What we learned from analysing museum educational material
How do curators prepare and personalise content to serve different learning goals, themes, topics and audiences? If we want to create an authoring tool that allows cultural heritage professionals to create adaptive digital content that augments and personalises the museum experience for different visitors, we need to gain insight into the curator’s current practices.
Corpus analysis
In preparation of the meSch pilot studies related to the Museo della Guerra’s domain, FBK analyzed their whole educational material: booklets and itineraries, worked with and animated by human guides inside and outside the museum. The corpus analysis and the additional exercise of manual construction of digital versions of the booklets helped in identifying general user requirements that will feed the design of an authoring tool that facilitates the creation of adaptive digital narrations.
![]() |
![]() |
Sample pages from an educational booklet in use @ Museo della Guerra
Understanding content variety and narrative structures
A sample of 12 booklets were selected from the catalogue of the material prepared at Museo della Guerra for the educational activities of the 2013-2014 school year: 4 booklets are associated to itineraries inside the museum, 6 booklets provide support to history labs, and 2 booklets have been prepared as companion material for trails on the territory. In our sample, 2 booklets are specifically targeted to the learning needs of the first level secondary school students (11-14 years old); 4 booklets are prepared for the activities of the second level secondary school students (14-19 years old); and the remaining 6 booklets are targeted to all secondary school students (11-19 years old). The booklets are composed of 12 to 16 pages, and contain different types of content: descriptions of historical events; descriptions of the impact of the war on the civilians and the society; descriptions of soldiers’ lives and equipment; information about the military organization; transcriptions of original documents (letters, diaries, official documents); technical descriptions of exhibits; scans of original documents, pictures or drawings; images of original objects. In total, our corpus includes around 34.500 words and 300 pictures.
Educational material dissected by meSch researchers.
All the booklets revolve around similar content and images but they are edited with specific educational goals in mind: the curators aim to provide different perspectives and different levels of complexity in order to meet the specific needs of the educational activities and of the targeted audience. Essentially, they are engaging in a process of content adaptation.
Important strategies as base for functionality
By studying curators’ practices we identified seven relevant strategies that are important in the design of an authoring tool for the creation of digital content adaption.
#1: Facilitate the search for specific content.
The curators most often know what to look for, because they have used the material before for other educational booklets, catalogues, exhibit labels or room introductory panels, or because the material belongs to their internal archives, which they know very well. For Museo della Guerra this is certainly due to the fact that, even if changes to the collection on display might occur due to renovations or temporary unavailability of rooms, the core topics presented to the public are relatively stable (of course, this might not be the case for other museums with very dynamic collections, that cover multiple domains). When searching for this “familiar” content, authors currently use specific strategies that should be further investigated to implement friendly functionalities in the authoring tool. A sample strategy would be not to search the content chunk directly, but to look for previously edited material/presentations that curators remember contain useful content.
#2: Inspiration comes from previous work: templates are useful.
When commenting the results emerged from the corpus analysis one of the curators said: “I wish I had more time to prepare the booklets. Being in a hurry, it was easier for me to resort to previous booklets, copy the structure and update the content.”. Authoring facilities should therefore support the easy definition of narrative structures in form of templates to be easily reused with different content.
#3: Have the freedom to create new narrative structures.
Even though many of the analyzed booklets share the same multi-topic, multi-thread narrative structure, there is a clear need for creating new templates. In the words on one of the curators: “If I had more time, I would have selected the content more carefully, using new photos and diary excerpts. I might have structured the content differently”.
#4: Baseline personalization by customization over topic, genre, thematic thread.
The pages/topics of the booklets typically correspond to hotspots in a museum guided itinerary or to objects/documents inspected during a history lab, that the human guide properly explains. The corpus analysis provided evidence to the fact that baseline content personalization needs to support the following major features: overall learning goal, current hotspot/object, and audience, that are used to condition the selection of appropriate topics, genres and thematic threads. A pre-packaged template could be offered to curators to experiment with, like the one implemented in the Companion Novel prototype for the multi-thematic exploration of heritage.
Parallel thematic threads traverse the sequence of topics in a booklet
#5: Reuse content, but with adjustments.
Content reuse is a de-facto common practice adopted by curators (at least those at the Museo della Guerra) in authoring their educational material. However, our analysis revealed that most often slight adjustments (e.g., different length of excerpt; different phrasing) are performed over the content before it is reused in a new context. This practice requires that the editing of the content is considered as an explicit step of the authoring process and an explicit management of different versions of similar content is necessary to help the author find the most suitable chunks to reuse.
#6: Different presentation strategies according for different media.
The preparation of content for an onsite/online/paper delivery requires different discourse and presentation strategies that should be clearly differentiated and supported in the authoring tool component. Indeed, the 2D content structure suitable for paper booklets requires a proper extension to include the temporal dimension and media assets synchronization when the content is to be conveyed as a multimedia online presentation. Onsite multichannel presentations additionally require shorter media assets, as well as context awareness.
#7: Presentation preview is crucial.
The possibility of previewing the material and of testing the alternative adaptive paths is essential. Preview functionalities are needed for tuning the most appropriate length of media assets, the effectiveness of temporal synchronization, the cohesion and coherence of the various presentation instances. In our exercise of rebuilding some of the booklets for digital presentation, several assemble-test-adjust cycles were necessary to reach an acceptable result.
Next steps
The requirements that have emerged from the corpus analysis complement the findings emerged from interviews and co-design workshops with museum experts and curators, and are now being translated into design decisions for an authoring tool to help curators compose adaptive digital content (work within the project workpackage WP3) and into ideas for personalization templates and rules (work within the project workpackage WP4).
More information about the work in progress for the organization of case studies at the Museo della Guerra in a previous blog post, that describes the porting of the Companion Novel prototype to the outdoor setting of the trenches of Mount Nagià Grom.
Material Encounters with digital Cultural Heritage