Disclosing Architecture: 18 Stories of Heritage and InnovationDisclosing Architecture: 18 Stories of Heritage and Innovation

Text Lois Hutubessy

The aim was to develop a user-friendly web-based system that would allow architecture and urban planning enthusiasts to interactively explore the Nieuwe Instituut’s extensive collection. To achieve this, the institute worked with a number of suppliers to develop various system components. Linked by Air designed the concept and style guide, Lifely developed the front-end component and Triply provided linked data products and expertise. The project also introduced new internal roles and responsibilities, including a product owner, editorial team and architecture group.

External parties were selected through a tender process to start the the project in autumn 2021. During the initial design phase, several concepts were compared, sketched and prototyped for remote testing with end users. These designs were then passed on to the development team, who defined the scope of the minimum viable product. The first version of the product was presented at the Disclosing Architecture conference, Disclosing Futures, Rethinking Heritage, in autumn 2022.

Feedback from the initial launch was used to shape a project reboot in 2023, during which significant changes were made to the platform’s design and infrastructure, project approach and team workflow. In addition to design and development, parallel tracks were established to improve data quality, strategy design and information architecture. This ensured a comprehensive and coordinated approach to the ongoing development of the platform.

Looking back, the project went through three distinct phases, each with its own lessons. First, the design phase involved several iterations with different stakeholders and user groups. It was only when the designs were handed over to the development team that the implementation hurdles became apparent. This led to a reboot of the project, focusing more on process management and adopting an agile approach to increase the efficiency of the scrum team.

However, the goal of maximising end-user value within the remaining budget did not allow the product to reach a level where it could be successfully flaunched for a wider audience. In the final iteration, the feasibility of the technical implementation became a priority. Designs were reworked to be development ready, and a fixed scope approach gave the team greater control over the end result.

A challenge throughout the project was aligning processes and workflows between multiple parties across different time zones and changing team composition. New agreements were constantly being made to adjust the project organisation and approach, establishing ‘ready processes’ and the ‘definition of done’.

Technically, the main challenge was to develop a robust, stable product that meets performance criteria and other non-functional requirements. To address this, an architecture group was established to continuously assess feasibility and review implementation. Another challenge was to improve data quality, which was addressed through a separate data cleaning initiative.

To make things even more complicated, the platform is fully compliant with linked open data standards and follows the Dutch Digital Heritage Network’s blueprint for building service platforms. Linked open data is used to link the Nieuwe Instituut’s collection with those of other cultural organisations. This allows visitors to the collection platform to browse data from multiple sources. An unforeseen challenge was the actual selection and matching with external sources that also publish their data as linked open data, as it appeared that not many sources were ready to be linked.

A key lesson from this project is the impact of data quality on the success of the product. Nieuwe Instituut’s collection includes data that has been manually registered and collected by staff members over decades, and which is often incomplete due to practical limitations. While the available data and user needs continually informed the design process, the initial conceptualisation phase focused more on innovative design solutions. As a result, rather than addressing the data gaps, the final product exposed them, sometimes leading to dead ends. In retrospect, a data-driven approach at the beginning of the project would have been beneficial.

At the same time, the project has also accelerated the improvement of data quality by providing a clear focus on which areas should be addressed first, such as those published online. It has also highlighted the need to enable external enrichment, whether through human intervention in the Asterisk* project or automated enrichment via linked open data connections.

The Collection Platform is scheduled to launch in autumn 2024. Future plans include its continued development, informed by feedback and insights into user behaviour. The next critical step is to add features tailored to the needs of researchers and archivists, facilitating the phasing out of the current search portal. Another priority is to use AI to enhance advanced search functionality, making the collection even more accessible. The ultimate goal is to continuously link the Nieuwe Instituut’s collection with those of other cultural organisations to create a comprehensive knowledge graph of the field of architecture and urban planning.