This weekend past I ran the Generous & Open Galleries, Libraries, Archives & Museums (GO GLAM) Miniconf at LinuxConf.au, with Bonnie Wildie. Being a completely online conference this year, we had an increased pool of people who could attend and also who could speak, and managed to put together what I think was a really great program. I certainly learned a lot from all our speakers, and I'll probably share some thoughts on the talks and projects later this year.
I also gave a short talk, about my new Library Map project and some thoughts on generosity in providing open data. Unfortunately, Alissa is completely right about my talk. The tone was wrong. I spoke about the wrong things and in the wrong way. It was an ungenerous talk on the virtues of generosity. I allowed my frustration at underfunded government bureaucracies and my anxiety about the prospect of giving a "technical" talk that "wasn't technical enough" for LCA to overwhelm the better angels of my nature. I won't be sharing the video of my own talk when it becomes available, but here is a short clip of me not long after I delivered it:
So I'm trying again. In this post I'll outline the basic concepts, and the why of Library Map - why I wanted to make it, and why I made the architecture and design choices I've made. In the next post, I'll outline how I built it - some nuts and bolts of which code is used where (and also, to some extent, why). You may be interested in one, or the other, or neither post 🙂.
The Library Map is a map of libraries in Australia and its external territories. There are three 'layers' to the map:
The libraries layer shows every public library in Australia, plus an indicative 800m radius around it. Also mapped on additional overlays are State and National libraries, Indigenous Knowledge Centres, and most still-operating Mechanics Institutes.
The Rules layer has two overlays.
The Fines overlay colour-codes each library service area according to whether they charge overdue fines for everyone, only for adults, or not at all.
The Loan Periods overlay uses patterns (mostly stripes) to indicate the standard loan period in weeks (2, 3, 4, or 6 as it turns out).
Library Management Software
The Library Management Software layer works basically the same as the Rules layer, except it colour codes library services according to which library management system (a.k.a Intergrated Library System) they use.
What is the library map for?
I've wanted something like this map at various times in the past. There is a fair amount of information around at the regional and state level about loan periods, or fine regimes, and even library management systems. But a lot of this is in people's heads, or in lists within PDF documents. I'm not sure I'd call myself a 'visual learner' but sometimes it is much clearer to see something mapped out visually than to read it in a table.
The intended audience for the map is actually a little bit "inside baseball". I'm not trying to build a real-time guide for library users to find things like current opening hours. Google Maps does a fine job of that, and I'm not sure a dedicated site for every public library but only libraries is a particularly useful tool. It would also be a nightmare to maintain. The site ultimately exists because I wanted to see if I could do it, but I had — broadly — two specific use cases in mind:
Mapping library management systems to visualise network effects
My talk at LCA2018 was called Who else is using it? — in reference to a question library managers often ask when confronting a suggestion to use a particular technology, especially something major like a library management system. This is understandable — it's reassuring to know that one's peers have made similar decisions ("Nobody gets fired for buying IBM"), but there are also genuine advantages to having a network of fellow users you can talk to about shared problems or desired features. I was interested in whether these sorts of networks and aggregate purchasing decisions might be visible if they were mapped out, in a different way to what might be clear from a list or table. Especially at a national level — I suspected there were strong trends within states in contrasts between them, but didn't have a really clear picture.
The State Library of Queensland was invaluable in this regard, because they have a list of every library service in the state and which library management system they use. When visiting library service websites it turned out that identifying the LMS was often the easiest piece of data to find — much easier than finding out whether they charge overdue fines! It turns out there are very strong trends within each state — stronger than I expected — but Western Australia is a much more fractured and diverse market than I had thought. I also discovered a bunch of library management systems I had never heard of, so that was fun. This layer is the most recent — I only added it today — so there may still be some improvements to be made in terms of how the data is displayed.
Mapping overdue fines
The second thing I wanted to map was whether and how libraries charge overdue fines, but my reason was different. I actually started the map with this layer, as part of a briefing I gave to some incoming Victorian local government Councillors about what they should know about public libraries.
Here, the goal is mapping as an advocacy tool, using the peer pressure of "who else is charging it?" to slowly flip libraries to go fine-free. Fines for overdue library books are regressive and counter-productive. I have found no compelling or systematic evidence that they have any effect whatsoever on the aggregate behaviour of library users in terms of returning books on time. They disproportionally hurt low income families. They need to go.
In Victoria there has been a growing movement in the last few years for public libraries to stop charging overdue fines. I wasn't really aware of the situation in other states, but it turns out the whole Northern Territory has been fine-free for over a decade, and most libraries in Queensland seem to also be fine-free. I'm still missing a fair bit of data for other states, especially South and Western Australia. What I'm hoping the map can be used for (once the data is more complete) is to identify specific libraries that charge fines but are near groups of libraries that don't, and work with the local library networks to encourage the relevant council to see that they are the odd ones out. I've worked in public libraries and know how difficult this argument can be to make from the inside, so this is a tool for activists but also to support library managers to make the case.
As if often a problem in libraries, I had to define a few terms and therefore "normalise" some data in order to have it make any sense systematically. So "no fines for children" is defined as any system that has a "younger than" exclusion for library fines or an exclusion for items designated as "children's books". Some libraries are fine free for users under 14, others for those under 17, some only for children's book loans and so on. On my map they're all the same. The other thing to normalise was the definition of "overdue fine", which you might think is simple but turns out to be complex. In the end I somewhat arbitrarily decided that if there is no fee earlier than 28 days overdue, that is classified as "no overdue fines". Some libraries charge a "notice fee" after two weeks (which does count), whilst others send an invoice for the cost of the book after 28 days (which doesn't).
As the project has progressed, some things have changed, especially how I name things. When I first added the Libraries layer, I was only looking at Victoria, using the Directory of Public Library Services in Victoria. This includes Mechanics Institutes as a separate category, and that seemed like a good idea, so I had two overlays, in different colours. Then I figured I should add the National Library, and the State Libraries, as a separate layer, since they operate quite differently to local public libraries.
Once I got to Queensland, I discovered that the State Library of Queensland not only provides really good data on public libraries, but also had broadly classified them into three categories: "RLQ" for Rural Libraries Queensland, a reciprocal-borrowing arrangement; "IND" for Independent library services, and "IKC" for "Indigenous Knowledge Centre". The immediate question for me was whether I would also classify any of these libraries as something different to a "standard" public library.
The main thing that distinguishes the RLQ network from the "independents" is that it is a reciprocal lending network. In this regard, it's much the same as Libraries Victoria (formerly the Swift Consortium), or ShoreLink. There are other ways that rural libraries in Queensland operate differently to urban libraries in Queensland, but I don't think these differences make them qualitatively different in terms of their fundamental nature.
But what about Indigenous Knowledge Centres? I admit I knew very little about them, and I still only know what I've gleaned from looking at IKC websites. The Torres Strait Island Regional Council website seems to be fairly representative:
Our Indigenous Knowledge Centres endeavour to deliver new technology, literacy and learning programs to empower our communities through shared learning experiences. We work with communities to preserve local knowledge and culture and heritage, to keep our culture strong for generations.
The big difference between an IKC and a typical public library is that the focus is on preserving local Indigenous knowledge and culture, which does happen through books and other library material, but is just as likely to occur through classes and activities such as traditional art and dance.
But the more I looked at this difference, the less different it seemed to be. Public libraries across the world have begun focussing more on activities and programs in the last two decades, especially in WEIRD countries. Public libraries have always delivered new technology, literacy and learning programs. And the Directory of Public Library Services in Victoria amusingly reports that essentially every library service in Victoria claims to specialise in local history. What are public libraries for, if not to "keep our culture strong for generations"?
Yet it still felt to me that Indigenous Knowledge Centres are operating from a fundamentally different mental model. Finally it dawned on me that the word "our" is doing a lot of work in that description. Our Indigenous Knowledge Centres, keep our culture strong for generations. I was taken back to a conversation I've had a few times with my friend Baruk Jacob, who lives in Aotearoa but grew up in a minority-ethnicity community in India. Baruk maintains that public libraries should stop trying to be universally "inclusive" — that they are fundamentally Eurocentric institutions and need to reconcile themselves to staying within that sphere. In this line of thinking, public libraries simply can't serve Indigenous and other non-"Western" people appropriately as centres of knowledge and culture. I could see where Baruk was coming from, but I was troubled by his argument, and the implication that different cultural traditions could never be reconciled. As I struggled to decide whether Indigenous Knowledge Centres were public libraries, or something else, I think I started to understand what Baruk meant.
I'd been thinking about this back to front. Indigenous Knowledge Centre is a usefully descriptive term. These places are centres for Indigenous knowledge. The problem wasn't how to classify IKCs, but rather how to classify the other thing. The activities might be the same, but the our is different. I thought about what a non-Indigenous Knowledge Centre might be. What kind of knowledge does it want to "keep strong for generations"? I thought about all those local history collections full of books about "pioneers" and family histories of "first settlers". If it's not Indigenous knowledge, it must be Settler knowledge. When I first saw this term being used by Aboriginal activists in reference to non-Indigenous residents generally, and white Australians specifically, I bristled. I mean, sure, the modern culture is hopelessly dismissive of 60,000 years of human occupation, culture and knowledge, but how could I be a "settler" when I have five or six generations of Australian-born ancestors? But a bit of discomfort is ok, and I have rather hypocritical ideas about other settler-colonial communities. It's exactly the right term to describe the culture most Australians live in.
So I renamed "public libraries" as "Settler Knowledge Centres". I initially renamed the National & State Libraries to "Imperial Knowledge Centres", but later decided it was more accurate to call them "Colonial Knowledge Centres". I also briefly renamed Mechanics Institutes to Worker Indoctrination Centres, but that's not entirely accurate and I realised I was getting carried away. I wasn't completely oblivious to the fact that this nomenclature could be a bit confusing, so I cheekily created two views: the "General" view which would be the default, and a second view which would appear on clicking "View in White Fragility mode". This second mode would show the more familiar names "Public Libraries" and "National & State Libraries".
While I was doing some soul searching this morning about my GO GLAM talk, I continued to work on the map. My cheeky joke about "White fragility mode" had made me slightly uncomfortable from the moment I'd created it, but I initially brushed it off as me worrying too much about being controversial. But I realised today that the real problem was that calling it "White fragility mode" sabotages the entire point of the feature. The default language of "Settler Knowledge Centre" and "Colonial Knowledge Centre" sitting next to "Indigenous Knowledge Centre" is intended to invite map users to think about the work these institutions do to normalise certain types of knowledge, and to "other" alternative knowledge systems and lifeworlds. The point is to bring people in to sit with the discomfort that comes from seeing familiar things described in an unfamiliar way. Calling it "White fragility mode" isn't inviting, it's smug. It either pushes people away, or invites them to think no more about it because they're already woke enough to get it.
So today I changed it to something hopefully more useful. General mode is now called Standard Mode, and White fragility mode is now called Colonial mode. It's the mode of thinking that is colonial, not the reader. Flicking to Colonial Mode is ok if you need the more familiar terms to get your bearings: but hopefully by making it the non-standard view, users of the map are encouraged to think about libraries and about Australia in a slightly different way. They don't have to agree that the "standard mode" terminology is better.
So that's some background behind why I started building the map and why I made some of the decisions I have about how it works. You can check it out at librarymap.hugh.run and see (most of) the code and data I used to build it on GitHub. Next time join me for a walk through how I made it.