Emerging from the First reflection workshop, London, July 22, was the idea that the National Collection was not something that already existing, patiently waiting to be connected via digital tools. Instead, we began thinking about the National Collection as something that is actively created through the connective work of humanities researchers, computer scientists, curators and archivists, and a whole raft of local and voluntary historians.

This meant that, as a project, we are not necessarily looking to develop singular pipelines aimed at creating a subject-specific instances of a National Collection, but rather, experiment with how multiple pipelines can feed into a National Collection that is created through different researchers undertaking cross-collections work. In a sense, we are using our systemic action research methodology to understand how multiple research pathways feed into the systemic whole.

Whilst understanding the role and technical infrastructural requirements of NLP, CV and other advanced digital techniques in helping to create connections between collections remains a primary interest of Congruence Engine, we are also interested in understanding how to develop and sustain participation in the construction of a digital national collection. Or, how the social infrastructural requirements - the relational aspects of a distributed, incrementally created national collection, can be properly considered and incorporated into future funding bids and National Collection projects.

National Collection as a verb has evolved into our work on National Collection as a social machine, but we still use it as a way of describing how the national collection does not exist waiting to be connected, but rather needs to be done, needs to be creative through active participation on national collection making.