Since 2016 the Autonomous Province of Bolzano-South Tyrol offers machine-readable access to selected cultural data sets as Open Data:
The first contains basic information about 160 different museums, collections and exhibition venues in South Tyrol: names, locations, contact information, geolocation, opening times, ticket information, all in three languages (English, German, Italian) (see https://data.civis.bz.it/de/dataset/musei-in-alto-adige). This data set is used by different touristic websites.
The second data set contains data from over 50 collecting institutions in South Tyrol: museums, archives, associations, other collections. This data set contains collection data of more than 310.000 objects, artefacts, art works, photographs, each linked to the respective institution (and its descriptive information). The institutions vary in size, resources and professionalization. People working in this area seldom are digital experts, a lot of them are honorary workers. This object metadata is structured according to national and international standards and comprises object names, description and historical-cultural comments, creator’s names (people and organizations), creation dates and places, titles, fonds, materials, techniques, dimensions, keywords, and image files (including rights information). Standardized terms (objects names, people, institutions, material) contain links to external, internationally established authority files (Linked Open Data). (Link to the data set: https://data.civis.bz.it/de/dataset/catalogo-beni-culturali, will be revised this summer). Data is provided mainly in German and Italian, parts also in English. This data is accessible for the public also on a website (https://objekte-museen-archive.provinz.bz.it/), The data is continuously processed, and more records are constantly being added. The data is used for example on a local (https://myargo.bz/) and national level (https://catalogo.beniculturali.it/). We do not know if others also are using the data on other endpoints.
A third bilingual (German and Italian) data set publishes historical photographs including printable image files (https://data.civis.bz.it/de/dataset/tyrolean-historical-photographs).
The talk will discuss the perspective of a cultural data provider and some of the challenges in this area:
1. The 50 institutions have different focus, approaches and resources at hand. Collections are vast, digitization/cataloging is time consuming work requiring specific skills.
a. Can different digital tools and AI be used to facilitate the work?
b. How much centralized data quality control is necessary?
c. How to keep such a diverse network alive, active and connected?
2. With some of the data stemming originally from historic record cards and inventory books, and after almost 20 years use of a mutual digital system in 50 different institutions, clean-up and standardization are central tasks.
a. How much standardization can the institutions handle?
b. What AI tools might be useful instruments for data clean-up? We use OpenRefine and have experience with database enrichment in an explorative AI project with Axiell using Claude. How can we find competent partners in this area?
c. The idea of clean and lean data is difficult/unrealistic to achieve. Is it still okay to publish data containing uncertainties, mistakes or is incomplete or should we further limit the publication of such data?
d. How to deal with the gap between the elevated trust people have in cultural institutions and the rising mistrust in online information?
3. In an age where all allocatable information is being harvested for the use of generative AI: Is an Open Data approach still helpful for cultural institutions? Should we not return to the old days of gatekeeping the knowledge/information in cultural institutions? Is there any way to ensure that this information is used adequately and respectfully? What kind of data is meaningful data nowadays?
