Return of the Coffeehouse: How to turn your library into an ideas factory

13 December 2011

Recently I read Steven Johnson’s Where good ideas come from: the natural history of innovation.  It’s a fascinating book, full of anecdotes about Charles Darwin and MIT’s Building 20.  Johnson uses Darwin’s life and work to highlight his main themes and uses coral reefs as an metaphor for “ideas ecosystems”, but of course I couldn’t help thinking about how libraries can use his ideas in the way we build and run libraries.

Throughout history libraries have been highly effective as what we might call idea storehouses.  Universities and schools have been highly effective as idea communicators.  But, particularly at a time when many are questioning the relevance of libraries (thinking in terms of the ‘storehouse’ model), might we develop libraries further as idea factories?  The place you go to generate ideas in the first place?

Johnson highlights what he considers to be the essential ingredients of an idea-generating environment: liquid networks, slow hunches, serendipity, error, exaptation & recycling and platforms.  Let’s explore what these mean in practice, and how libraries can encourage them to become idea generation spaces.

Liquid Networks

The concept of liquid networks is what in other contexts is called cosmopolitanism or multiculturalism.  Someone with liquid networks mixes with lots of different types of people,  picking up ideas and practices they would not encounter in a mono-cultural, small network.  The concept of ‘liquidity’ is important, because it comes with a related idea of ‘idea spillage’.  When the physics department and the School of History share a tea-room, ideas will ‘spill’ from one to the other, potentially leading to completely new concepts or discoveries.

In public libraries at least, we like to think we naturally encourage ‘liquid networks’ and ‘spillage’ because we are open to anyone.  But is this really true?  Is your library open at times that are convenient to a range of people?  If you are only open 10am to 6pm Monday to Friday, your library is ensuring that full time workers will rarely mix with retirees, the unemployed or students.  Is your library really welcoming to all, or is it always full of middle-class educated people looking for the latest Booker prize winner?

Citizens, no less than guests at a party, need 'hosts' to show them how to come together. - Alain de Botton

If you want to ensure your library encourages liquid networks, you need to go out of your way to ensure spontaneous (and planned) encounters between different types of people occur.  Should you be running programs where students teach computer skills to retirees and learn in turn about gardening, or crochet, or small engine repair?  Could you run a hotrod-themed poetry slam night and attract two different interest groups?  Do you simply need to extend your opening hours?  Could you run an ‘investor club’ where local business-people meet young people with ideas for start-ups?  Even something as simple as ensuring your library is easily available as a place for clubs and hobby groups to meet will help make your community a place where ideas can happen - Johnson tells us that inventors are more likely than others to have a large number and wide range of hobbies.

One of the things that intrigued me about Johnson’s descriptions of effective ‘liquidity’ is that if networks are too open it actually detracts from successful idea-generation.  Anyone who works in an open-plan office knows what he means.  He cites MIT’s Building 20 as the perfect balance.  Buliding 20 is famous for being the home of an abnormally high number of successful new concepts, ideas and discoveries.  Johnson argues that this is because it was always considered a ‘temporary’ structure and thus when researchers and students felt the need to internally reconfigure their working space (eg by knocking out or buliding a wall, or running a cable from one office to another) they simply did it, often without worrying about asking permission or having to get professional workmen in to do it.  This meant that although researchers still had their own offices to develop their ideas and work in some quiet, they could easily adjust the workspace as needed.

For libraries, this gives us something to think about during major renovations or new builds.  Many libraries, particularly those built since the 1960s, are an open-plan design.  This is great for showing of airy and light-filled architectural masterpieces, but is also the genesis of the daily struggle librarians have trying to reconcile the differing needs of mature-age students and self-directed learners in study areas with boisterous storytimes.  A mix of shared and fit-for-purpose spaces may be best for generating ideas.

In my own library service, as in many others I know of, we constantly struggle to balance the requirements of a twenty-first century information service with the physical restrictions of 1970s-era buildings.  Library managers need to ensure that the architect’s brief includes a requirement to build flexibility in to the design right from the beginning.  The ability to run cabling through the ceiling or floor (or along tracking) easily to anywhere in the building is essential.  But try to think about building flexibility in to the building in other ways as well.  Can you easily put in or take out internal walls if necessary in the future?  Could you put all your shelves on castors so they can be re-arranged on a day to day basis for events?  Could you use the same space for events in the evening and study space in the day time?  Or vice-versa?  No doubt we’ll be using technology in 20 years that doesn’t even exist now.  Your building is likely to last longer than your career - ensure that it’s just as flexible as you will need to be.

Slow Hunches

The slow-hunch is the opposite of the ‘Eureka moment’.  Johnson argues that most good ideas are actually ‘slow hunches’ that eventually come together in a coherent argument, rather than sudden flashes of insight.  Slow hunches need to be nurtured since they are easily crushed by the business of normal life, but also by lack of support and insufficient information to back them up.

Libraries can help with slow hunches by doing what we have traditionally done best - providing a wide range of information sources on a wide range of subjects, sensibly organised and easily available.  This allows people with hunches to develop them further.


Serendipity is a popular concept in libraries at the moment, but mostly in the area of recreational reading.  Mostly the work is being done in the form of ‘face out’ displays to encourage readers to choose books they wouldn’t have looked for, but surely we can do better than this.  In the developing digital environment, the challenge will be to avoid simply trying to replicate the ‘serendipity’ possible when library patrons are ‘in the stacks’ looking for something.  Physically browsing for a title and finding something else unexpected does have its romance, but too many authors have waxed lyrical about why this means libraries should never throw out books or go digital.  It means nothing of the sort.  Donald Barclay wrote provocatively about this last year in an American Libraries article - “The myth of browsing”.  Barclay points out that the idea of browsing in academic libraries is fairly new in historical terms, and not nearly as ‘serendipity friendly’ as its supporters claim.  What all this means is that we need to think about new ‘born digital’ ways to provide a superior serendipity service.  Neither ‘searchability’ nor ‘browsability’, but something else.  The recent focus on Discovery Layers is a step towards this, but I suspect the current batch will be seen as a bridging technology while we work out how to really solve the problem.


Johnson’s chapter on error is particularly challenging for libraries.  One of the things we prize above all else is accuracy.  Whether it’s the obsession with ‘reputable sources’ or the anal-retentive stubbornness of the average cataloguer, the whole profession of librarianship is based on accurate organisation.  When it comes to encouraging new ideas, however, it turns out that a few mistakes along the way are often the key to ultimate success - taking the example of evolution in the natural world (as Johnson continually does), without error in DNA replication we would all still be single-celled ocean creatures.

The idea that we should deliberately produce more errors in the way we organise libraries is not necessarily one I endorse.  In my experience errors in the library catalogue are more likely to obscure information and ideas than generate them.  Perhaps all we need to consider here is whether our collection development policies are flexible enough.  Where is the line between different interpretations of the world and material that is simply wrong?

There is also a link here to serendipity.  If your cataloguing staff have made a mistake and given a book the wrong call number, it may well be discovered by a bunch of people who otherwise would never have seen it.  Maybe we should simply worry less.

Exaptation (and recycling)

Exaptation is the act of using something designed to be used in a completely different way.    Johnson gives the example of Gutenburg using wine-press technology to create the printing press.  He goes on to use some other literary examples, such as the detective genre emerging from a few minor characters in early novels.

Exaptation is the pointy end of ‘good ideas’ and is fostered by the previously mentioned factors - people serendipitously coming across information through their diverse, liquid networks are more likely to develop slow hunches resulting in the exaptation of an existing technology or idea into a new one - possibly as the result of a mistake when trying to build something else!

For libraries, we can foster this by working on the previously mentioned aspects, as well as ensuing that our libraries are full of information (or access points to information) on a wide range of existing technologies and ideas.  In other words, the best thing we can do to encourage exaptation is probably to keep doing what we do best. Developing depth of collections is important here too - it’s not enough to know the basic overview of a concept, if you’re going to develop it into something else you probably need to know the detail.

The concept of recycling is also important.  In the natural world, the simplest form of exaptation is found in the idea of ‘waste as food’.  Whether it’s a coral reef or the Serengeti Plain, this is the foundation of all eco-systems.  McDonough and Braungart write about remaking human manufacturing processes to follow this principle in their excellent book Cradle to cradle.

For libraries, the important thing here is not really the recycling of physical matter (although I thoroughly endorse this) but rather ensuring the recycling, recirculation and re-use of ideas.  Science, the humanities and literature are no less prone to trends and fads than any other area of human endeavour.  Libraries must ensure that we don’t limit our range of materials and information to just what is popular right now.  This is the flip side of the desire to be ‘responsive to our community’.  The community might be demanding lots of novels about teenage vampires, but that doesn’t mean you throw out all your copies of Catcher in the Rye.


Platforms allow a common framework upon which new ideas (or species) can be tested and developed.  Soil is a platform.  Forests are a platform.  So too are cities, nation-states, universities, languages, laws and so on.  Johnson writes about ‘platform stacks’, whereby a series of platforms build upon each other.  (He claims this is a common term in sofware development, alas the hive mind of Wikipedia refers only to a solution stack.)  Recently I saw a presentation on libraries in Timor Leste.  Part of the presentation talked about how Australian librarians were training Timorese library staff to use the Browne System for keeping track of library loans.  I’d never heard of it, probably because Australian library circulation systems have been mostly computerised for my entire life.  Why would they use the Browne System when they could use one of the many computerised library management systems that exist?  Let’s look at some of the ‘platform stack’ required:

Library Management System XML SQL TCP/IP C Operating system Computer hardware Electricity distribution network Electricity generation plant

Unfortunately for Timor Leste’s libraries, everything we take for granted in Australian public libraries in terms of our circulation systems ultimately sits atop the platform ‘electricity generation plant’.  Timor has limited and unreliable electricity generation.  Thus, they rely on a different platform stack:

Browne System literate staff writing mathematics and spoken language

Right at the beginning of Where good ideas come from we explore the concept of ‘the adjacent possible’.  In a platform stack what is possible is determined by what is adjacent to where you are at a given time or place.  ‘Web 2.0 discovery layers’ are an adjacent possibility (or a reality) in Australian libraries.  In Timor, this possibility is not adjacent - they don’t have reliable electricity and the number of internet connections is miniscule.

For libraries in Australia and other nations lucky enough to have reliable electricity networks and highly developed technologies, thinking about platforms in the context of idea generation is more complex.  How many platforms will you stack underneath or on top of your library platform?  This might mean you have a collection in a range of languages, but might also include greater use of online communications tools like Twitter and Formspring.  Can you use platforms such as to organise and market your events?  How does the range of ‘platforms’ available to your potential patrons affect who has access?  Are you on a good public transport route, or do people need to own a car to get to the library?  Do you have multilingual staff?  Do you provide programs in a wide range of formats, or are they all ‘talking heads’ events?  Are there services you’d like to offer but you need to go one step back and collaborate with others to build the platform first?

A final question, which I hope to explore in a future post, is how libraries themselves can operate as an ‘open platform’ encouraging new ideas, new thoughts, experimentation and innovation.  Whilst the traditional role of libraries includes aspect of this, with modern technologies a wide range of new possibilities has emerged.

If we, as librarians, consider all these ideas we can transform our libraries into idea factories - the modern-day equivalent of the sixteenth-century coffeehouse.  Instead of being defensive about the role of libraries, we can develop an exciting story to tell funders and decision-makers - the library as an engine for innovation.  I’m excited about the possibilities.

Discussed by

A Record Store Where Nothing is For Sale

16 New Library Tech Stories You May Have Missed

Over the Transom: Volume 4

Libraries, Coffeehouses, and Third Places