Prima Facie: Inventories and Documentation

So far, we’ve talked about ethics and collections.

Inventories and documentation

Acquisitions, long-term loans and accessioning; Inventory control and cataloguing; Syntax and terminology; Object numbering, labelling and marking; Location and movement control; Backlog accessioning, inventory control and cataloguing; Manual and computer-based cataloguing and retrieval; Images; Web access to the information about the collection; Staff and Financial Resources; Recommended cataloguing fields.

George: Let me start by saying that, even with just 10 objects in the collection, I was already nearly overwhelmed by the prospect of maintaining inventory and docs. But, importantly, my primary goal for the entire exercise was to generate content about it, the exercise itself.

WordPress pages = object records

As I noted in Gathering Digital Content for the Archive, one of the research ideas we’re working on is making digital content, news and opinions about museum objects part of their official records. Our idea with the WordPress pages for each object was to give each object a home online, and then to add information and content to that page as we found it. We built out the initial views for each object from British Museum tombstone records, and other information uncovered on the highlight pages there. Harriet added Google maps where we could see a location, and images of the objects.

It was also good to use WordPress, I think, because we just get a whole bunch of web-native features right off the bat. Thirty8 Digital has been doing great museum-y WordPress-y work for ages.

Collecting more data gently about objects

This is an important idea for The Small Museum. How can we use the breadth of web content to our advantage? How does information and knowledge about these things change over time? This is what we’re referring to on the Good, Form & Spectacle homepage when we say “contextual benefits of networked metadata.”

To date, we’ve just added information about what we did with each object in the collection over the course of the residency at Somerset House. I can easily understand how that could become a full time job. I’m also interested in seeing what web scraping technology, RSS and other types of “web listening” could annotate an object’s official record with context from the web. All sorts of questions there.

This kind of idea is a bit different from tools like Archive-IT, from the Internet Archive. That’s designed to archive individual websites to see how they change over time. I’m imagining something that’s at the same time more fine-grained, but also with a broader reach. Something like a slurper that’s like Google News for each object.

As an example, it seems like the British Museum should be collecting the sort of content and opinion that’s being written about the Parthenon Marbles on The Guardian. Don’t you think?

Github, google doc

I’ve been thinking about the idea of a “minimum viable record” since I led the Open Library project at the Internet Archive.

The basic step is to collect as many identifiers about an object as you can. Once you have those stored in a giant array, you can use them to do lookups into their host services. It’s a giant headache to get two (let alone fifty) slightly different semantic systems to play nice, so I thought of this as a way to avoid that issue. (There’s still the challenge of recognizing an object across two systems though, to be sure. Even the distinction between vernacular, official and colloquial gets tricksy with primary/original objects, like Parthenon Marbles vs Elgin Marbles.)

You can see the early seeds of this minimum viable record in the columns we chose for the data, where we’re immediately trying to create connections from our records into BM, Wikipedia, and the 3D model repository, My Mini Factory:

Object Name, MIAB_ID, MIAB_URL, BM Explore URL, BM Collection URL, BM Open Data URL, Wikipedia, MyMiniFactory ID, Country of Origin, Place Found, Period, Date, Measurements, Material, Notes

I was also encouraged to see that Getty released its database of some 650,000 artists names onto the web the other day, available as Linked Open Data under the Open Data Commons Attribution License (ODC-By) 1.0. This is just good. It would be nice to be able to annotate that with IDs and URLs and URIs that refer to the same people. I’m also not the only person thinking about the need for radical simplification.


Harriet made some really handy labels for each object. They were handy because they contained basic metadata about their object, but could also be easily moved and positioned around the museum space. Here’s one for Goddess Hathor:

Day 6 Object of Focus
The Small Museum Version One The Small Museum Version One

They were simple and manoeuvrable.

Donation! Bequest!

We were all a bit surprised and excited when Dr. James Lattin brought us one of his own Museums in a Box – our first donation! The only inventory we’ve done so far of this event (and the new object) is the blog post I just linked to, a tweet or two, and another mention right here. This is exactly the sort of mention or tendril about an object it would be good to collect methodically.

Our first bequest was also utterly charming, thanks to its creators, Arthur and Henry Maxwell, aged four and two.

File 31-03-2015 10 36 13

Both of these events made me ever so slightly nervous that we’d quickly become a Museum of Museums in Boxes. Not the end of the world, but, I suspect that’s how a lot of small museums grow quickly.

There’s also a digital element to acquisitions here, about digital rights management. Acquiring something new is a great opportunity to clarify how a digital surrogate of an object may be used. Of course, I’m all talk, because I completely failed to do that in the course of these two events.


Prima Facie: Collections

So far, we’ve talked about ethics.

Collections management

Developing a collections management policy; Acquisition and accessioning; Deaccessioning and disposals; Numbering and marking of objects in the collection; Loans condition reports; Collections storage; Handling and moving collections; Photography; Insurance; Public access to collections; Display and exhibition galleries and rooms; Research of collections.

Harriet: Collections data has always struck me as hugely annoyingly inconsistent. How is it so hard to agree some basic principles and everyone work to them? Trying to find a specific British Museum object in the collections database has always flummoxed me. I never quite know which of the various numbering systems I should be using. Though I love that one of the systems is awesomely titled the ‘Big’ number. This helpful tip for collections online says it all really: The user may have to try several punctuation alternatives … it may be in a non standard format such as dashes rather than commas and full-stops or a zero appended to a number.

How can it be so hard?

Oh, turns out that it is really hard.

We had just 10 objects in The Small Museum Version 1. We put them in a spreadsheet and added the basics (name, place, date). Turns out those aren’t so basic after all. I’m still not sure our names are particularly consistent, we have Goddess Hathor, Buddhist goddess Tara, and Figure of Xochipilli. Should we have used Hathor, Tara, Xochipilli? And places and dates come with their own issues (more on that with the Inventories chapter). Luckily we have a straightforward 8-digit unique (big?) number…for now.

The web pages were another hub of information, I admit I was quite nerdy about these – the inner content producer/editor came out – but it was difficult to maintain consistency across just 10 objects. We didn’t have the same information for each and invariably some had more interesting acquisition stories and others had more interesting contextual stories. In the end I learnt to let go a little!

And it’s just struck me that we had two acquisitions/gifts/bequests. One from Dr James Lattin and one from Arthur and Henry. While we documented these on the blog, they haven’t been formally added to the register.

I think we had (and still have) good intentions about our data management – but it’s harder than I thought.

According to the handbook ‘…the most important museum collections document is the Collections Management Policy’. We haven’t got one of these. In fact the objects we 3D printed were those for which we had relevant files. A number of visitors asked ‘why did you select these objects?’ and that was a tricky one to answer. Perhaps we should get that policy document underway.

George: We were truly liberated by two things: 1) having a very small collection of ten objects, and 2) having those objects be facsimiles of existing things from another institution. The small collection let us not be distracted by having to deal with a giant collections management issue right off the bat. Having a starting position of 200,000 archaeological artifacts, as suggested by Ed Rodley in his 2012 Making a museum from scratch: Part One blogpost just immediately puts you underwater, and perhaps unable to consider questions other than what to do with those dusty, unclassified objects. Maybe, by flattening or simplifying that issue, but still having some objects, you can find a space to think about things other than the sheer weight of your collection. Is that crazy?!?

That said, we did still have a small collection to manage. We did some simple things to get started:

  • Made a simple spreadsheet to list each object
  • Gave each object a unique ID
  • Added links to various BM collection data sources
  • Made a page on the web for each object
  • Added imagery and narrative about each object through research of the BM catalogue

Digital Catalogue
The catalogue was very simple. Here are the column headings we selected, with an emphasis on digital descriptions and links into other datasets: Object Name, MIAB_ID, MIAB_URL, BM Explore URL, BM Collection URL, BM Open Data URL, Wikipedia, MyMiniFactory ID, Country of Origin, Place Found, Period, Date, Measurements, Material, Notes. We published it on Github too, because that’s what you should do.

The crazy part is, we’re already a bit behind, in particular because we had planned to supplement our original pages for each object on WordPress with information and content we made during our residency at Somerset House. You can see we’ve done that for some objects, like the Colossal Foot, but haven’t quite finished yet.  We generated a ton of great content about our daily displays, but it’s now scattered across several services online (Flickr, Twitter etc.) Ideally, each object’s catalogue record should connect to that documentary content too, so people can see the sorts of displays and blog posts we wrote about them. Content is hard to maintain!

Donations / Bequests
We were also absolutely thrilled and surprised to receive two museums in boxes from visitors! The first was a donation, and we decided to make the second one a fun bequest, because it was made by Arthur and Henry Maxwell, two smalls who belong to Harriet, and had made their own museum in a box after visiting. IT WAS ADORABLE, but we haven’t catalogued either of them anywhere yet, except a reference blog post.

Ephemeral Content/Connections
Gathering related digital things is a huge interest of mine. There’s so much the web can say to supplement the classic “tombstone” records that most collections contain. We made a gesture towards this idea with our addition of Wikipedia links in the catalogue, but there’s a LOT been written about each of our ten famous objects. The question begs, whether or not institutions should be starting to collect broader digital metadata and context from the web about their collections. I think the answer is yes, but, it’s bloody hard. Even just representing the residency itself across the various publishing platforms we used to make the thing is difficult to manage, and that’s just for ten things and ten days.

A later approach might be to consider a tool like the Internet Archive’s Archive-IT service, or look into the work of Ilya Kremer‘s pywb, both of which are actively developing around the idea of crawling and indexing specific slices of the web.

I know understand why it’s possible and quite common for folks who work in institutions to rarely be certain of the entire contents of their collections. It’s no longer a surprise that some institutions make amazing discoveries within their own collections. (Sandwich Magna Carta, anyone?) I now sympathize with all the archivists I’ve met who apologize for the state of their archive.

Content development and catalogue maintenance are laborious endeavours. You have to be on the ball and proactive, and that’s all there is to it.