Digital sustainability in art museums
Art museums don’t just need to provide access to art in their buildings, they also need to make their collections available online in a sustainable way. Meemoo has therefore been working together with museums from the former Flemish Art Collection (now meemoo) on various projects since 2012.
Using analyses of existing infrastructures as our starting point, we’ve collaborated with Contemporary Art Heritage Flanders (CAHF) and the Flemish Community Collection (Collectie Vlaamse Gemeenschap – CVG) to develop a data hub that connects different collections.
1. Review the technical structure of VKC’s online catalogue (2012 – 2013)
In 2004, the former Flemish Art Collection launched one of the first online collection catalogues in Flanders. This catalogue had the ambition of providing full access to the collections from Flanders’ three main art history museums in virtual form. It was hoped that presenting these hidden collections online would stimulate more collection registrations. The range of collection catalogues available online has expanded dramatically since, and so too has the complexity. At the start of 2012 we reviewed its technical structure to see if it still satisfied digital strategy requirements. We formulated three research questions:
What functional requirements are set for the online catalogue?
Which computer systems make up the technical structure and how are they linked together?
How could this technical structure be simplified and made more efficient?
We wanted this review to provide clarity about the existing technical structure behind the online catalogue, and offer recommendations to simplify the structure of this architecture and make it more efficient. The conclusions and recommendations from the review were incorporated in an action plan, which was started in 2013.
2. Persistent identification (2013 – 2016)
Making a collection findable online is a first step. Keeping it findable requires persistent identification. This is done using a persistent URI: a web address that’s composed using certain rules, with a clear and consistent structure, which is easier to manage and use, and is stable (unchangeable) and accessible in the long term.
We ran a series of pilot projects in 2016 in the framework of the ‘How do I become a data publisher?’ and ‘Sustainable links between artworks, archives and publications’ projects (pages on CEST only in Dutch). We linked the activated persistent URIs for artworks to related library and archive content, and vice versa. This interconnected the museum, library and archive collections and made them accessible in an innovative way. The people who visit (digital) museum collections can now easily find their way to libraries and archives, and vice versa.
3. Event-based object descriptions (2014 – 2015)
Un a sub-project we focused on the concept of events: clusters of information about where and when certain actions are taken, and by whom, with regard to specific artworks. The aim of this project was to investigate how the broad range of contextual information about artworks can be clustered in a generic data structure. This information should be coded in a machine-readable way using LIDO XML and made re-usable for web applications. The project looked at the new possibilities that are created when you can visualise an artwork’s history with this enriched information.
4. Publishing linked open data with Wikidata (2015 - 2016)
Making the metadata from six art museums available via linked open data (LOD) requires quite a lot of work. In this project, we investigated how we could use the Wikidata platform to help us. The aim was to see if the infrastructure that Wikidata makes freely available can be used by museums to put their open data policy into practice. This would mean they could publish their collection data as LOD without needing to make any large IT investments. This further enriches the information available in Wikidata, makes all artworks in the seven museums findable in Wikidata, and gives the museums a powerful tool to involve Wikipedia users in the further development of their collection data. So everyone wins.
5. Layered licence model (2014 – 2015)
When enabling the re-use of content, it’s important to provide clarity about user rights and licences. We therefore started a project in 2014 with the aim of developing a clear matrix of content, user rights and licences for re-using data and images from the museums. The layered licence model was published on CEST and indicates which content from the registration systems and from the museum’s image databases can be re-used by the public and under what conditions.
6. Implementing a data hub for museums (2016 – 2018)
The previous projects were a useful prelude for creating a shared data hub for (art) museums. The data hub is a sustainable aggregator that makes it possible to search through data about art collections from collection management systems from various affiliated museums. The data hub uses an API to make the data available for re-use in various formats.
The objectives of this project were:
to design a reference architecture for a data hub. We work out a specification for a set of web services for publishing and exchanging collection data in an open, standardised and reliable way.
to develop an open source data hub framework package. This means that we build an open source software library which allows us to realise the specifics for the reference architecture.
to roll out a reference implementation of the data hub using the collection data from the museums that are part of the former Flemish Art Collection (VKC) and Contemporary Art Heritage Flanders (CAHF).
to design and develop a dashboard that measures the quality of museum collection data, visualises it in graphs and statistics, and presents it in a user-friendly way. The application offers museum employees clear insights into the current status of the registered data for large groups of museum objects.
to develop strategies and expertise to clean up collection data in a targeted and systematic way. The project wanted to convey these attitudes and transfer this experience and knowledge to museum registrars by organising practical, hands-on data cleansing workshops in the museums.
to include the data from CAHF museums in the VKC Arthub. We helped the CAHF museums prepare their data for inclusion. This preparation essentially involved the steps taken by the VKC museums in the persistent identification process, which are necessary for enabling the technical inclusion of data in the Arthub.
to build an open source community of data hub developers and users to take on management of the software.
Consult the data on Arthub Flanders >
Discover the REST API endpoint and the OAI-PMH endpoint of the data hub >