You'd think we'd know this. After all we're the Collections Trust. But a collection can be a remarkably tricky thing to pin down. They can be a group of items, or a single item, or (as we learnt some time ago during the Cornucopia project) a house. It can be a collection because somebody collected it, or because all of the things in it are linked thematically. One thing can be part of lots of collections, and it is even theoretically possible to have a collection with no things in it at all (because they have been deaccessioned, but their ghost remains in the form of a collections-level description).
The Collections Trust has been giving a lot of thought to this subject recently because of the work we're doing on SPECTRUM Digital Asset Management, due for publication some time in April. What we're trying to do is show how the management of digital assets and their associated information can be integrated alongside the existing practices of curatorship, collections management, preservation and interpretation.
To do this, we need a reasonably logical and simple model of what the stuff actually is that museums are managing through these processes. Hence the question 'What's a Collection'?
So we'd like to propose the following to you, dear reader, and ask for your views and thoughts on it. Very broadly, the model we are developing looks like this:
The concept underpinning this is that museums have increasingly had to adopt management methodologies which combine content creation, knowledge management, rights, access, user control, location and a range of other activities into a coherent framework. This is a challenge not only for the systems they use, but in terms of skills, capacity and cost. The broad headings in the diagram break down more or less as follows:
These are the material objects in the collection. They might be beetles, rock samples, works of art or warships, but it remains a fundamental role to collect, preserve, interpret and provide access to groups of physical things. These things are likely to have physical attributes (size, shape, colour, weight) as well as material properties (hardness, resistence to light, temperature, humidity) which in turn have an impact on how they are managed, stored and moved.
This object-centric model of environmental management and conservation is the central tenet of the BSI Publicly Available Specification 198 was developed by the National Archives with support from the Collections Trust (see the excellent TNA blog 'A Really Useful Environmental Standard').
The brilliant 'Guide to Digital Asset Management' from the Canadian Heritage Information Network (CHIN) defines Digital Assets as follows:
“Digital materials created or owned by your institution.
Digital assets exist in a variety of formats, and can include text, web, audio, video and image files. Digital images of objects in your collection are digital assets, as are logo image files, corporate Powerpoint presentations and any other digital resources created by your institution that generate revenue or that provide valuable content to employees or clients.
Digital assets may be used in many contexts, including sales, marketing, education, web development, collections management and digital preservation. Sometimes you will see the term 'media asset' used to refer more narrowly to audio or video content.”
Galleries, libraries, archives and museums are all managing ever-increasing quantities of digital assets, whether they are born-digital, acquired, captured, digitised or otherwise created. Although they are made of electrons rather than atoms, code rather than wood or steel, digital assets nevertheless require storage. They have different formats, requiring different tools for management, playback and re-use.
Whether you are a gallery, library, archive, museum, cathedral or private collection, making the most of your physical and digital collection means creating and capturing 'administrative' information about it. This kind of information forms the primary focus of the SPECTRUM standard, with its units of information and information groups. It includes things like location, valuation, risk, conservation and the kind of information that tends to crop up as the objects and digital assets move through their lifespan in the organisation.
This administrative information arises from the processes of managing the physical and digital collection. In the 80's and early 90's it was all about efficiency and accountability - where is it? Why have we got it? Can we find the shelf it's on? More recently, however, we have started to think of this administrative information as 'metadata' and as such intended to support strange new processes like discovery and use.
This covers knowledge about the subject, thematic or informational content of the object. It is generally, but not always, created by the organisation and can be based on expertise or research in a particular subject discipline. This kind of knowledge used to be called 'scholarship', and has been the subject of intense debate in recent years because of the apparent partisanship of expertise and the tendency it has to make people outside the organisation feel disempowered. In practice, however, museum audiences depend to an extent on there being someone in the organisation who knows about the subject to which the collections relate.
If 'Collections-based Knowledge' is, more or less, fact-based, then 'Narrative Information' is much more subjective - it includes stories, thematic links, conceptual ideas provoked by the collection as well as some kinds of user-generated knowledge and reflection. It can run the gamut from professional technical expertise to a deeply personal emotional reaction.
We think that the future relevance and adaptability of museums (and to an extent libraries, archives, galleries etc) depends on being able to bring all of these elements naturally into orbit around each other, and to continue to develop them so that we can develop services that are more open and participatory without losing the essence of professional intepretation and research.
But this is just one possible model. We would love to hear from you about whether this reflects your working practice. What have we missed? Does this model make things artificially tidy? Please add your comments below!