Summary
This week, Hugo went to a conference about Digital Humanities in the Netherlands, as well as a workshop on the use of crowdsourcing in a cultural heritage setting. This resulted in more ideas on enriching and visualizing data in a research and institutional context. Furthermore, we made a start with obtaining additional sets of book (meta)data, and worked on the prioritization of project activities.
Specific activities
Visit of the Creative Amsterdam - An e-Humanities Perspective conference. 28-29.10.2016. This conference looked at:
- The relationship between cultural artefacts (art, books, etc.) and the urban networks and spaces in which they were conceived, (re)produced, distributed, mediated, and consumed”, and at “how such issues can be studied by means of existing and novel (digital) methods, as well as comparative and long-term approaches.” The attended sessions handled a wide variety of topics, including the history of “creativity & the city” (Prof.Van Damme, Antwerp Univ.), the concept and visualization of “missingness” in performing digital history research (Matthew Lincoln, Getty Research Institute), and the lessons digital humanists can draw from past adoption of new technologies by historians (Scott B. Weingart, Carnegie Mellon).
A full report is available here. Specific sessions had the following topics:
- Uses of Digital Heritage.
- How can you “experience” architecture through digital technologies, such as virtual reality and augmented reality (e.g. an “augmented facade”, point phone camera at the facade of building and view old/new/inside of building), and how can you blend the physical and “virtual” reality (e.g. organizing exhibitions). A second presentation was about digital museums and networks of innovation, tracing the popularity of museum objects through analyzing their popularity online (online visits, Wikipedia) and offline (visitor statistics). Museums that are most popular in terms of physical visitors (e.g. the louvre) are not necessarily most popular online. Third, a presentation by the Amsterdam Museum talked about how engagement with digital content (digi depot) may be measured.
- Digging into (Linked) Data.
- How can one link various datasets and tools (e.g. Getty LOD, Monk tool, transcribus). In the “Golden Agents: Creative Industries and the Making of the Dutch Golden Age” project they focus on ontologies created by experts. In a second presentation, the topic of Historical Event Search (“Thematic Exploration of Parliamentary Data using Knowledge Bases”) was discussed. The target topic is scattered across parliamentary proceedings (e.g. references to Golden Age -- a highly complex concept to find in the dataset). To prepare for this task, entity linking was performed, tagging and disambiguating named entities and concepts, then using the DBpedia semantic network to dynamically build more comprehensive representations of historical events, and creating a simple yet powerful querying interface. Finally, Lab1100 presented their database for scholarly research (NodeGoat research environment), available for free for individual researchers.
- New tools for Cinema History and Visualization.
- A first presentation discussed the “Heuristic value” of tools for research (including an interactive map depicting early cinema in the Netherlands). This may serve to get ideas in initial phase of research, but is not necessarily exhaustive or reliable. Again via a map interface, other researchers explored patterns of cinema locations in relation to (sub)urbanisation over time. Finally, another research presented “Cinematic cartography”, geovisualizations of cinema-going experiences in 1950s Italy, based on 109 interviews (oral history). An issue in visualization of these things is how to intersect temporal and spatial perspectives via maps.
- Other. Other talks explored how data processing and visualization can be created to create an “information time machine” (see also this TED talk), building a “multidimensional model of 2000 years of European history”, and depicting 1000 years of the history of Venice. A European grant proposal is prepared for this project. Finally, the same researchers presented about the European circulation of people, goods and patterns. They used machine learning to detect features of images, so researchers can search via their prototype for similar motives across paintings.
- A first presentation discussed the “Heuristic value” of tools for research (including an interactive map depicting early cinema in the Netherlands). This may serve to get ideas in initial phase of research, but is not necessarily exhaustive or reliable. Again via a map interface, other researchers explored patterns of cinema locations in relation to (sub)urbanisation over time. Finally, another research presented “Cinematic cartography”, geovisualizations of cinema-going experiences in 1950s Italy, based on 109 interviews (oral history). An issue in visualization of these things is how to intersect temporal and spatial perspectives via maps.
- Visit workshop "Two birds, one stone: Bridging cultural heritage collections with crowds and niches", 2016-10-31. Institute for Sound & Vision, Amsterdam. A full report is available here. A summary of the presentations:
- A first presentation discussed Waisda, a video tagging game for audiovisual heritage collections, and how to utilize the “cognitive surplus” of website visitors for enriching and annotating heritage collections. Learned lessons from three video crowdsourcing projects include the importance to aim for attracting niches, not the general public, that using open knowledge structures is preferable, and that it is necessary accept and respect multiple authorities and perspectives with respect to an institution’s own collection.
- A second presentation discussed “Crowdsourcing and Co-curation from a Natural History Perspective”, and mentioned nederlandsesoorten.nl (species thesaurus), a crowdsourcing “specimen label transcription” project (via velehanden.nl), and xenocanto.nl, a world-wide collaborative bird sound database. Learned lessons emphasized the importance of shared goal for the users (“solve puzzles together”), allowing for autonomy and mastery (e.g. customized personal user pages). But also content issues arose when doing crowdsourcing (disagreements, and copyright problems). (Lack of) Available expertise may also be a bottleneck.
- The Rijksmuseum discussed the crowdsourcing of their large online collection, 274K hires images. Structured crowdsourcing is done via the Accurator toolset, aiming for specific niches. Various issues may arise, such as the consolidation of previous research results, the ingestion of elicited metadata in the collection, infrastructure management, and data modeling.
- Finally, the DigiBird project aims at “on the fly collection integration using crowdsourcing”. It aims to create “hub” between projects using shared vocabularies as a bridge, and allows for integrated searching in different crowdsourced bird-related audiovisual collections (e.g. for a specific bird). Institutions are able to monitor their (crowdsourcing) progress dynamically.