Usage of custodialHist and custEvent

[tomhigginsuom]

At Manchester we've used custodialHist and custEvent in some collections to document exhibitions - is this a recommended use of the element?

e.g. https://www.digitalcollections.manchester.ac.uk/view/MS-PERSIAN-00051/1

<custodialHist>
<custEvent type="exhibition" from="1992-09-09" to="1992-12-21">
<p>Exhibited in <title>Gilded Word and Radiant Image</title> at the John Rylands Library, sponsored by <orgName ref="http://viaf.org/viaf/122924270" role="spn fnd" key="person_122924270">Altajir Trust</orgName>, <date from="1992-09-09" to="1992-12-21">9 Sept. to 21 Dec. 1992</date>.</p>
</custEvent>
</custodialHist>

[holfordm]

This strikes me as an appropriate use of the element and it would be good to agree best practice (the example looks good in fact) , document it in the guidelines, and create one or more scenarios for display in the XSLT transformation

[Dr-James-Freeman]

Looks good.

There has been some uncertainty here about whether custodialHist and custEvent ought also to be used to record superseded classmarks - or whether they should only go into altIdentifier within msIdentifier near the top of the file.

[ElizabethGow]

This element can also be used to record a photography/reproduction/digitisation events - we haven't really done this yet, but thinking about it for some 'treasures'.

[ElizabethGow]

Re. former references. We consistently use <altIdentifier type="former">. Would <custEvent> duplicate this information or replace it? The advantage of <custEvent>would be that the usage or change could be dated, which could support digital humanities reuse of the data (e.g. for visualisation). But I think it would be harder to search/browse and (perhaps) make it more difficult to create digital reconstructions of historic collections.

[Dr-James-Freeman]

I like the idea that custEvent permits a classmark to be datable. I think this element is preferable to altIdentifier type=former (and lessens the risk of confusion about which classmark is current - or (more likely) whether it is necessary to cite the present classmark and the others that are listed nearby (readers often don't know which to use, so to cover themselves use multiple classmarks - and separating the former classmarks from the current one by use of custEvent would help us to get round this issue.

[Dr-James-Freeman]

A further thought: altIdentifier is probably better applied for current but not necessarily internal reference numbers that are not the item's classmark as such: as the schema has it on ll. 758-782, for the summary catalogue, Oxyrhynchus papyri and Trismegistos references. These are all valid, current and alternative ways of referencing an item, whereas a former classmark is not.

(My one concern for the UL is that, applying this approach, the running numbers assigned to the manuscripts in the 19th-century catalogue might be taken as a valid way of referring to the collection, when they cover only a part of the materials we hold and have never been widely used as a means of referencing our collections, whereas the two-letter classmark sequence has.)

[mjhawkins]

It has been decided that custEvent is how we should record this. Our documentation needs to be improved to reflect these uses.

Further discussion is needed on type values.

[mjhawkins]

I have added placeholders for the longer documentation on custEvent as well as the elementSpect element (either so we can constrain attribute values, update the description that appears in oXygen or to include examples).