The Paradoxical Art of Designing for Emergence

Abstract

This chapter explores the role of digital technologies in collaborative open space learning programs developed at Northern Beaches Christian School in Sydney, Australia. The chapter begins by recounting the role of an in-house innovation incubator called Sydney Centre for Innovation in Learning in nurturing a school philosophy that values emergent interactivity, and offers justification for this philosophy using Self-Determination Theory. It describes a number of ensuing open space learning designs, making specific reference to the practical use of digital technologies. The chapter concludes by proposing ecological design language that interprets structures in physical, virtual, and cultural space by their ability to facilitate emergent, unscripted interactions between people, their environment, and information. It emphasizes the paradoxical importance of linearity, constraint, and expert teaching in learning designs that set the scene for emergence to occur.

A Design Framework for Emergent Learning

All the students were on-task, all the time. If I didn’t know better I’d have thought that the students were actors—and I was in some kind of Truman Show experience. —Andrew Douch, after a half-day visit to our school (Douch, 2012)

In this chapter I explore the place of digital technologies within an ecosystem of structures which can promote or constrain interactions between people, their environment , and information. I base this exploration on a range of designs that teaching staff evolved during an intense 10-year journey of innovation at a K to 12 school in Sydney, Australia .

A Lived Experience of Innovation and Transformation

The key source material for the reflections in this chapter is the lived experience of Northern Beaches Christian School (NBCS), through my own interpretation in the period from 2002 to 2015 as first a classroom teacher , then school leader, and eventually a consultant working on behalf of our school .

I would describe NBCS in 2002 as “industrial” in structure, consisting of single cell classrooms populated with rows of desks facing the teacher’s space at the front. The default pedagogical practices relied heavily on teacher-delivered content according to a teacher-crafted schedule.

Our principal Stephen Harris had joined the school in 1999 and had set about disrupting these structures with some success. In 2005, he greatly accelerated our change journey by announcing the launch of “Sydney Centre for Innovation in Learning” (SCIL) as a research and innovation unit embedded in the school. SCIL became a rather informal banner under which we could champion and celebrate innovative culture, processes, people and projects. For instance, a teacher could call themselves a “SCIL Associate” or launch a “SCIL” project. Certain leaders were associated with SCIL and had responsive wildcard functions, fanning into flame moments of inspiration among staff whenever these occurred with collegial and logistical support . Its mottos were “do, then think,” “ready, fire, aim,” and “question everything.” The SCIL leadership roles were also deliberately subversive, allowing ideas to be implemented without having to navigate a formalised bureaucracy.

David Price (OBE) of the UK Innovation Unit has characterised SCIL as “what would have happened had Thomas Edison been an educator , rather than an inventor.” (Price, 2013)

Between 2005 and 2010, under the banner of SCIL, a range of new learning structures were prototyped and tested by a group of enthusiastic teachers that might be called the “coalition of the willing.”

These prototypes were often inspired by ideas we had identified in a series of international research tours, documented in Harris (2005), Harris (2006), and Linfoot (2006). The establishment and early operations of SCIL within Northern Beaches Christian School, and translation of research insights around virtual space into day to day practice , are documented in detail in Harris (2008), Linfoot (2007), and Harris (2010a, 2010b).

After 2010 we experienced a cascade of further innovations around the school as the “coalition” tipped and became a majority.

Our rationale for change during this time was largely philosophical, driven principally by a vision for high student engagement . We grew a shared school narrative that interpreted our core challenge as being how to engage and activate all our students to the highest degree possible. The data we used to measure engagement was sensory and informal, and processed through shared conversations. We didn’t feel we needed a number or a metric to tell us what we could see with our own eyes. In practical terms, were the students moving, talking, contributing and taking initiative? If so, we had a promising design. If not, we needed to change it quick.

I have recently found in Self-Determination Theory (SDT) better language to describe our philosophy and evaluative process (Deci & Ryan, 2008). SDT identifies three critical ingredients that promote growth and wellbeing in individuals and communities: autonomy , relatedness, and competence. I will use these terms in this chapter as a more robust proxy for what we as a school called “engagement” when evaluating our progress. The proxy is a fair one; the SDT terms simply bring nuance:

SDT suggests that it is part of the adaptive design of the human organism to engage interesting activities, to exercise capacities, to pursue connectedness in social groups, and to integrate intrapsychic and interpersonal experiences into a relative unity. (Deci & Ryan, 2000, p. 229)

If our measurement of success consisted solely of student competence then we might have focused entirely on data about learning achievement to guide our learning designs. While we certainly did consult such data, and while it steadily strengthened during this time, we were more concerned with getting the energy up in the room: having students moving, talking, taking initiative; concerns that fit neatly into the SDT categories of relatedness and autonomy . It did not surprise us that in pursuing deeper student engagement , we saw a steady improvement in data from external testing showing improved learning outcomes. However, our pursuit of engagement was not a means to end, but was valued in its own right, for both students and for teachers. This was matter of school culture , identity, and philosophy.

I have used the phrase “emergent learning” in the chapter title to capture some of this philosophy. The sense of the word “emergent” is that unexpected and unscripted interactions take place between people, space, and information. Emergence is implied by the SDT categories of relatedness and autonomy ; both require genuine freedom and therefore transcend our ability to script and control.

In the same way as we pursued designs that created space for emergent interactivity , the evolution of the designs themselves was emergent. Under the banner of SCIL we embraced trial and error, and grew a high-trust environment. We did not have a scripted master plan for school improvement; instead we felt our way forward through a series of iterated experimental designs.

Below I describe some of these designs. Then, I explore new language that we developed to identify a repertoire of design structures that emerged over the years as aligning with our philosophy. In particular, I explore how digital technologies found their place within a holistic design framework whose core affordances and constraints (Greeno, 1994) are mapped back to the interactivity suggested by both SDT and our stated ambition to have students “engage.”

An Early Prototype using Bloom’s/Gardner’s Matrix

One of the early prototypes at our school saw a team of five High School teachers come together for a few hours a week for a program integrating History, Geography, and English , which we called “The Matrix.” The three subjects still continued in traditional form—“The Matrix” was an additional subject using time peeled away from those main linear programs. It took place in a large shared space created by opening the concertina doors separating five traditional classrooms.

During Matrix time, students opted into one of a large number of learning challenges presented on a giant poster with vertical and horizontal grid lines. The horizontal axis had rows categorised by Gardner’s multiple intelligences, and the vertical axis had columns categorised by Bloom’s taxonomy. This structure is well known by teachers , and was apparently first conceived, popularised, and christened “the Matrix” by Ralph Pirozzo (Rao, 2009). In our implementation, each table cell contained a brief description of a challenge keyed through to learning outcomes. Students would choose an activity, and then find a “stimulus card”—a corresponding laminated A4 card with more details and resources .

Some tasks were open ended, and some quite prescriptive. Each task was worth a number of points, with each student setting targets for how many points to earn over a 10 week term. Each term we moved to a fresh Matrix with a new theme and new target for points.

Despite some fairly obvious design flaws (a bewilderingly high number of choices, no expert input, no linearity at all, no formalised collaborative structure) the Matrix was a stunningly popular with students and the teaching team.

We filmed various students from Year 7 and from a similar program in Year 5 and 6 as they explained why the liked the program. You can watch the Video 1.

A dominant theme in the student feedback is autonomy : being freed from the tyranny of teacher talk, and being free to choose what to do and when. Yet this freedom was experienced within carefully curated structures. It was certainly not a vacuum.

A critical element to this early design was the notion of some kind of map that students could consult of their own accord that set out a series of curated learning challenges . In the Matrix program this took the form of a poster on the wall and a series of laminated A4 stimulus cards. In later designs we used a web-based Learning Management System (LMS) as a space where we could communicate different options to students. However the LMS was following the same principle is the original Matrix poster on the wall. If we are going to give students choices, but don’t want them to have to choose in the context of a vacuum, how will they know what choices are available? Where will the scaffolding go? We needed a space from which to broadcast this information (Figs. 15.1 and 15.2).

Fig. 15.1

First two rows from a Matrix poster

Fig. 15.2

A Matrix stimulus card

Both our Matrix poster and later web-based matrices had the same design function: to broadcast information to students at the point of need. Both increased interactivity in the program design by releasing the teacher from having to be the broadcaster of information, and the students from having to be quiet and listen. The information was there when it was needed.

Later Designs

Our original Matrix program ran from 2007 to 2012. During this period a wide range of other designs emerged around our school.

This time-lapse video shows a number of these designs in succession. The video pauses at key moments and I shall refer to each of these in sequence below (Video 2).

In each design multiple spaces and classes have been brought together and teachers are working in a team.

The first scene shows a Year 7 Visual Arts lesson . At the first pause two teachers communicate briefly before class begins. There is no formal start to the class—students experience high autonomy from the moment they arrive.

In this design, students choose from six different projects, completing one each term. Instead of scaffolding these projects with a poster or task cards, a richer set of multimedia scaffolding was brought together by the teaching staff and published on our Learning Management System (LMS) —Moodle . Moodle allows teachers to design their own web pages, populated with resources and interactive tools —e-mail forums, quizzes, wikis, and the like.

The screen-shots in Figs. 15.3 and 15.4 show the Moodle pages for the Visual Arts program. It includes “flipped learning” videos created by the program teachers.

Fig. 15.3

Students click to choose a project

Fig. 15.4

Instructions and videos on our LMS

All students blogged their progress and tracked their projects and blog sites on a central collaborative Google Doc. These are shown in Figs. 15.5 and 15.6.

Fig. 15.5

A student blog post

Fig. 15.6

Google doc tracking dashboard (deliberately blurred)

The notion of online learning can conjure up a vision of students buried in laptops, cut off from the world. However, in this program, where the technology is used in the context of a holistic design built to enhance engagement, there is a great deal of collaboration among students, and lots of 1 on 1 “on the shoulder” guidance by the teacher. The second pause in the video shows such a moment.

A close examination of the video reveals a group session lead by the other teacher in the background. This was run for students who are pursuing one of the projects, and was a compulsory session to allow the teacher to provide necessary expert input.

This teacher-lead session has an informational function that is comparable to both the original printed Matrix poster and our use of our LMS to scaffold choices. In each instance, information is being broadcast. So what is the different between a teacher and a poster on the wall? In my observation, if teacher instruction is delivered to large groups of students with minimal interactivity, there is very little difference indeed. In the time-lapse video, we see a far more interactive form of expert input. The students gathered around in a small group, sharing a space, in a special moment in time.

The second scene of the time-lapse video, shows a program of greater scale and complexity. The program takes place in what we call the “Zone.” The Zone consists of 180 students from Year 5 and Year 6, and their six teachers, who work all day, every day, in a very large shared space. The video shows an integrated project with a Science focus, which runs for about a term. The video only shows half of the space, with the other half accessible down a large central staircase in the middle of the building.

In the video background you can see teacher Mr Daniel Wearne providing a teacher-lead session related to one of a number of Science activities that were available to students. Mr Wearne repeated the session several times over a fortnight, with different students opting in at different times. Other teacher-lead sessions were available in other locations, including outside (Figs. 15.7 and 15.8).

Fig. 15.7

Clickable graphic on our LMS in the Zone program

Fig. 15.8

Part of a digital stimulus card in the Zone

In the foreground, a student works by himself, but is visited by two of his peers for a short interaction where they appear to discuss his progress.

It is worth noticing the behaviors of the other teachers in the video, moving from student to student or group to group. There is a special moment where teacher Ms Katie Morrison emerges from the stairs and scans the space to see where she is needed.

In the third scene of the video there are two Year 8 Music classes in a shared space. The program was developed and is led by Mr Brad Fuller and Mr Peter Orenstein. The learning space consists of a series of stations in a sequence that mimics a typical compositional and performance process. In groups of up to seven, students compose music at one station, jam together with instruments at the next, transcribe the music at the next station, perform it to the class on the stage, and then record it in one of two side music studios. Students work in the same group all year, rotating from station to station every 15 min according to a dashboard set up on a shared Google doc (Figs. 15.9 and 15.10).

Fig. 15.9

Clickable music stations

Fig. 15.10

Music online instructions and video

Two of the stations involve direct teacher input. One of these stations is visible on the right of the screen.

The final scene of the video, before returning to the Zone program from a different angle, shows two Year 8 French classes in a similar paired combination. The particular unit on the video is called “Bootcamp” and consists of six cycles around a variety of communicative learning outcomes. Each cycle begins with a teacher-lead session, followed by video training and then a mini-Matrix with a range of language games and activities that address the learning outcome (Figs. 15.11 and 15.12).

Fig. 15.11

French clickable graphic

Fig. 15.12

French flipped learning video

Most of our other learning designs, from Kindergarten to Year 12, involve similar structures to varying degrees and in varying combination. For instance, a small senior elective class might take place in a single-cell classroom with just one teacher, but still offer choice and flipped learning resources . Our kindergarten program does not require students to consult a web portal, but is co-taught in a flexible learning environment with a program that values student choice.

I have noticed several misconceptions that are very common among fellow educators when they first observe our programs: First, that greater student autonomy correlates with less structure and constraint in our program designs, whereas we believe our programs simply have more sophisticated structure and constraint . Second, that teacher-delivered instruction has to be de-emphasised to make room for student-driven learning, whereas we believe that because teacher expertise is so powerful we need to be savvy about offering it to the right learners at the right time. Third, that, given choice, some students will choose not to engage, whereas our experience has been that they all engage to a deeper level than in a traditional set up. Finally, that students will get lost in the complexity of the programs, whereas we have very deliberate tracking mechanisms to make progress transparent.

There is something of a paradigm shift implicated here. Many schools are seeking alternatives to default modes of teaching. Our challenge is perhaps that the defaults are so well known, whereas the alternatives are less familiar.

As we have chartered unfamiliar design territory we have had to find new language to normalise a repertoire of new structures. I will unpack this language below.

The Learning Landscape

Our previous more traditional pedagogical designs tended to sequence teaching and learning activities in a linear process, leaning heavily on certain concepts of cause and effect: the teacher does THIS, then the students do THAT, and the result is learning. Of course, great teachers improvise and are highly responsive. Our challenge was to allow such responsiveness to permeate the entire learning design.

We have used the phrase “learning landscape” to invoke a vision of complex interactive and responsive cause–effect loops between people, space, and information: in short, an ecosystem. In a learning landscape, teacher–student interactions may still be centrally important, but are not exhaustive and all-dominating.

Yeoman explores the complexity of such emergent interactions in her thesis “Habits and Habitats: an Ethnology of Learning Entanglement” (Yeoman, 2015), where she applies the Activity Centred Analysis and Design (ACAD) framework of Goodyear and Carvalho (2013, 2014a, 2014b), to her observation of the Zone program at my school that we saw previously in the time-lapse video. Her observation was extensive in scope, comprising no fewer than 549 h throughout 2012.

Yeoman’s refers to both “material ecology” (Yeoman, 2015, p. 26) and a related concept : Ingold’s (2010) concept of entanglement: “a meshwork of interwoven lines of growth and movement” (p. 3) and “an ontology that assigns primacy to processes of formation as against their final products” (p. 2).

Our internal idiomatic phrase “learning landscape” evolved separately to these works but is very much in the same spirit.

Linearity and Personalization

The term “landscape” deliberately activates teachers’ visual imaginations and taps into cognitive architecture that is already adept at tracking and making sense of complex interactions .

A landscape has a mix of structures that create freedoms and constraints . It may contain pathways—even a main pathway with a beginning, end, and checkpoints, but can also contain side paths, alternative routes, or even outright wilderness.

There is an interesting tension here between relatedness and autonomy , because a deep sense of relatedness can in fact be created during one-size-fits-all experiences. For instance, in our Zone program we routinely have all 180 students sit together for one experience or another. This “coming together” with its associated rituals, routines and rules, creates bonding, a shared identity, and common culture . These moments are defined by their very lack of personalization, or individual autonomy . No matter how we feel, or where we’re up to, or what our agenda is, in this moment we’re going to meet together and operate as one unit.

The concept of a learning landscape allows for such solidarity while providing ample space for personalization, in the form of side paths or a Matrix. These can nurture relatedness, too, of a different sort, as students help each other or cross-pollinate from different pathways.

I mentioned wilderness above. There are parts of landscape that have not been mapped, curated or resourced . They are unknown, even to the teacher , at the outset of the journey and may unfold from moment to moment. Consider the significance of the term “landscape” versus “map.” A map is artifice. In contrast the landscape is intractably complex and infinitely fractal. The term “learning landscape” intentionally embraces both of what Yeoman calls “design-in-advance” and “design-in-the-doing” (Yeoman, 2015, pp. 12–13).

Synchronizing Physical, Virtual, and Cultural Space Design

Where do the “digital technologies ” referred to by the title of this book fit in the design repertoire that I am exploring? I’ve already pointed out the similarity between a web portal and our Matrix poster. If cast in terms of information flow these both share a similar function. I wish to pursue this insight through to a main thesis that sees design entirely through the lens of relational affordances between people, space, and information.

Let me start with a strange contradiction: in the early 2000s our school actively pursued a digital technology infrastructure in the form of hardwired computer rooms. These were very popular and in high demand by teachers, especially after we introduced our LMS in 2005. In response to demand, our IT team created more computer rooms. At our peak in 2008 we had 500 desktop computers installed for a school population of about a thousand students.

Typically, the desktops were installed around the perimeter of classrooms, or in a horse-shoe configuration, or both.

We experienced in these configurations a strange contradiction: the technology offered a wealth of new information and interactive possibilities to students, while simultaneously making face to face conversation more difficult by blocking line of sight and restraining physical movement.

The contradiction lead us to conversations about matching the physical characteristics of a space with the technological characteristics of a space. How did they fit together? Or fail to fit together?

Our term for “digital technology ” was Information Communication Technology (ICT) . Our dilemma was that our ICT design was allowing online information flow, but hampering the flow of information between the people inhabiting the physical space where they were installed.

At about the same time we were experimenting with writing surfaces created by applying whiteboard paint to entire walls or building tables out of old whiteboards, shown in Figs. 15.13 and 15.14.

Fig. 15.13

Students work collaboratively on a whiteboard wall

Fig. 15.14

Senior students collaborating on a whiteboard table

Unlike the desktop computers, our whiteboard surfaces were allowing students to contribute their thoughts freely to anyone within line of sight. These, however, did not have the transcendent capacity of a computer to save, duplicate, and transmit information across time and space.

It made sense for use to the term “virtual space” to refer to any mechanism for the transmission of information. The desktop computers and whiteboard paint were serving the same function: shaping the flow of information. To be strictly consistent in this functional definition—that virtual space is information flow—handwriting, and even vocal cords have to be included.

This definition holds great explanatory power. Consider the popularity of flipped learning videos. A lecture might be ostentatiously interactive, but a flipped learning video can be paused and rewound and fast forwarded. The video essentially separates the information from the living speaker, so that it can have a life of its own. This is how virtual space works: information is unhooked from live communication and becomes a strata in its own right. One need only think of a “keep off the grass” sign or other printed admonition. Whoever created this little bubble of virtual space is now absent, but the sign stands, broadcasting an agenda.

The disruption created by a free economy of information is far from recent. For instance, Socrates expressed annoyance at how freely available knowledge could be obtained by reading. He questioned its value. He complains that readers “will be hearers of many things and will have learned nothing; they will appear to be omniscient and will generally know nothing; they will be tiresome company, having the show of wisdom without the reality.” (Phaedras, 275b) In his view, knowledge should not come so easy.

The invention of the printing press took the dilemma up a notch: in Ann Blair (2003) quotes Conrad Gesner complaining in 1545 about the “confusing and harmful abundance of books.” Both Socrates and Gessner are coming to terms with an intensifying information ecosystem. The internet is simply continuing the trend.

The practical upshot of this definition is the question: what sorts of information flow do we want to make possible in our learning landscape?

The question can then be answered with a single, unified design that synchronises physical affordances and constraints with virtual affordance and constraints .

Considered from this perspective, our old traditional classrooms were actually very well synchronised: the physical rows of desks faced a front where information would be transmitted from the teacher and the whiteboard. Both layers—the physical and virtual—worked together, albeit in a very constrained way.

Continuing this perspective, in our quest for increased autonomy and relatedness, the desktop computer rooms appear as an awkward and desynchronised middle step with unresolved conflict: the physical design hampered interactivity while the virtual promoted it.

In 2010, we resynchronised our physical and virtual spaces by moving to a “Bring Your Own Device” program. We maintained our enthusiasm for using walls, tables, and windows as writeable surfaces. This allowed information to flow freely in and out of the classroom to the world outside, and within the classroom, and permitted similar freedom of movement through physical space.

We also speak of a third layer of space: cultural space, referring to such elements as routines, permissions, behavioral scripts and norms, and shared values. Just like physical and virtual space, a classroom’s cultural space has affordances and constraints and can be shaped by design. Cultural space can synchronise with its siblings or be in conflict with them. During our journey we have worked with intent to shape a set of permissions, routines, and the like to allow students and teachers to feel navigate the higher degrees of autonomy and relatedness that we have established in our physical and virtual spaces.

Caves, Campfires, and Watering Holes

As we sought to find learning designs that synchronised physical and virtual spaces, we were delighted to come across a specific taxonomy of spaces proposed by Professor David Thornburg (2007).

Thornburg proposes the terms “cave,” “campfire,” and “watering hole” as metaphors that refer to spaces that encourage certain interactions between people and information.

A cave allows someone to relate to their own inner world, a campfire allows a guru to share insights to a group, and a watering hole allows free interaction . I use shorthand to understand these three relationships: 1:self, 1:many, and many:many, but they can also be termed reflection , expertise, and collaboration .

The terms apply elegantly to both physical and virtual spaces. A secluded corner is a physical cave; a blog is a virtual cave. A set of chairs can make a campfire, and so can a flipped learning video. A physical watering hole can be created with grouped tables or agile furniture, or equally in virtual space with a shared Google Doc, an Edmodo social networking site, or indeed any Web 2.0 tool (Fig. 15.15).

Fig. 15.15

Students from Australia and France communicate using a virtual watering hole: Edmodo

In physical space, the spaces tend to exist in discrete units, but in virtual space some very interesting mash-ups can occur.

For instance, I have referred to a blog as a cave space—it can certainly feel that way to the blogger . However, it can also function as a campfire, sharing thoughts from one to many, and potentially getting the blogger in trouble for their unguarded candor. Twitter can function as a cave, campfire, and watering hole all at once.

Our writeable surfaces—virtual space—share this dual functionality. Students can write reflectively on a wall or table (in a cave space). Then, for as long as the writing remains, their thoughts are accessible by anyone within local line of sight (a campfire) (Fig. 15.16).

Fig. 15.16

Definition and examples of caves, campfires, and watering holes

In the time-lapse video you can see these different spaces in action. For instance in the video of the Zone program, the child working by himself is in a cave space, while the teacher at the television screen behind is offering a campfire experience, and most other students are interacting in watering hole mode.

The use of our web portal to present and resource various learning activities on a menu of choices is, in these terms, a virtual campfire, as was our original poster showing the Bloom’s/Gardner’s Matrix.

In a co-teaching environment that includes a laptop program and online videos, students can engage with many varied sources of expertise. One teacher can offer a campfire session in one physical location, while an alternative is offered elsewhere by their colleague, and online explanations are available on tap as the need arises.

Similar opportunities could be replicated even in an environment with few computers and no internet. Flipped learning videos could be made available on the hard drives of the available laptops, or via a DVD on a television screen. Students could walk to one of these stations, by themselves or in groups, and select the video explanation they need.

These sorts of designs don’t necessarily need an open physical space either. If co-teaching is desired but the only classrooms available are single cell, a pair of teachers could still work in separate spaces but allow movement by the students, perhaps establishing a campfire in each space, or a campfire in one and a silent cave space in the other, or a watering hole, or mix. Low-tech virtual watering holes can be created with butcher’s paper taped to tables or walls, or by using chalk on concrete floors.

None of these design elements—the learning landscape, linear paths, personalised paths, and caves, campfires, and watering holes make for a necessarily good design. A good design is a design the works according to success criteria. As I lay out at the beginning of the chapter, our criteria were simple and philosophical: we simply wanted all students to be engaged and activated in their learning. We recognised good design with our eyes and ears, seeking out structures that prompted the students to move and talk and contribute. As we embraced these structures we found that ecological language came naturally and intuitively, and helped us make sense of what really mattered in our designs.

Emergent Learning

Emergent learning has been central to our journey for teachers and students alike. For teachers, because we could only develop new designs by trial and error, each prototype opening up new avenues and closing others. For students, since letting go of teacher control meant anything could happen.

When everyone in the room is both individually activated and collaboratively connected , a world of possibility emerges. Learning can happen in powerful and unscripted ways, coming at any moment from the confluence of interactions occurring between people, space, and information.

Counterintuitively, such dynamic learning environments are not devoid of structure, but nurtured through deliberate shaping and resourcing. In particular, digital technologies can find their home within a broader ecosystem that liberates its inhabitants from the tyranny of constant one-to-many delivery or linear cause–effect chains. In our best programs there is just the right combination of structures—physical, virtual, and cultural, to set the scene for the unexpected.