Controlled subject vocabulary

[etonkin]

Hi,

I note that a previous issue has identified the Getty Thesaurus of Geographic Names (TGN) in use as a controlled vocabulary. I was wondering whether you could tell us anything more about the controlled vocabulary in use in the subjects section of the JSON files? for example:
"name": "actions: postures and motions"
or "name": "individuals: female"
From a linked data perspective any further information you could provide about this vocabulary (origins, [relationship with] standards/practices within the domain...) would be very helpful. Apologies if it's already documented and I've missed it!

Thanks so much for making this great resource available.

[sebastfr]

Hello - i think this is just the Tate taxonomy (some staff are curating it) - you may see some of it under "Find similar objects" on artwork's pages e.g. http://www.tate.org.uk/art/artworks/millais-ophelia-n01506 - I'm happy to get you in touch with one of the (data) curators if you want more information, let me know!

[alexpilcher]

Hi Emma,

Here's some further detail that I've been given by our main indexer. (I'm afraid that this is a long-winded way of reinforcing Seb's point that this is a bespoke taxonomy.)

"The subject index was originally created and developed alongside the digitisation of Tate's collection (a process which began in the late 1990s) as a means of providing extra keys into the collection by enabling visitors to search artworks via subject as well as artist name or artwork title. The design of the hierarchical structure and initial tagging of the bulk of collection artworks was carried out by a team of three indexers recruited for this specific purpose (all of whom had an art history background); with Tate's curatorial team (one designated member of staff from each specialist collection area) acting as advisers where necessary. The initial structure and key word terms were based partly on a previously developed card index which Tate's Information team had compiled in response to popular enquiries by in-gallery visitors.

"A number of pre-existing systems were consulted in the approach to planning the index. ICONCLASS was examined closely to help determine the logic of Tate's subject index – and its top-level hierarchies heavily influenced the structure of Tate's index. (The project’s senior indexer joined the project with six years of experience using ICONCLASS for the classification of images at the Witt Library). But the academic focus and complicated structure of ICONCLASS was not suitable for the sort of public-facing index we wanted to create.

"A system much closer to the type of index we wanted to create was COLLAGE – public-access system for the Guildhall Library and Art Gallery (http://collage.cityoflondon.gov.uk/collage/app). It enables users to search works of art by a variety of criteria including subject, and it uses a similar three level structure.

"Other systems evaluated and useful in planning the index were the Getty classification systems and the National Gallery's 'Micro Gallery'. The Getty's TGN (Thesaurus of Geographic Names) was useful as reference for the subject index's topographical 'places' category, and its online search facility was used as an authority reference for checking topographical locations.

"In terms of deciding the actual subject headings, we were primarily led by the nature of Tate's Collection. The index is very much a bespoke one in that it is designed specifically to reflect the scope of Tate's Collection – and many of the second and all third level terms were added during the process of indexing.

"The creation of the headings was done in consultation with an internal Client Board that included collection curators, education curators, Information staff and library and archive staff. Samples of indexed works were also sent to a group of specialist subject advisors made of Tate curators and interpretation staff throughout the course of the indexing process, who provided feedback and comments. These contacts were also available to answer queries we had in relation to subjects and art historical groups and movements, where no published interpretative information about the work was available."