There is a creative company in Seattle that has been steadily publishing content for a single project for the past 25 years in around a dozen languages to a global community of tens of millions of people who not only engage with it individually but actively gather specifically to enjoy this work. The experiences it creates with its extraordinary mix of narrative, visual art, and design, are an ongoing process of discovery and exploration shared by a community as big as a medium-sized nation.
The company is Wizards of the Coast (WotC), and the endeavour is the original collectible card game, Magic: the Gathering. For the past 25 years, aside from a couple of hiccups, Magic has been growing steadily to its current considerable size. What’s more, the average length people stay engaged with the game is nine years – the typical game lasts one or two – and many play once or more a week, as opposed to the typical more sporadic play. In commercial terms, in terms of the dedication of its audience and the people-hours spent engaged with the work, in terms of its ongoing longevity and engagement with the public, and in artistic terms (or design terms if you have trouble with the idea of game design as art), Magic is one of the most successful creative works in history.
On the topic of success, Mark Rosewater, Magic’s Head Designer, has a saying he uses often when interacting with the players of his game: “Success breeds repetition.”
Most often he says this to explain the straightforward commercial mechanism that informs WotC’s design processes: if people like some element of a particular expansion, and therefore buy a lot of it, WotC’s designers tend to make more of that sort of thing.
But sometimes, especially when discussing game design, Rosewater uses his dictum in a more nuanced way. He is a big believer in consciously examining the structures around creative work and engineering them to prevent stagnation and creation-by-default. For example, he takes considerable care to ensure that he approaches every game expansion he designs from a different starting point to the previous sets he has made. This may be why he has been making Magic for 22 years straight, has made nearly 100 such projects in that time, and since he became Head Designer a decade and a half ago has presided over a long run of both quality design and audience growth.
When Rosewater says “success breeds repetition” in this context, it is a caution against success leading us to repeat things unthinkingly. Magic has been so successful for so long that it is easy even for a team of dedicated, highly-trained innovators to miss obvious and easy improvements. (For a somewhat involved example of an embarrassingly obvious fix to a nagging systemic problem that was missed for over a decade, see the extended endnote Solving the Small Set Problem.)
The key takeaway from Magic’s experience is this: it feels difficult to justify questioning your own assumptions when you are clearly doing well. First there is the initial trap of thinking that your success means there is no improvement to be made, or at least none necessary. Even if you avoid that, if your services are popular, it will always seem like a smarter move to devote all your resources to keeping on doing the thing everyone wants you to do, rather than taking time out to wonder whether you should still be doing it in exactly that way.
In other words, success breeds repetition even if repetition is undesirable and will get in the way of further success. The author or actor whose breakout hit leaves them reprising the same material over and over understands this only too well.
Unfortunately, I don’t believe libraries do, and I suspect that we may be on the road to learning it the hard way.
Parelthocracy in the library
Public libraries, and the library sector generally, are examples of the best and the worst tendencies of parelthocracy.
Benign parelthocracy: the library and the living past
On the one hand, the past is vitally important, and libraries’ commitment to giving their communities access to the past through their non-fiction and fiction collections is at the core of what it means to be civilized.
Without freely accessible evidence of where we have come from, both in the form of important historical works and in the form of new collections and interpretations of information about the past, both at the global or national level and at the local level, the public is ill-equipped to understand the present day and to avoid repeating the mistakes of the past.
And there is undeniably value in the inherent history of the library itself as an institution. In the same way that walking into a library connects us with all the other libraries throughout the world, there are connections to libraries and by extension human communities throughout time as well. (Pratchett’s L-space gave us a magical metaphor for this.)
The profound significance of all these things – the pragmatic, the symbolic, and the emotional – is a living example of the very best of bringing the past into the present. An institution without these things would be a book depository rather than a library.
Toxic parelthocracy in libraries
But in terms of decisionmaking, libraries (public libraries in particular, as that is where my personal knowledge is broadest, but I have seen these same tendencies across the sector) are prone to both palaeocracy and notocracy.
The problem is that we are so popular in our current configurations that we feel that that means we must be doing our job pretty well. In some respects that’s true; there’s no doubt that we add tremendous value to the communities we serve, with a recent study of libraries in Victoria finding that we give value to the community over 6 times greater than we cost to run. As the rest of government increasingly moves online, and libraries become one of the few places people can come for internet access and support in using computers, this number can only have gone up.
But let’s be real here, overly relying on public satisfaction figures is pre-emptively letting ourselves off the hook. We are:
- pretty much the only place left dedicated to genuine sharing, where people can get cool, useful stuff for free, with nothing asked except to return physical items on time and still in shareable condition (and an increasing amount of free electronic stuff as well),
- pretty much the only enclosed public space which is freely accessible to all and not dedicated to specific pre-programmed activities,
- pretty much the only place with a mandate to help people navigate the world of information and culture, in a time when vital services are moving online and becoming more bureaucratic as a result of automation,
- doing all the above in a framework of genuine, all-inclusive, secular public service.
In this context, getting 95%+ satisfaction ratings in customer surveys isn’t a definite sign of exceptional good work. All we can be sure it means is we aren’t visibly doing anything to anger significant numbers of our community enough to offset everything in the bullet points above.
But that’s not immediately apparent when we’re contrasted against other public infrastructure bodies, whose ratings are often far lower, because their work is designed to be taken for granted (e.g. rubbish collection) or is unwelcome to some part of its users (e.g. parking inspectors). Both the external bodies from whom we receive funding (and through them ultimately the public who take an interest in these things), and we ourselves, can be fooled into thinking we’re doing our job that much better than other public servants, and that there’s no pressing need to be critical of our own work.
In other words, because what we do is so important and makes us so well-loved, in some important respects we are failing to do our job at all.
This seems like a shocking statement when we work as hard as we do with the limited resources we have, but let me explain.
My understanding of the role of the library is that we enable free sharing of and self-directed access to culture and information, as well as a community space amenable to these sorts of pursuits. Our job is to know what’s out there, think critically about it, and offer our community tools, expert guidance, and assistance in engaging with it. A century ago, the best way to do this was to buy a bunch of books, store them in one place, sort and index them to make them easy to find, and provide knowledgeable staff to help people learn how to use our systems and access the information they needed, as well as provide useful context and guidance about further research.
But in the networked age, the possibilities for ways to empower our patrons (and the creators who serve them) are far broader and more powerful than that.
As a community of so-called information management professionals, we have been shockingly complacent about our role in building the new information economies and ecologies, instead leaving that to engineers who all-too-often think that solving a technical problem such that the “good” numbers go up and the “bad” numbers go down is an ironclad guarantee that you’re actually making the world a better place (and that there are no thresholds that those numbers should never cross), and who are unambiguously building for profit and therefore power, rather than placing the public good above other considerations.
Libraries, with our commitment to universal humanistic ideals and our deep, rich, ancient knowledge of what constitutes a healthy community of truth-, beauty-, wellbeing- and joy-seekers, of weighing the rights of individuals against each other to maximise everyone’s freedom and wellbeing, could only have had a beneficial influence on the evolution of the internet. We could have… if we’d only rolled up our sleeves and mucked in, instead of sitting on the sidelines watching other people make the informational tools and processes that shape public discourse, grumbling to ourselves about their inadequacies, but nonetheless teaching people to use them with only token warning about their flaws and perils.
And that missed opportunity to make a difference matters. Given the prevalence of misinformation and divisive, inflated, disrespectful rhetoric, and their effects on the present state of the world, it is no exaggeration to say that if there had been just 1% more library in the internet from the early days, 1% more resistance to bullying and bulldust, global geopolitics would look radically different in 2018.
Not only that, but we are systemically biased against new models of funding, publishing and distribution that have sprung up. This sounds harsh, but how many libraries have systems in place to monitor these channels and acquire works through them, despite crowdfunding platforms supporting literally billions of dollars’ worth of creative and informational projects? How many of us do anything to help our patrons find the considerable volume of quality culture and information released under various free licenses, or to support the creators who are generous enough to freely share their work? We are even passing the buck on our responsibility to curate our own commerically-published collections to external suppliers, who take our funds and give us slush that doesn’t sell elsewhere, and then we act dismayed when it doesn’t get borrowed much either. Given the contrast between the ways we could be spending our collections budgets and the ways we actually do, it is not unreasonable to characterise the way many of us purchase collections as being a form of corporate welfare for publishing conglomerates.
Why is this? Are we lazy? Indifferent? Corrupt?
We are none of these things. I am proud to be part of a hardworking, passionate, principled profession. But, ironically for information management professionals, we are not looking at the big picture and thinking critically about what we measure and what questions we ask, and therefore what we could be doing to fulfil our timeless mission of empowering our communities by sharing and helping people navigate information and culture. As a result we are letting our passion for our communities and our immediate goals keep us working hard at tasks that only partly fulfil our mission – living by our principles, but not reflecting on them.
And that is not entirely our fault. As we’ve noted, our funds are consistently less than we need to meet existing demand. The funding bodies that give us money ask us to measure things that reflect those old ideas of what libraries are and do, and expect us to do more of those things but with the same or fewer resources. The public who love us do so for what we have already done and want us to do more of the same – and fair enough.
Success is breeding repetition.
The thing is, it’s hard to blame the funding bodies or the public for that.
The funding bodies see our overwhelmingly positive feedback from the public – why would they ask us to change things?
And the public are busy leading their own lives – they trust us to think about the worlds of information and culture. While of course we should be open to good ideas from anywhere, especially our communities, it isn’t for them, or the funding bodies, to imagine that we could be and steer our profession towards that.
We’re the professionals. It’s on us to resist the siren song of success and make sure we give ourselves room to not just repeat our achievements, but build on them.
And that’s where we’re failing.
Let’s look at the specifics of how, broken down into notocracy and palaeocracy.
Notocracy in the library
I’m starting with notocracy because, thankfully, notocracy among library staff themselves is relatively rare. Where it exists, it takes the form of active resistance to media other than books, regarding them as inferior or, more generously, “not a library thing”; and generally, the ”newer” the media to libaries, the greater the resistance. (By contrast, non-notocratic staff love books but are not hostile to other media per se.)
Literacy and books are absolutely core parts of the library mission, but to disparage other media rather than dispassionately recognise the strengths, weaknesses, and value of all media and help our communities engage accordingly is to live in denial of the full range of ways in which ideas and experiences can be expressed and shared. It’s hard to see how we can adequately discharge our duty to the public under those conditions.
Notocracy is more common among the ancillary industries who make their livings from libraries’ existing ways of doing things and are understandably reluctant to see competition for collections and library managament system (LMS) budgets. The vested interest there makes it not only easy but imperative to set aside their concerns: their job is to empower our work, not the other way around.
Palaeocracy in the library
As I hope the above makes clear, palaeocracy is the overwhelming majority of toxic parelthocracy in libraries.
It takes two forms: systemic palaeocracy, where our systems are designed in such a way as to leave little to no room for experimentation and innovation, especially of any substantive kind, and cultural palaeocracy, where the importance of the work we already do blinds us to the necessity of applying our core mission and principles to the culture as it currently exists, not as it was last century.
Both are driven by a sense of insufficient time and resources to do our jobs, and in some cases, particularly smaller rural library services, it’s hard to argue that.
But if your library service has a collections budget that is more than 10 times the wage of a single worker, and hundreds of linear metres of shelves which are regularly two-thirds full or more, you clearly have some wiggle room. And if you are routinely weeding large numbers of books that are still lendable in terms of both condition and currency (for example, a mint condition Windows 3.1 manual would not be current), either you have an incredibly neat and considerate community, or – I would argue – you are overspending on collections that your community isn’t using.
This is not a problem in and of itself (better a slight oversupply than an undersupply), but does suggest that you are underspending on other resources, staffing, and/or tools and programs to help your community make the most of the materials you do stock. Which in turn certainly means you are underspending on business intelligence, strategy development, and innovation.
Preventatives for palaeocracy
What does adequate spending on these things look like? Well, it involves actually having budget lines for them, for starters; I’m not sure many libraries do.
It requires looking for (and spending money getting) insights not only into the library trade but into the worlds of culture and information beyond. (It should not be so easy for a random solo punk like me to surprise heads of library services with facts about where and how the public are spending their cultural dollars, but in my consulting work I do – and not only when it comes to games.)
It means understanding that marketing is not just promotion. Marketing involves listening to the market and using that information to shape your offering as much as attempting to push your own product once it’s made. How many libraries spend much time monitoring their community on social media and sharing those insights with the staff body?
It means setting time and resources aside for staff to meet to share ideas, insights and tips more than a few hours of a highly structured and top-down-directed agenda every few months.
It certainly entails recognising and cultivating the staff who contribute ideas, looking into barriers that might prevent other staff from contributing too, and having a channel to meaningfully and visibly feed frontline staff ideas into decisionmaking and resource allocation conversations.
It means understanding that an “innovation” budget that’s entirely predetermined by the same people who make the other budget decisions a year or more ahead of time, and what the innovation is going to be, is largely missing the point, and far less effective than having some capacity to allow staff to follow interesting ideas as they arise. Budgeting for innovation means being ready for opportunities and ideas when those ideas, and especially opportunities, arise – not a year and a half later after a budget submission and approval process. It also means being far more willing to make the case for varying budgets than libraries, public libraries at least, traditionally are.
It also requires that we recognise that part of our role is not just to offer things to the public but to tell the story of why they might matter. Underestimating the extent to which the public needs assistance to even recognise or understand your offerings, and how they might be useful to them, is a particular problem when it comes to new tech.
For example, I’ve heard complaints about 3D printers and makerspace tech generally being underutilised, but I’ve also seen library services whose communities make good use of them. Now, it is certainly true that there will be more interest in some communities than others, but there is always, always a strong correlation between uptake and the energy and visibility with which the devices have been promoted to the community. Maker tech is potentially relevant to DIYers, kids, STEM students, design students, designers, artists, crafters, tchotchke-makers, random tinkerers… the list goes on. But most people are still only dimly aware of them, and have no idea of the kinds of uses they can have. To judge the relevance of entirely new tech – and 3D printers are very new, and transformative in all kinds of ways – without first taking the time to ensure you’ve given the public plenty of opportunity to grok it is a highly palaeocratic move.
(And don’t even get me started on “online safety” classes. Fiddling with your Facebook privacy settings is a sick joke if you’re not first having a serious conversation about Facebook itself, and the hidden empires of profiling algorithms that chitter and scurry behind its façade – and much of the rest of the web as well.)
The wisdom of risk
But of course, what I’m arguing for here is devoting time and resources to things we don’t know will work, when there are so many things we know do work. It’s understandable to want to stick with the old reliables – doing anything else feels like a gamble.
Here I’m going to back to Magic Head Designer Mark Rosewater again. This may seem of dubious relevance, but hear me out: Magic, as I said in the introduction, is a game of discovery and exploration; and libraries are places of discovery and exploration.
From time to time, the Magic team make a misstep and put out a card that turns out to be so unexpectedly powerful in some way that it needs to be banned or restricted in tournament play. Rosewater always acknowledges the specific mistake, of course, but makes the point that if this never happened he would be more worried – because it would indicate that they were being too conservative in their card designs. Or as he puts it, “Never taking any risks is the biggest risk of all.”
In other words, human fallibility being what it is, Rosewater recognises that to achieve exceptional things you need to accept that failure is possible. Look to prevent it, and to minimise it where prevention fails, by all means, and to learn from it when it happens so that you don’t make the same mistake twice, but accept it as the inevitable price of the excellence and innovation that you are striving for.
Are lessons from a game design company applicable to a public institution like a library? Are we supposed to be as driven to innovate?
Well, I’m not actually arguing that we should be as driven to innovate, but I think the burden of proof is on those who argue that at least some of that drive shouldn’t be systemically built into what we do.
We know that the world is rapidly changing, especially in the spheres of information and culture, and the demands and opportunities it creates for our communities are changing with it. Our mission is to help our communities engage with the world’s information and culture. Given these two facts, we have two alternatives: either we expand the things we do, to ensure our mission keeps pace with those wider changes; or we curtail our mission so that it is only about doing the specific things we already do.
I am not inclined to agree that our mission is worth limiting in that way.
And given that, we have a clear need to foster innovation, not only to respond to the ceaseless stream of invention and creation in the wider world but in order to proactively promote the core library values of democratic inclusion; truth and wisdom; free inquiry and exploration; and beauty, fun, and joy.
If we were building a library for the first time now, without any baggage of historical assumptions, but with an eye to the needs of our community now and into the future, what would it look like? That’s the question we need to answer; and the answer is what we need to strive towards.
Endnote: Solving the small set problem
To understand this example, you need to know two things about Magic: the Gathering.
- Magic is often played in a draft format, where the players take turns choosing cards from the same pool. In the most common draft format, each player is given three packs of cards which they open, draft a card they want to use in their deck, and then pass the remainder around. This pattern continues – draft a card, pass the pack on – until the pack is all drafted, at which point the next pack is opened. Once all three packs are drafted, each player takes the cards they drafted and builds a deck, with which they then compete in a tournament.
- Magic releases new sets of cards regularly, and from about the third year of the game, these sets were made in “blocks”, groups of three sets in a kind of trilogy. The first, large set in the block provided baseline effects, established the gameplay themes and the narrative premise and setting, and was followed by a couple of smaller expansions that continued the narrative and evolved the mechanics of play.
For around the first 15 years of Magic, or more precisely of drafting and the block structure, drafting used to follow the same pattern as the block: start with one or two packs of the large set, then move on to one pack of each the small sets which evolved from the first. (So AAA when only the large set had been released, then AAB when the first small set was released, then ABC when the whole block was out.)
But WotC had a problem with the small sets: despite by their nature being interesting evolutions on the themes of the first set, they didn’t feel like they had enough impact on the draft, which made them feel less exciting to a sizeable contingent of players, which in turn reduced their sales and meant that, while the world in which the block was often well-known and -loved, players often had no idea of how the second and third parts of the blocks’ stories turned out.
It took over a decade before anyone suggested the simple expedient of opening the cards from the newest set first and allowing those to set the agenda for subsequent drafting. Exactly the same mix of packs, just change the order (from AAB to BAA or ABC to CBA), and all of a sudden the small sets are impacting the draft much more effectively.
Since breaking free of the old default, they have gone on to change the numbers of each set in the drafting mix (BAA became BBA), to change the default composition of blocks (1 large set, 2 small became 1 large, 1 small), and finally to do away with “blocks” altogether and just design large sets for each quarterly release. Escaping the grip of that legacy decision not only fixed the problem they could see, the “small set problem”, but freed them up to experiment and innovate to something that works far better.
Now, this is a company full of literal geniuses. The game was created by a professor of mathematics who went on to design many more smash hit games and to write a pioneering textbook about the history and design of games, and other designers have included accomplished writers, artists, biologists, ecologists, and an actual-no-jokes rocket scientist.
And yet although the problem I’ve described was clear in retrospect (though bear in mind that the simple fact of me having to explain the problem in a way that is comprehensible to readers with no knowledge of the game serves to make it even more readily apparent here than it was at the time), a solution that simple eluded them for years. Why?
Well, part of the problem was that – as we’ve seen – success breeds repetition.
At the time they made the change, Magic was a tabletop game with an international audience in the high millions. It has since grown into the tens of millions, which gives you an idea of both its growth trajectory at that moment and the success of these changes. The sets weren’t performing as well as WotC would have liked, but they weren’t exactly failing. And it’s not illogical to think that if you want something to succeed more, you just do more of the thing that’s making it succeed, namely design it as well as you can.
(The term for this particular trap is “local maximum”. The phenomenon is analogous to trying to climb the highest mountain in the world by always going upwards from wherever you are right now. Unless you’re lucky enough to actually be at the base of the mountain, and have a straight upward path to the top, at some point you will be at the top of a hill and be unable to go higher.)
And when you’re thinking about the sales outcomes of a particular set, the default is to look at the properties of the set itself. Magic’s designers were smart enough to also consider their small sets in the context of the design of the whole block, and even of the blocks either side, so they were already thinking more strategically than usual.
The other key part of the problem was that thinking about the draft order wasn’t really anybody particular’s job. It was the designers’ more than anyone else, but they had a whole slew of much bigger and more clearly defined responsibilities – namely, designing all the cards that would be printed – to tight deadlines every quarter… and then, as the popularity of the game grew and they started making more products, more frequently still.
So even at a company rightly famed for innovation and creativity, a combination of success and poorly-structured distributions of work can produce “innovation dead zones” – areas of work where defaults go unquestioned and cause systemic problems for years. Frankly, all things considered, I think it’s to WotC’s credit that they spotted the problem as soon as they did – though I have wondered if a helpful fan or two might have helped prompt the change (one of the benefits of having, and listening to, such a large community – the old open-source truism Linus’s Law, “given enough eyeballs, all bugs are shallow”).
What are your institution’s blind spots? And to whom are you willing to listen to find them?