Data Catalogue Launched!
One goal of Unpath’d Waters is to make it easier for researchers and the public to find maritime heritage resources.
The types of resources are rich and varied. They range from simple text entries that record that the wreck of an unnamed vessel is known to be at a specific location on the seabed, to large databases of finds from underwater excavations, such as that of the Mary Rose.
As digital resources are scattered across collections, we aim to create a unified catalogue. This should make it easier to find information, wherever it is held.
One of the goals of the Towards a National Collection programme (which provided funding for Unpath’d Waters) is to investigate the challenges of bringing information about such collections together, and hopefully to find some solutions.
Getting the data together
As a result of our data audit, we have already identified over 300 collections of interest, each one of which may contain thousands of items. We have already catalogued over 50,000 shipwrecks in UK coastal waters, ranging in date from the Bronze Age to the Second World War. We anticipate that the Unpath’d Waters catalogue will eventually hold over 100,000 items.
This information is currently managed by lots of different organisations, held in different databases, and sorted by different classifications. For example, there is no agreed national classification for types of ships across the UK.
Our data sources include national heritage agencies, museums, archaeological contractors, and specialist maritime bodies. These Unpath’d Waters partners are based across England, Scotland, Wales, Northern Ireland, and the Isle of Man.
Creating an ‘ontology’
To be able to search these collections we need to create an agreed way of describing all the different digital resources, or ‘ontology’. We used work by the Archaeology Data Service as inspiration. They worked on ARIADNE, which makes it possible to search over 3 million archaeological data resources from over 20 countries. They did this by creating their own ‘ontology’.
The ARIADNE ontology is based on the fact that almost all data resources can be described by a combination of
- What?” (the subject matter of the resource)
- “When?” (the archaeological period of the resource)
- “Where?” (the location of the resource)
We can use this ontology with just a couple of changes (such as also recording the depth as well as latitude and longitude to describe where a wreck is). However, we cannot force everyone to use the same ontology in their collections. This would be impossible given they use their own processes based on their specific collection. Instead, we will map each database to the Unpath’d Waters ontology. For example, rather than asking everyone to change their subject classifications, we’ll map their terms to an international standard. In this case, we’ll use the Getty Institute’s Art and Architecture thesaurus, which covers the majority of archaeological artefact and monument types.
Similarly, organisations define archaeological periods differently. For example, the Iron Age or the Viking Age have different start and end dates in England and Scotland. This doesn’t matter, as long as we know that. For this we are working with a US web service, PeriodO, which defines periods according to geographical regions.
Goals for the catalogue
Using these methods, we’ve already started to import standardised information (“metadata”) about Unpath’d Waters resources into our catalogue. We want anyone with the right skillset to be able to build websites and interfaces to investigate any aspect of the UNPATH catalogue they are interested in.
They should also be able to link it to other datasets, such as information about the natural environment. To allow this, the catalogue is held as Linked Open Data in an RDF triplestore. We’ll also share the catalogued data to existing online portals, including ARIADNE and MEDIN. This includes creating a version of the ARIADNE portal which gives a UK-only view of marine data for Unpath’d Waters.
In particular, this portal will allow users to ask complex and useful questions, like “What information is there about submarines that were sunk during 1914-18 within 40 miles of Liverpool?”
We’ve only just started this work, but already it’s clear that it will allow researchers to ask an entirely new set of research questions and give the public more access to the UK’s rich maritime heritage.
Also of interest
