On data spaces, cakes, poetry and hats: #Europeana2022

Reflections from the conference

Europeana 2022 – making digital culture count, took place from 28 – 30 September 2022. It was a celebration for at least two reasons.

• First, this conference was the first Europeana event bringing together participants face-to-face after the social distancing restrictions of the previous years. It was a hybrid event with hundreds of participants from some 80 countries around the globe, but the KB, the National Library of the Netherlands in the Hague buzzed with several venues and a big community coming to the event on the site.
• Second, this is the beginning of the transition of Europeana into gradually shaping in the future years the Europeana data space for cultural heritage. It was selected by the EC to implement its vision of the European data space for cultural heritage. This is yet another step in the constant expansion of the work on making digital heritage accessible at the fingertips of multiple communities with different needs. The journey of Europeana is now continuous for 17 years, and it seems the whole data space business will mark its stepping into maturity if we use as a metaphor the human life.

Such an event is an unmissable opportunity for me. I am really curious about how this new technological development will leave a mark on the digital heritage and culture sectors, and also I am very involved with the larger community around Europeana being a member of the ENA Management Board.

The bold new world of data spaces? It’s a misleadingly easy concept

The conference started with an appeal to put ourselves into a learning mindset which was setting a huge expectation that the event would help the audience understand better what data spaces are. The first panel of the event looked at this matter with the most extensively used metaphor being the one brought in by Esther Huyer from Capgemini on cakes. Yes, there are multiple admirers of cakes (aka users in the data space realm) with different needs and even restrictions (e.g. allergies). There are different potential producers and suppliers of ingredients, and someone can even be an intermediary and bring the table where the cakes will be assembled, displayed and admired (and potentially eaten) by the users.

Everyone loves a good metaphor, but this one is somewhat oversimplistic: the complexity of data spaces is coming from the diversity of what is ‘baked’ for the users, not many varieties of the same kind of a ‘product’. Data users within the digital cultural heritage domain are notoriously diverse and will probably be interested in all sorts of things which are not always easily translated into the traditional combinations of search keywords within the current digital solutions.

The mass culture already provided us vivid images of data spaces albeit not being named that way: I am sure everyone could recall a favourite scene from a movie where supercool agents (or hackers) are pulling on large touch screens and tables data from all sorts of sources to find the answer to a complex question very quickly. Has the time come to bring the power of analytics and flexible intelligent solutions within a digital cultural heritage environment, where a user can not only search for examples within a sea of objects but use analytical tools with ease (as in not being a data scientist)? Can for example any user discover the best documentary heritage sources to be used to illustrate a class in history (or music performances for a class in music?), or extract a dataset which can support answering a proper research question, or even construct their family tree with a simple request to the future data space?

The ideas on the potential uses of data spaces will most probably evolve with examples of use cases and scenarios. I have already argued that a great source of inspiration for use cases would be the work done by innovation labs. But there were some inspirational stories at the conference on approaches which can contribute to the construction of the data space.

One unconventional proposal presented during the conference was that data spaces can be built reorganising the content in a way which guarantees each object resides in one place using the conventions and tools of Solid Pods and developing a linked infrastructure allowing to pull the same content into multiple platforms. This vision was also coming with an essential question, what would be the role of Europeana if data spaces develop in this way? What exactly will be in its’ pod?

These examples show that there is still a need to arrive to a shared understanding what data spaces are within the cultural heritage domain. It looks like this is an intuitive concept, we all know what data is and what spaces are, but the big questions of what happens in the data space, how it empowers its users still require more insights. Here are a set of my personal questions for the future in no particular order:

• Where is actually this data space? How are users ‘entering’ there – do they go anywhere specific, or will it underpin our web searches on any platforms we might be using (as far as we access the web the way we do today)?
• What is expected the user to know for using the dataspace? If the space allows the use of intelligent tools, does this mean that the successful users will be some sort of geeks or data scientists?
• How the current content providers will contribute to this dataspace – should they change something in their infrastructures? Are any new providers going to emerge? What would be their roles?
• Which institutions will be able to act as data owners? What conditions will guarantee an institution will be successful as a data owner contributing to the data space? Are the small institutions at a disadvantage here?
• How many intermediaries – data providers – will be needed and what exactly are they going to do? Is Europeana of today a particularly strong data provider which can grow further into becoming the data space enabler?
• What technologies can be orchestrated within the data spaces to support the mixing, matching, reuse and packaging of results for the users?
• What will be the right implications of mixing objects from different data owners and data providers? How will rights be inherited? How the data space will identify cases where the current rights’ statements are problematic?
• What platforms and tools are already there to help develop the data space? What will be the underpinning architecture?

This list could be expanded with additional questions which were raised during the vivid and engaged discussion on Europeana Research Futures during the last conference day. While the discussion did not delve into the research use of the future data space, it brought a number of ideas on what more is needed to make the content and tools used within the current Europeana more helpful for those who have a specific research question in mind. And there was a palpable feeling of the desire to organise more active events such as datathons/hackathons for the research community.

A lovely addition to the serious discussions was

the poetry in the data space

One of my favourite discoveries at this event was meeting Mr Gee, a data poet. He put into work his ability to masterfully summarise in poetic form the various topics and ideas floating around.

Being very far from Mr Gee’s articulate style, here is my own haiku-style summary:

Hey data spaces,
what a great inspiration!
How do we start now?

And ‘my’ metaphor: inspired by The Hat Exhibition in De Passage

The cake metaphor we started with is a good one for diversity in demands and supply, but there is no strong space component in it. I loved the story coming from one of the elegant spaces in the Hague, De Passage. Some 3000 hats, forgotten in a vacant space close to a hat shop, had been discovered some 20 years after the shop closed. What a time capsule! About two-thirds of these hats are now displayed around De Passage in an installation by the Hague designer Pink Steenvoorden of Einstein Design.

I really liked several features of this story:

• Space: The hats are grouped in different spaces, and one can get an idea of some of them with a closer look – but also see that there are many more around.
• Appealing to our physical senses visually: you can see a close-up and also realise there is much more. We have a fundamental issue with the digital collections – they give the impression everything in this domain is digitised when we know it is some 15% of all cultural holdings across Europe. We are not very good though to show what is available in a digital form and to what extent it is representative of the whole population of similar artefacts/objects. When we navigate spaces, we rely on our senses. How are we going to ‘involve’ the senses in the digital heritage space?
• Diversity: the layout of the installation does not group similar items together but illustrates the diversity. Can we achieve this in a digital space not being overwhelming to the users?
• Discoveries are serendipitous. as far as we have a flexible and appreciative mindset.

Data spaces depart from the mass scale aggregation and open the door for intelligent tools and protocols which can communicate with data owners and data providers to answer specific requests of data users. It’s a great promise, and I am looking forward to seeing how the vision and implementation of it will take shape with Europeana Foundation, Europeana Network Association and Europeana Aggregators Forum at the helm.

Disclaimer: the views in this blog are my own and do not represent the opinions of the Europeana Network Association Management Board or any other entities mentioned throughout the text.