Search

We present a modeling of document dates such that it can allow scholars to express uncertainty when annotating data, act as a differentiable loss function for training models and allow for unbiased interpretable performance evaluation under uncertainty.

In the ERC-Advanced Grant Project "DiDip – From Digital to Distant Diplomatics" (PI Georg Vogeler) we want to show which charter features are identifiable automatically and test which other features are recognized by the machine that can be useful for Diplomatics research.

The poster will present the project "Between Composition and Reception: the Authority of Medieval Charters" and discuss what role machine learning methods can have for the human interpretation of late medieval charters as means to convey authority, esp. regarding their graphical aspects

Our strategy to support filtering the descriptive texts and the transcriptions in Monasterium.net seeks to assign semantic categories for the legal acts they record, such as confiscation, donation, or property sale.

The presentation discusses the notion of a "Proto-edition" which combines a pure visual representation of text with a deep description skipping the transcription of the text. It relates it theoretical models like Patrick Sahle's text wheel (2013) and the use of artificial Intelligence for creating "artifical editions" (Stutzmann at IMC 2019).

Book of Abstracts from the Digital Humanities Conference 2023, held from 10th to 14th July in Graz, Austria

Monasterium.net ist mit seinen über 600.000 Urkunden eine Hauptanlaufstelle für die Forschung mit mittelalterlichen Urkunden. Das ERC Projekt "From Digital to Distant Diplomatics" beabsichtigt, diese virtuelle Forschungsumgebung auf einen KI basierten modernen Stand der Technik im Bereich Computer Vision und natürliche Sprachverarbeitung zu bringen. Das Poster beschreibt das Projektvorhaben und die in seinem ersten Jahr erreichten Ergebnisse. Ein Beitrag zur 9. Tagung des Verbands "Digital Humanities im deutschsprachigen Raum" - DHd 2023 Open Humanities Open Culture.

Die Einleitung kontextualisiert den Band und stellt ihn in Zusammenhang jünster Entwicklungen und Forschung zum Thema „digitale Edition“. Dieser Band versammelt Forschungsergebnisse aus einem vom österreichischen Bundesministerium für Bildung, Wissenschaft und Forschung finanzierten Projekt. Das „Kompetenznetzwerk ‚Digitale Edition‘“ (KONDE) hatte sich zum Ziel gesetzt, die österreichischen Akteure im Feld digitaler Edition so mit einander zu vernetzen, dass Synergien sichtbar werden, die Partner sich im Austausch gegenseitig befruchten sollten und nationale Infrastrukturen entstehen.

Bericht ��ber die Entwicklungen am International Prosopographical Interoperability Framework (IPIF). Ein Beitrag zur 8. Tagung des Verbands "Digital Humanities im deutschsprachigen Raum" - DHd 2022 Kulturen des digitalen Ged��chtnisses.

This volume is based on the selected papers presented at the Workshop on Scholarly Digital Editions, Graph Data-Models and Semantic Web Technologies, held at the Uni- versity of Lausanne in June 2019. The Workshop was organized by Elena Spadini (University of Lausanne) and Francesca Tomasi (University of Bologna), and spon- sored by the Swiss National Science Foundation through a Scientific Exchange grant, and by the Centre de recherche sur les lettres romandes of the University of Lausanne. The Workshop comprised two full days of vibrant discussions among the invited speakers, the authors of the selected papers, and other participants.1 The acceptance rate following the open call for papers was around 60%. All authors – both selected and invited speakers – were asked to provide a short paper two months before the Workshop. The authors were then paired up, and each pair exchanged papers. Paired authors prepared questions for one another, which were to be addressed during the talks at the Workshop; in this way, conversations started well before the Workshop itself. After the Workshop, the papers underwent a second round of peer-review before inclusion in this volume. This time, the relevance of the papers was not under discus- sion, but reviewers were asked to appraise specific aspects of each contribution, such as its originality or level of innovation, its methodological accuracy and knowledge of the literature, as well as more formal parameters such as completeness, clarity, and coherence. The bibliography of all of the papers is collected in the public Zotero group library GraphSDE20192, which has been used to generate the reference list for each contribution in this volume. The invited speakers came from a wide range of backgrounds (academic, commer- cial, and research institutions) and represented the different actors involved in the remediation of our cultural heritage in the form of graphs and/or in a semantic web en- vironment. Georg Vogeler (University of Graz) and Ronald Haentjens Dekker (Royal Dutch Academy of Sciences, Humanities Cluster) brought the Digital Humanities research perspective; the work of Hans Cools and Roberta Laura Padlina (University of Basel, National Infrastructure for Editions), as well as of Tobias Schweizer and Sepi- deh Alassi (University of Basel, Digital Humanities Lab), focused on infrastructural challenges and the development of conceptual and software frameworks to support re- searchers’ needs; Michele Pasin’s contribution (Digital Science, Springer Nature) was informed by his experiences in both academic research, and in commercial technology companies that provide services for the scientific community. The Workshop featured not only the papers of the selected authors and of the invited speakers, but also moments of discussion between interested participants. In addition to the common Q&A time, during the second day one entire session was allocated to working groups delving into topics that had emerged during the Workshop. Four working groups were created, with four to seven participants each, and each group presented a short report at the end of the session. Four themes were discussed: enhancing TEI from documents to data; ontologies for the Humanities; tools and infrastructures; and textual criticism. All of these themes are represented in this volume. The Workshop would not have been of such high quality without the support of the members of its scientific committee: Gioele Barabucci, Fabio Ciotti, Claire Clivaz, Marion Rivoal, Greta Franzini, Simon Gabay, Daniel Maggetti, Frederike Neuber, Elena Pierazzo, Davide Picca, Michael Piotrowski, Matteo Romanello, Maïeul Rouquette, Elena Spadini, Francesca Tomasi, Aris Xanthos – and, of course, the support of all the colleagues and administrative staff in Lausanne, who helped the Workshop to become a reality. The final versions of these papers underwent a single-blind peer review process. We want to thank the reviewers: Helena Bermudez Sabel, Arianna Ciula, Marilena Daquino, Richard Hadden, Daniel Jeller, Tiziana Mancinelli, Davide Picca, Michael Piotrowski, Patrick Sahle, Raffaele Viglianti, Joris van Zundert, and others who preferred not to be named personally. Your input enhanced the quality of the volume significantly! It is sad news that Hans Cools passed away during the production of the volume. We are proud to document a recent state of his work and will miss him and his ability to implement the vision of a digital scholarly edition based on graph data-models and semantic web technologies. The production of the volume would not have been possible without the thorough copy-editing and proof reading by Lucy Emmerson and the support of the IDE team, in particular Bernhard Assmann, the TeX-master himself. This volume is sponsored by the University of Bologna and by the University of Lausanne. Bologna, Lausanne, Graz, July 2021 Francesca Tomasi, Elena Spadini, Georg Vogeler

Starting from the observation that the existing models of digital scholarly editions can be expressed in many technologies, this paper goes beyond the simple opposition of ‘XML’ and ‘graph’, It studies the implicit context of the technologies as applied to digital scholarly editions: embedded mark-up in XML/TEI trees, graph representa- tions in RDF, and stand-off annotation as realised in annotation tools widely used for information extraction. It describes the affordances of the encoding methods offered. It takes as a test case the “assertive edition” (Vogeler 2019), in which the text is considered in a double role: as palaeographical and linguistic phenomenon, and as a representation of information. It comes to the conclusion that the affordances of XML help to detect sequential and hierarchical properties of a text, while those of RDF best cover the representation of knowledge as semantic networks of statements. The relationship between them can be expressed by the metaphor of ‘layers’, for which stand-off annotation technologies seem to be best fitted. However, there is no standardised technical formalism to create stand-off annotations beyond graphical tools sharing interface elements. The contribution concludes with the call for the acceptance of the advantages of each technology, and for efforts to be made to discuss the best way to combine these technologies.

TEI offers a great vocabulary for text encoding in a default set of genres and use cases. Minor use cases are usually covered by customisations. This approach is not very effective, if other use cases provide already a formalized vocabulary in particular if this vocabulary is implemented with non-xml schemata like RDF. This contribution considers the problem on the example of study of medieval charters, the field of research called "diplomatics". We propose a method of encoding concepts of diplomatics in digital scholarly editions. For diplomatics the CEI (http://www.cei.lmu.de) and the proposals by Sean Winslow (2019, jTEI forthcoming) take the first way. Bradley et al. (2019) have proposed an alternative approach. The diplomatics community has created a vocabulary for terminological use already in the 1990s with the Vocabulaire International de Diplomatique (Cárcel-Orti 1997, https://www.cei.lmu.de/VID). Our proposal uses a combined approach: using standard TEI, enhancing it with concepts from the VID. This method can make extensive use from a well established TEI annotation method in the globally available ana-attribut. We will demonstrate this on the example of the Serbian medieval charters, as they are processed in the project Diplomatarium Serbicum Digitale (DSD). The VID is applied for specialisations of generic TEI elements. The effect of this approach is that a semantic web based model of charter descriptions can be extracted from TEI encoding.

Presentation for the panel discussion at FORGE11

For almost 20 years, the digital repository GAMS (Geisteswissenschaftliches Asset Management System) has been providing long-term preservation of Humanities’ resources offering both a technical solution and a way of achieving sustainability in the handling of research data. Such domain-specific digital repositories rely on specialised design principles and a strategic orientation to balance the need for long-term sustainability and the use of advanced technologies. The repository solely relies on open source software due to its commitment to open access and open science. Its design principles intend to foster compliance with international best practices like the COAR Community Framework for Good Practices in Repositories (COAR 2020) and the certification guidelines of the CoreTrustSeal, and following the FAIR principles of findability, accessibility, interoperability and reusability. Long-term preservation entails migration to new software versions and reflecting advancements in technology as well as responding to security issues; increase in projects and services require a more elaborate and stable system architecture. This submission describes the conceptual framework and architecture of a large digital repository infrastructure that specialises in the preservation, management and presentation of research data in the Humanities, as well as the maintenance and migration of this exemplary research data ecosystem.

This article contributes to the ‘Confronting the Digital debate by set-ting out the case for using the methods of an ‘assertive digital edition to enrich historical research and to make archives accessible to a range of users. It illus-trates its case with data modelling, data analysis and personal stories arising from the use of digital resources derived from the British hearth tax records (166295). Parts 12 address the nature of the debate for academia, libraries and ar-chives, and their relationships with commercial organizations. Parts 34 explore the work of ScotlandsPlacesand Hearth Tax Digital, setting out the methods of an assertive digital edition in enhancing record access, research and long-term preservation, and assesses the benefits for academic research. Parts 56 demon-strate how this work has enhanced life skills and increases peoples confidence outside the academic world through research undertaken in schools and colleges and by the University of the Third Age.The article shows that an assertive edition approach brings considerable benefits to thestudyofpre-modern sources dis-persed across archives, therebyenabling historians and other users to conduct qualitative and quantitative research both within and outside academia on com-munitiesand ordinary people from the early modern period. Thisnot only en-hances research knowledge but also increasesconfidence among diverse groups. Version of record

The digital scholarly edition of the Basel City Accounts has been successful in its combination of TEI/XML encoded transcriptions with RDF representation of the historical content of the accounts. The contribution discusses the use of the RDF rule language SHACL to define a formalism to convert the implicit structure of the XML encoding into explicit RDF statements. It suggests to use precedence and dominance of XML elements as indicators for the implicit information. SHACL can serve as the formalisation of constraints. used to extract XPath expressions, which include missing explicit information from elements in the ancestor or preceding axis. With this method, a formal description of the target data model can be used to generate XSLTs converting the TEI/XML into RDF/XML. Version of record

A single abstract from the DHd-2020 Book of Abstracts., Sofern eine editorische Arbeit an dieser Publikation stattgefunden hat, dann bestand diese aus der Eliminierung von Bindestrichen in Überschriften, die aufgrund fehlerhafter Silbentrennung entstanden sind, der Vereinheitlichung von Namen der Autor*innen in das Schema "Nachname, Vorname" und/oder der Trennung von Überschrift und Unterüberschrift durch die Setzung eines Punktes, sofern notwendig., {"references":["https://doi.org/10.5281/zenodo.3666690","https://github.com/DHd-Verband/DHd-Abstracts-2020"]}

Unter dem Gesichtspunkt der „Rekontextualisierung“ erweist sich das Semantic Web als eine eigenartige textuelle Ausdrucksform, die als Prototyp einer digitalen Rekontextualisierung verstanden werden könnte: Es besteht konstitutiv aus Aussagen (Basisprinzip Triples), die über das Internet miteinander verknüpfbar sind (Basisprinzip IRI). Im Unterschied zum Hypertext sind die Aussagen aber hochgradig fragmentiert und formalisiert. Sie sind damit „digitaler“ als Hypertext, da sie weit stärker als dieser diskret konfiguriert sind. Zusätzlich sind sie einigen formallogischen Methoden zugänglich, welche die für einen Text notwendige Kohärenz erzeugen. Aus textlinguistischer Sicht könnte man also das Semantic Web nicht nur als „Giant Global Graph“ sondern auch als „Giant Global Text“ zu beschreiben versuchen. Damit wird der Begriff der Rekontextualisierung für die Verwendung des Semantic Web in den DH produktiv, weil er darauf hinweist, dass das Semantic Web nicht ohne menschliche Intervention Text produziert. Daraus ergibt sich aber auch eine »Unsichtbarkeit« des Semantic Web. Die Rekontextualisierung mit Methoden des Semantic Web kann deshalb auch darin bestehen, den „Giant Global Text“ mit jeweils neuen Paratexten zu versehen und so neue Rezeptionsformen zu ermöglichen.

The paper discusses the question whether methods applied in work with prosopographical data can be integrated into an ”International Proposography Interoperability Framework” (IPIF) comparable to the IIIF (International Image Interoperability Framework). It suggests basing this upon a RESTful interface defined in a publicly available definition of an API. The API is based on the three-partite factoid model, thus keeping the identification of the person, information on the person, and the source documenting this information separated, aggregating it in a “factoid” with the metadata of its creation. The API definition follows the rules of OpenAPI which allows automatic code generation. As a proof of concept, the API has been implemented in the context of the APIS project, containing the information of the Austrian biographical lexicon, and with data from monasterium.net, the world wide largest database of medieval charters. It introduces a wrapper for LOBID and wikidata and a standalone server created with code generation from the OpenAPI specification as a second proof of concept implementation. In a third proof of concept it showcases the reuse of tools originally developed for APIS in other services.