Digital research infrastructure for archaeology, ancient history & historical evidence
An open ecosystem of tools for documenting, structuring, analysing, relating and evaluating archaeological, textual, material, spatial and chronological evidence.
The Digital Laboratory explores how digital methods and artificial intelligence can help transform fragmentary, heterogeneous historical evidence into knowledge that remains traceable, contestable and reusable. Provenance, uncertainty and scholarly judgement remain part of the research process.
01 / The research problem
Historical research works with incomplete evidence: archaeological contexts, objects, inscriptions, papyri, manuscripts, buildings, images, chronologies, spatial relationships and earlier interpretations.
Digital systems can blur the differences between what was observed, what was computed and what was inferred. This Laboratory studies how to keep those distinctions inspectable and connect each historical interpretation to the evidence and reasoning behind it.
02 / From evidence to historical knowledge
A source is not automatically an observation. An observation is not automatically an interpretation. A hypothesis does not become a historical claim without review.
An object, papyrus, inscription, archaeological unit, photograph, building, text or map that can be identified as an evidential origin.
03 / Evidence architecture
Different evidence types can inform a proposition while retaining their distinct origins and epistemic status.
Read the methodologyConceptual demonstration · Experimental framework
This schematic demonstrates how a claim might connect documentary evidence, a person, an office, a place and a date. It is not a live database and does not assert a historical identification.
04 / Research tools & infrastructure
The applications contribute to distinct research stages. A conceptual relationship between tools does not imply automatic data exchange.
Loading research tools…
25 distinct catalogue tools are listed from the current public page. Tool descriptions and statuses reflect published information; details that could not be confirmed are omitted.
Evidence-specific method notes
These methodological principles clarify how particular evidence types move through analysis. They describe research practice, not a claim that each step is implemented in every tool.
Processed archaeological images must remain conceptually distinguishable from the primary visual record. The methodological objective is to preserve the relationship between original image, analytical transformation, parameters and resulting observation whenever technically possible.
Image, transcription, reading, expansion or restoration, interpretation and translation are separate operations. Automated or AI-assisted readings are propositions, and human scholarly validation remains necessary.
A chronological equivalence is an argued transformation, not a context-free value. Preserve the source of calculation, conversion rule, date range and uncertainty where available.
Name similarity alone does not establish that two attestations refer to the same person. An identity proposal should be assessed with its confidence, justification, evidence and validation status. This is a methodological principle, not a claim of a particular implemented ontology.
05 / Research laboratories
Oxyrhynchus and Heracleopolis Magna offer complementary contexts for examining methods across textual, archaeological and material evidence.
Documentary density brings papyrological, epigraphic, archaeological, architectural and material evidence into close relation, across Roman and Late Antique urban histories.
A different archaeological and urban sequence offers a complementary setting for examining how methods respond to other relationships between textual and material evidence.
The objective is to develop transferable methods for integrating heterogeneous historical evidence, not digital systems limited to two cities.
06 / Scholarly responsibility
AI-generated readings, classifications, translations, reconstructions and interpretations are research propositions, not definitive historical evidence.
The source and process behind a proposal should be identifiable whenever technically possible.
Research proposals must remain open to correction and scholarly revision.
Alternative readings, ranges and unresolved questions should remain expressible.
Scholarly interpretation and final academic responsibility belong to researchers.
07 / Research conditions
Historical claims gain context when their sources, transformations and limits can be examined.
Every observation, transformation and interpretation should remain connected, where possible, to its source, author and analytical history.
Alternative readings, dates, identifications and interpretations may coexist until further evidence enables evaluation.
Where technically possible, preserve inputs, transformations, parameters, versions and outputs so workflows can be inspected and repeated.
Progressive integration of relevant standards is under evaluation; compatibility is not claimed unless a tool documents it.
08 / Research outputs
Publication and citation records will be added when verifiable information is available. No DOI or publication is inferred here.
Research applications, with versions and preferred citations added when documented.
Benchmark datasets are planned research infrastructure and are not presented as available resources.
PLANNEDA space for research using the Digital Laboratory. No publication records are listed until verified.
RECORDS TO BE ADDEDDocumentation of workflows, evidence models and methodological change as they develop.
09 / Research software lifecycle
Tools move through experimental, prototype, beta and stable stages. Methodological change should be documented to support reproducibility and citation.
A related research agenda
The Digital Laboratory provides an experimental and methodological environment for questions explored within the OPENPAST research agenda. The framework is under development. Explore the functional v0.1 prototype →
Open science
Apply FAIR principles progressively to software, workflows, metadata, documentation and research outputs where feasible.
FFindableDiscoverable records
AAccessibleAccess within rights and licences
IInteroperableStructured exchange where appropriate
RReusableAttribution and documentation
Credits & licences
Licensing must be determined separately for author-developed software, datasets, original images, third-party images and texts, external data, software libraries and APIs. No single licence is asserted for all Laboratory content. When third-party media are added, their source, institution, creator, licence and original URL should accompany the item.
Software citation details, DOI, repository, release and licence are shown only when verifiable. Unassigned citation metadata is not fabricated.
Open access depends on applicable licences and third-party rights. Primary records and source images remain subject to their own access conditions.