential to make cultural heritage metadata FAIR. The proposed activity will bring together specialists from cultural heritage, data science and digital humanities to provide feedback on a practical application of this approach over two sessions. Cultural heritage practitioners hear a lot about the transformative potential of AI, but opportunities to get hands on with cultural heritage collections are rare. They might also be aware of the problems of AI, especially LLMs, but would like to understand more about limitations and their relevance to heritage.

The Cultural Heritage AI Cookbook, offers a practical method of enriching metadata. Funded by the Netherlands eScience Center and the Lorentz Center@lambda, the convenors and twenty others created a set of Google Colab Notebooks for the recognition, disambiguation and relations of named entities using LLMs. The next step is to integrate LOD more deeply within the cookbook to address issues with the accuracy, validity and biases of outputs.

This activity will be divided into two sessions, scheduled for 9 Dec at 4pm UTC and 10 Dec at 9.30am UTC. Please sign up for further information. The first examines the cookbook as it currently stands, providing an introduction, practical examples of its use and an opportunity to provide feedback. The output will be a feedback document used to update the cookbook. The second session will focus on future directions. Discussion will encompass the challenges of embedding the cookbook, the approach, and AI more generally in cultural heritage practice. The activity will offer an opportunity for cultural heritage professionals to express their interest and concerns. This second session aims to draw up a plan in the form of a pathway towards impact encompassing how this might be used in cultural heritage organisations and possibilities for enriched FAIR data to develop new forms of engagement with collections.

Sign up to the activity: https://forms.gle/DErqFUuY2Vn79X3t6

Details of the cookbook can be found here: https://pelagios.org/llm-lod-enriching-heritage/intro.html

The notebooks are available on this Github repo https://github.com/pelagios/llm-lod-enriching-heritage

The notebooks will be updated based on feedback and made available openly. Other outputs such as updates to explanatory text, descriptions of the notebooks and plans for future work will be added at the above links.

Linking Knowledge Through Place: ISHI, WHG, and the Future of Gazetteer Collaboration

Convenors: Ruth Mostern (University of Pittsburgh), Stephen Gadd (World Historical Gazetteer), Alexandra Straub (Pittsburgh), Palak Vashist (Pittsburgh)

Tuesday, December 9 15:00 GMT–16:30 GMT.

This activity focuses on advancing collaborative methods for historical place alignment across gazetteers and related datasets within the Pelagios community. It brings together members of the Institute for Spatial History Innovation (ISHI), the World Historical Gazetteer (WHG), and the Pelagios Network to develop shared conceptual models, metadata practices, and documentation that support this work. Historically, alignment has focused on matching discrete identifiers or reconciling lists; this activity reframes alignment as the creation of linked, attestive ecosystems capable of representing how places are named, described, bounded, and understood across time, language, authority, and cultural tradition. Building on the forthcoming Linked Places 2.0 graph model, we aim to surface community priorities for modeling temporal scoping, variant semantics, overlapping geometries, uncertainty, and source-level provenance.

Our agenda recognizes that alignment is both conceptual and infrastructural. Participants will explore how meaning is produced at the level of attestations and how richer relationships can support comparative research across regions and periods. Asynchronous lightning presentations will be shared across the two-week LP11 period, followed by collaborative documentation efforts, discussion threads in the LP11 Slack workspace, and optional synchronous working sessions scheduled to accommodate diverse time zones. Planned outputs include a revised charter for the Gazetteer Alignment Activity, community-authored documentation on best practices for relationship modeling, and a contributor readiness checklist that supports smaller projects preparing their place data for integration. Participation is open to researchers, developers, heritage practitioners, and community contributors of any dataset scale, regardless of prior experience with alignment workflows.

How to participate: Participants may sign up using a short online form: https://forms.gle/dzSq2GibT17nckim6

Information and outputs: Programme details, documentation drafts, and working group activities will appear on the ISHI website and associated GitHub spaces. Documentation will remain available after the symposium to support ongoing community engagement into 2026.

Pelagios Network Annotation Working Group

Convenors: Elton Barker and Anne Chen

This activity represents a regular meeting of the Pelagios working group on Annotation. The use of semantic annotations lies at the heart of the Pelagios vision of lightweight approaches to data connectivity. Specifically, they are intended to be easily produced conceptual bridges between independent online resources. The benefits range from mutual contextualisation to the derivation of geographic coordinates for automated mapping. Nonetheless, a variety of technical and social barriers impede their creation – from consensus on digital formats to the challenges of identifying the relevant entity in a gazetteer. The Annotation Working Group is dedicated to reducing barriers through consensus building, the development of best practice, and the implementation of assistive software.

Partners of this Pelagios Working Group range from tenured academics to PhD students, from DH researchers to curators or software developers. We meet every month at a regular office hour (Stammtisch) to hear from Partners, host guest talks, or discuss particular issues. For this meeting in December, we will hear from Rocio Da Riva on her project RelNet: religious networks in Late Babylonia. Using I millennium BCE Babylonian cuneiform ritual texts as sources and Nodegoat as a research platform, RelNet scrutinises sacred landscapes and the connections between Babylonian cult centres, analysing temple festivals, religious journeys and their participants and venues. While cuneiform texts offer infinite possibilities for this, they also present a series of issues and challenges related to both the nature of the sources and the methodology.

To sign up to the activity: email the convenors: annotation@pelagios.org

Pelagios Network People Activity: Sample Data Hackathon

Convenors: Gabriel Bodard (University of London); Jun Ogawa (University of Tokyo)

December 1 (Mon), 12–14 GMT & Dec 4 (Thur), 11–13 GMT. Online only.

This activity is a hands-on session exploring some sample person-data (extracts of between 100–1000 records per project in a variety of formats) collected by the Pelagios Network (PN) People Activity.

Participants will be introduced to sample datasets (https://github.com/DigiClass/LOD-People/tree/main/sample-data) with the option to clone or download the repository, or explore the metadata. The goal of the session is to explore alignments and interoperability between datasets. Two working groups will approach this via different methods:

  1. Individually look at the data and make note of any links or alignments between them, including shared terms or vocabularies, common persons or other entities (facilitated by Gabriel Bodard)
  2. Experiment—through hands-on activities—with processes such as converting sample data into Wikidata-style RDF, performing data reconciliation, and storing and querying the data in a triplestore (facilitated by Jun Ogawa)

Considering these tasks, the range of background and expertise expected is quite wide, including participants with various backgrounds from those with interest in cataloguing or prosopography to those familiar with more technical aspects of Linked Data (e.g. OpenRefine, triplestores, SPARQL, etc.).

These two groups will then reconvene to discuss and document some of the alignments and reconciliations identified. Discussions of other aspects of the PN People Activity are also expected, and everyone interested in the topic is welcome regardless of level of technical expertise and areas of knowledge.

The anticipated outputs of this activity (such as sample datasets and RDF triples) are planned to be published in our GitHub repository or other triplestores, and the activity itself is also scheduled to be reported as an article for the Stoa Review.

Sign up for this activity: https://www.sas.ac.uk/digital-humanities-research-hub/events/pelagios-network-people-activity-sample-data-hackathon

  • The activity will be discussed and planned on Ancient People mailing list
  • Programme will be updated on the LOD People Wiki
  • Outputs will include new data posted to LOD People git repo
  • A follow-up event will be held at Linked Pasts Japan II (Feb 2026)

Promoting Collaboration, Linking Communities: building Pelagios connections through the Digital Classicist Wiki

Convenors: Tom Gheldof (KU Leuven) & Sarah Middle (Archaeology Data Service)

The Digital Classicist Wiki (DC Wiki) is a community-authored resource describing digital projects, organisations and tools, relating to the global ancient world. Pages in the wiki are grouped by thematic categories to facilitate discoverability and serendipity, as well as forging connections. Contributions are made by volunteers on an ongoing basis, with monthly topic-based sprints being a particular catalyst for activity. While many Pelagios Network Partners already appear in the DC Wiki, some pages are out of date and there are currently few links in the wiki between them and the Pelagios Network itself.

In this sprint we will aim to improve this situation by:

  • Enhancing existing content: updating pages with current information and applying the ‘Pelagios Network members’ category to create a means of easily grouping and accessing all Pelagios Partner entries;
  • Adding new content: using an amended version of the existing ‘To do list’ to highlight Pelagios Partners who  do  not  yet  have  a  DC  Wiki  page  and  prioritising  them  for  creation.

There will also be scope for an alternative/bonus task to add and/or update pages in the ‘Linked open data’ category, a theme relevant to both Pelagios and Linked Pasts. For those new to the wiki, tutorials and documentation will be made available in advance, with all resources linked from the associated Google Doc. Support will also be provided during two synchronous sessions, to optimise coverage for multiple timezones:

  • Monday 1 December, 3.30-5pm GMT
  • Monday 8 December, 8.30-10am GMT

This activity will not only enrich the quality and connectivity of the DC Wiki, but will showcase the diverse projects and institutions affiliated to Pelagios, making their information more discoverable, reliable and current. We additionally hope that this activity will result in the extension of the DC Wiki community and inspire participants to attend future sprints.

Participants can sign up to one or both of the synchronous sessions using this Google Form. The form includes a question about whether the participant already has a DC Wiki account; if not, we will create one in advance and send details via email. All information about this activity will be added to this Google Doc.

Reconstructing events in time and space

Convenors: Yu Lee An and Ryan Shaw, University of North Carolina

Wednesday, December 10, 10:00 AM – 1:00 PM (US Eastern Time, EST).

  • US Pacific Time : 7:00 AM – 10:00 AM (PST)
  • United Kingdom (GMT): 3:00 PM – 6:00 PM
  • Germany (CET): 4:00 PM – 7:00 PM

The goal of the activity is for participants to learn about tools and techniques for modeling, reasoning about, and visualizing historical events, using as an example Yu Lee’s dissertation work on creating a knowledge graph of the British music trade from the late eighteenth to the early nineteenth century. People, places and events from the British music trade are modeled by hand using the CIDOC-CRM ontology. These assertions are then augmented with inferences made using rules written in Notation3 and the EYE reasoner. A major goal is to model events so that qualitative temporal relations can be inferred from partial information about dates and times. We plan to meet one time, via Zoom, for three hours with a coffee break in the middle. After an introduction and demonstration of Yu Lee’s work, we will lead participants through the process of modeling events using the CIDOC CRM in Google Sheets, transforming it into RDF, enhancing the RDF with rules-based inferencing, and visualizing the results. Our intention is for this to lead naturally into a group discussion of the interpretive nature of historical data modeling, balancing conservative approaches that capture only what is explicitly stated with bolder ones that infer implicit information, such as likely participants in court proceedings. Such modeling choices raise recurring challenges in representing presence, agency, and documentation within the CRM, where distinctions between active and passive roles (e.g., ecrm:P11_had_participants vs. ecrm:P12_occurred_in_the_presence_of) highlight broader semantic issues.

Sign up via this google form: https://docs.google.com/forms/d/e/1FAIpQLScbBjMIBD8I_gVfJ0THIuhrjCWcRXKeduSltFks026c-9XJHg/viewform

We will create a repository and web page for the activity under the Distributed Knowledge Graph Lab Github Organization: http://github.com/dkglab/

SNAP:DRGN secondary recommendations: TEI XML and Wikidata

Convenors: Gabriel Bodard (University of London), K. Faith Lawrence (The National Archives), Camillo Carlo Pellizzari di San Girolamo (SNS Pisa), Charlotte Tupman (University of Exeter)

We will run a short workshop to build on the recommendations of the SNAP Cookbook <https://github.com/SNAP-DRGN/Cookbook/wiki>, which specifies a light RDF model for encoding an interoperability focussed subset of person data from digital prosopographies, catalogues and similar datasets. SNAP recommends a short list of essential fields that are common to many person records: URI, type, citation, collection, names, attestations, identifiers, place, date, titles, events, gender, relationships.

After a short introductory video meeting on Monday 1st December at 14:30 GMT, participants in this workshop will invited to contribute to one or both of two strands:

  • Wednesday 3rd December, 15:00 GMT
    SNAP in TEI XML. This session will map the key features of a SNAP person record to elements in the TEI Personography module, including the reference to URIs for entities and vocabularies where these exist. This recommendation will cater to projects that already use TEI XML for their prosopographical or person authority data, or individuals who know TEI better than they do RDF, and open a pathway to producing SNAP RDF via shared XSLT stylesheets (partly already in existence).
  • Thursday 4th December, 15:00 GMT
    Wikidata properties, classes and data models. We will start the process of mapping the key elements of SNAP interoperability data to a person record in Wikidata, primarily by identifying (or in some cases perhaps proposing new) properties and other Wikidata URIs for these features of a person record. The goal is to enable maximum compatibility with existing datasets in what is already a massive, valuable interoperability medium.

Both of these activities will be held via Zoom calls, with shared docs, Git repos and Wiki also available for final outputs of the workshop. After the initial meetings, further discussions may follow in the new year as required.

Register for this activity: https://www.sas.ac.uk/digital-humanities-research-hub/events/snapdrgn-secondary-recommendations-tei-xml-wikidata

Where to find out more about this activity:

  1. Working document at https://docs.google.com/document/d/1bM6OIAri46szls4w_1dOExU7DsMfODkjXN8GDh3x-Uc
  2. Eventual inclusion in the SNAP Cookbook <https://github.com/SNAP-DRGN/Cookbook/wiki>
  3. Announcements and discussion on the Ancient People discussion group.

Wikidata/CIDOC property mapping: discovery and planning

Organizers: Anne Chen, Kimiko Adler, Katherine Thornton, Florian Thiery, Daria Stefan, Maxime Guénette

Kick-off meeting: Tues. Dec. 2, 11am EST/4pm GMT
Summary discussion/planning next steps: Thurs. Dec 11, 9am EST/2pm GMT

CIDOC-CRM (and variants like Linked Art) provides a long-established and important framework for semantic knowledge organization related to cultural heritage. While standardization is important for building interoperability in the data landscape, how we arrive at standardization matters, especially in contexts entangled with legacies of colonialism. The imposition of rigid knowledge organization frameworks developed without input from marginalized stakeholder communities risks perpetuating existing biases in such contexts, especially where Eurocentric knowledge organization has delegitimized other ways of knowing.

With its multilingual infrastructure and existing multilingual support community, the Wikidata environment holds promise as a place to connect with globally-situated colleagues to experiment with less-prescribed LOD methods that hold space for more epistemic flexibility generated in a bottom-up capacity (Chen et al., forthcoming). Importantly, mechanisms that support the ‘mapping’ of properties between external ontologies and equivalents in the Wikimedia ecosystem can be mobilized to bridge across LOD approaches. This activity proposes to coordinate an asynchronous “discovery” phase aimed at documenting the state of property alignment between CIDOC/Wikidata, summarizing challenges, identifying existing features that can be cooperatively mobilized to promote better future interoperability, and suggesting steps for future community action.

Planned Outputs: This activity proposes to create queries and documentation summarizing the results of the “discovery” activity.

How will meetings/collaboration be arranged? We propose to kickoff with a brief introductory meeting via Zoom on 2nd December to coordinate discovery tasks, and will use a collaboratively edited GoogleDoc to lodge insights/notes from asynchronous discovery work completed independently. We propose to hold a second 1 hr discussion meeting (via Zoom) toward the end of the LP11 schedule to discuss the results of asynchronous discovery work and compile a list of suggested concrete next steps.

Registrants can fill in the short form here: https://forms.gle/w6mpocsve2RdDAFX7

Where to find out more and where will outputs appear? We have established Wikiproject CIDOC/Wikidata Interoperability as a place to document the results of this discovery activity and to facilitate future cooperation: https://www.wikidata.org/wiki/Wikidata:WikiProject_CIDOC/Wikidata_Interoperability

Writing SPARQL queries with GitHub Copilot

Convened by Duncan Hay (University College London)

Date and Time: 12/12/2025, 2-3.30pm UK time (UTC+00:00)

As many people working with Linked Open Data will attest, writing SPARQL queries is hard. Happily, Large Language Models, especially tools such as GitHub Copilot, are very good it. In this workshop, you will be introduced to how to set up and use Visual Studio Code and Github Copilot to write SPARQL queries and test them against the Tools of Knowledge database of scientific instrument makers.

After the demo and a chance to run your own queries, we will conduct a facilitated discussion around possible use cases, and the advantages and disadvantages of this approach. Questions addressed will include:

  • How might this be used in different work and research contexts?
  • How else might these tools be used to work with Linked Open Data? (for example GraphRAG and related methods)
  • How good are the queries that Copilot generates?
  • What gets lost when you offload your thinking to Copilot in this way?

Prerequisites

Before the session, you should have:

Joining Instructions

If you’d like to join, sign up using this form https://forms.office.com/e/543xaivwjb, ensuring that you include your email address. You will be sent a meeting link before the workshop.

Information about the workshop and an outline of possible next steps will be available on this web page: https://ddunc23.github.io/workshops/sparql-with-llms/.