Introduction

The ADS has contributed the largest amount of records to the ARIADNE portal and was the test case for the new AO-CAT data schema when this was implemented for the Catalogue in the ARIADNEplus Project. The AO-CAT was based on the original ACDM version developed for the ARIADNE Project (2014-19) using the experience gained to both streamline and extend the schema to encompass the depth and variety of data associated with archaeology, and aligning it with the CIDOC CRM. The ADS has now taken on the task of uploading and updating its own data in the ARIADNE Portal. In addition, it is using a tailored version of the Portal for the UNPATH’D Waters Project led by Historic England and involving a large consortium of UK universities, cultural heritage institutions and government agencies concerned with maritime archaeology. UNPATH’D Waters aims to bring together various diverse collections and archives to create a national collection about the UK’s rich and varied maritime past.

A logical progression

The ADS had already seen good results in loading its own metadata (datasets and grey literature), and inventory data from UK National Heritage Agencies, within the main ARIADNE portal. As the UNPATH’D datasets were from the same sources (so using the same native metadata schemas and native vocabularies), re-using ARIADNE was a logical progression. The UNPATH’D Waters Portal was, therefore, straightforward to implement using the AO-Cat resource type of ‘Maritime’ as the search filter and by changing the interface appearance (i.e. branding and images, etc.)

Learning the process

As part of the ARIADNEplus Transnational Access (TNA) programme, three of the ADS team attended the Implementing Interoperability Summer School organised by CNR-ISTI which provided them with a much wider understanding of the infrastructure as well as details on services such as the 3M mapping tool which is an integral part of the data preparation for the Portal. The team also used the detailed User Guide, the ARIADNEplus Data Aggregation Pipeline, which defines, for the archaeological data providers, the process by which their data should be uploaded to the ARIADNE Content Cloud, so that it appears in the ARIADNEplus Catalogue. After a short demonstration by Alessia Bardi, who was responsible for the data aggregation at CNR, the ADS team members did not have any difficulties in learning the process for managing their own data and that for UNPATH’D Waters.

Challenges that were overcome

One of the main challenges that several of the data contributors encountered in ARIADNEplus was getting the mappings of inhouse metadata schema to the AO-CAT right, i.e. consistent and expected  results when searching the uploaded datasets. Good subject knowledge and experience is required and it can take a few attempts, making small adjustments and enhancements before satisfactory results are obtained. ADS also found this when loading of the metadata stalled or went wrong, that it took time to understand why. Problems such as the use of the wrong mapping, or problems in the data have been overcome, but it took a little time and trial and error and the team have obviously gained experience from this.

Benefits of using ARIADNE

Despite the challenges, ADS appreciate the benefits of managing their own data for both Portals. When asked about this, Dr. Tim Evans, who oversees the ADS team, said:

“The process is flexible yet repeatable. So, for Unpath we’ve been able to wrangle ‘messy’ metadata into a simple iteration of Dublin Core XML, have this mapped to the AO-Cat, load, and then repeat. When we have more time we can get this formalized into something like OAI-PMH, but for the time being, it’s relatively easy to quickly turn something around and load. This flexibility means we have capacity to load more data should the opportunity arise. 

However, the main benefit is the end-product: we have data from across the various agencies that hold heritage data in the UK and the Isle of Man cross-searchable. This is a significant achievement and wouldn’t have been possible, so quickly, without the Ariadne infrastructure.”

Recommendations

Based on their experience, ADS would make the following couple of recommendations for other providers who may wish to take on their own data management:

  • Start with a smaller dataset with a relatively simple structure/schema. It’s easier to quality check the mapping onto the AO-CAT, but also to identify issues in your data.
  • It’s also beneficial to have a team or group of people working on this, so that skills/knowledge are shared.