Just received the great news that the video recordings from #DH2025 have been published on the conference website https://dh2025.adho.org/. But it would be great if these were published with an explicit license and a unique URL (or DOI) for each recording to allow for re-use and distribution.
There are probably a plethora of open, non-profit platforms that could be used, but https://av.tib.eu comes to my mind as an almost perfect fit. This would also reduce the strain on the ADHO infrastructure, which currently serves the video I am interested in at less-than-ideal 500 kB/s.
#dh2025 @dhtech mini-conference has evolved into full Technical Track at #dh2026 conference 👀 👏
https://dh2026.adho.org/cfp/
It was a pleasure to present at Digital Heritage 2025 conference in Siena the results of the Mini-Muse research project — a preliminary study that combines #NLP algorithms and #DataVisualization to enhance access to digitized publications on history. The research team includes Fabio Rinaldi and Joseph Cornelius. The project was made in partnership with ETH-Bibliothek.
You can read the paper at link below.
One of the most important tasks for a historian is to identify key historical figures across multiple cultural archives and analyze their impact on history. The extensive effort of cultural institutions in digitizing historical archival materials and distributing them through online digital archives have significantly enhanced the study of historical figures. However, current historical digital archives, which rely on keyword-based search methods, often return numerous but imprecise results making it challenging for historians to understand chronological and contextual events surrounding a historical figure. We would like to present the result of the Mini-Muse project. It was a preliminary research project aimed at identifying data visualization models and user-friendly interface features to help historians visually explore historical figures and their actions. The project leverages Natural Language Processing (NLP) algorithms to extract metadata from unstructured text and generate structured data about key figures. It also applies data visualization techniques to support the visual analysis of each figure's timeline. The project adopts a user-centered design approach to ensure that the user interface features meet the needs of historians. It involves a pool of Swiss and Italian historians to gather insights on their research practices and validate a working prototype. The findings of the preliminary research project suggests that the introduction of an ''action flow view'', an interactive timeline displaying the historical figure's actions extracted automatically, can significantly improve the identification and study of historical figures.
How can AI help us understand both 17th-century church administration and today’s public institutions? Join Christoph Sander and Jan Mendling for a thought-provoking talk on efficiency, corruption, and the power of algorithms. Please note that the event will be conducted in German.
We are deeply saddened by the tragic news from #Lisbon, where we had the chance to stay in July for #DH2025.
Our photos are of a different funicular railway (Elevador da Bica) & from a much happier moment.
Estamos profundamente tristes com as trágicas notícias de #Lisboa, onde tivemos a oportunidade de ficar em julho para o #DH2025.
As nossas fotos são de um funicular diferente (Elevador da Bica) & de um momento muito mais feliz.