“For the partisan, when he is engaged in a dispute, cares nothing about the rights of the question, but is anxious only to convince his hearers of his own assertions.”

- Plato, Phaedo

Too often in problem solving, we become fixated on the answer, and in the single-minded pursuit of that answer – that one, singular solution – we seek out those who we think might possess it. They are the experts, the specialists, the consultants; we know that when we have a problem, a question, then what matters most is to find a person with the answer.

But to have an answer is to ignore the question – to come in with an “expert point of view” is to assume a kind of intellectual partisanship in the problem-solving process. Far more powerful is to give due time and attention to the process of questioning.

I recently heard “killer questions” defined as questions that can’t be answered directly – these questions, and the exploration necessary to follow them through, can help unpack assumptions and gain new perspectives – but more importantly, can allow answers to emerge.

This is important in the typical consulting process. Clients typically ask a question or present a problem. The consultants go off and undertake their process of discovery, then return to present their answers. Sure, there may be a menu of answers to choose from, but they are answers nonetheless.

The great loss to the organization is that it did not go through the process of questioning, which in the end, is far more valuable than the answer it produces.

Why? Because questioning creates understanding. Because deep questioning is a creative process. Because going through a process of questioning together helps us to believe in our answers…and because going through the process keeps our answers from being stagnant and fixed – answers become as fluid and emergent as the questions that lead us to them.

So while it is helpful to have experts involved in the process of problem-solving, they are better involved as an input to the process, rather than as an output; their particular ideas and points of view should be swept into a process, instead of speaking for it.

I set out before that in stringing knowledge games together, the designer needs to consider three dimensions progress; the tactical, emotional and conceptual. In thinking through the idea again, I tried to relate my own design process to each of these dimensions, and came to a few realizations. The most common design models I use are the Scan-Focus-Act model and the Creative Process Model, both from the MGTaylor method of designing collaboration. I’m not sure about other MGTaylor practitioners, but here’s what I realized about my design process when I thought about it in terms of these dimensions:

1. I tend to design for information, first. At its most basic, the Scan-Focus-Act model is a model of information distribution. Scan: Bring in disparate data sources, ensure a common understanding of information across the target group of individuals. Focus: Allow for the reconfiguration of information and prototyping by the group. Act: The best ideas are developed more fully from the prototypes to bring together a working model. In designing this process, I often focus on the development of this information through the lifecycle of the interactions.
2. While I consider the other dimensions as design considerations, I don’t specifically design them. I think a lot about what the individuals going through the process will experience, and use that as a consideration in the design of the information flow, but I don’t directly design the experience with the same calculated rigor that I apply to the tactical dimension. The emotional dimension is something I’ve thought of as an element you need to make space for, but which is a by-product of other factors and processes. The conceptual dimension I have treated mostly as emergent.

This leads me to the following questions. First, can you design emotional experience? Clearly a great number of artists, writers and filmmakers would argue that you can. But where does that fit within the realm of collaboration design? What models inform the designing of emotional experience?

Secondly, as the conceptual dimension is critical to the ideas that come from collaborative interaction, how do you design for emergence?

To answer some of these questions, I’m going to be writing a couple of pieces both here and on the Knowledge Games blog. First up will be Designing for Emotion, followed by Designing for Emergence. I’ll give shorter treatment to Tactical Game Design, since there’s lots more out there on that, and then I’ll reexamine an approach to bringing the three together. In the interim, ideas and input on any of these questions would be appreciated!

There’s an extra image on Dave’s flickr stream here which equates the journey towards a fuzzy goal as a journey. Going picture for picture, here’s a few thoughts.

First, when I think of how fuzzy a goal can be, I come back to the idea of tiered goals, or layers.

Goal fuzziness as it relates to layers.

When I think of how fuzzy a goal can be, the first question is this: fuzzy to whom? Assuming we are talking about the people who are in the process, as opposed to those who are managing the process, I think the model here applies.

Collaborators need to be very clear on the goals of the task at hand; the need to know what they are doing now.

What can be a little more fuzzy is what comes next. In fact, too much focus or too much supposed clarity on what’s next can be a distraction from the task at hand, and can lead to rigidity. If you always know what’s next, it’s harder to change based on evidence on the ground.

Fuzzier still are the results, both of the immediate task and the tasks that follow. On the outer edge of fuzziness is “where is this all going?” This is the bigger picture, which in knowledge work is only created through iteration and, as I am now convinced by Dave, progression. Progression toward something broader, which becomes iteratively more clear as each “experience” builds on the last.

To put this into the context of Dave’s drawing of the ship on a journey of discovery, I think we end up with not only the levels of fuzziness, but also the roles in knowledge work.

The core of knowledge work is the experiential process itself and the emergent concepts which result from interactions of people, thoughts and influences. But what differentiates knowledge work from just knowledge or unguided thought is the process.

The roles in knowledge work as we move towards fuzzy goals

Taking the same labels from the last diagram, we have a person (group of people) focusing on what is being done now, living in the moment and tackling something defined.

To allow these people to truly focus on the immediate task, we have another who can, based on progress, guide them towards “what’s next.” The interplay between what’s happening and what’s next is mediated by someone who guides the process based on the results, and it all comes together as we plot our map, or bigger picture, of where it’s all going. I imagine that this person is a map maker as well as a navigator; they chart based on the knowledge we have, but add as they explore new territory.

So when I look at Dave’s picture, that’s what springs to mind.

My scratchy brainstorming notes

Now having had more time to reflect on Dave Gray’s talk on what he’s calling “Knowledge Games”, I’ve put together a few thoughts that were niggling in the back of my head. I think that, for those of us working in the collaboration space, the concept of knowledge games is a great way of encapsulating and explaining the complex “play” that we enable to help people solve difficult problems.

Part of Dave’s concept, as I’ve understood it, is that in knowledge work – as opposed to process driven industrial-type work – structure, teams and motivation need to be modeled differently. Instead of process, then, there is play; essentially, resources need to be “managed for unpredictable results,” as opposed to the traditional, industrial model of managing for consistency and predictability…which is a split which I’ve been exploring here.

I liked the idea of framing these ways of working as games, but for my purposes, I’m going to take it down a level from game, and call it a designed interaction; which is how I think of them when thinking through a collaborative design.

I think that the three or four people who actually read this blog are already bought into the idea of play as work (anyone who designs/delivers DesignShops), so there’s no need to dive into that, but I’m excited by the idea of play as work getting more mainstream attention, because it gives a better framework for the core question of collaborative design; if games are used to work, how do you align multiple games to your goals as a group or organization?

Three main models came to my mind in thinking this through; one from Christopher Alexander’s The Timeless Way of Building, the other from Michael Anton Dila, who drew it on a whiteboard in this video, and the other which I came up with while trying to make sense of it all. Horrible sketches of all three models are in my notes, below.

So first, quite simply, you can’t achieve integrated outcomes through a single game. A cluster of activity creates something, but it often isn’t the something that you can “take to market”, so to speak. Enter Christopher Alexander, and his patterns. These patterns were expressions of interactions between people, places and things which – if properly aligned – would produce a desirable result (a horrible simplification, I know). What sticks out for me in his design approach is the way in which he sketches his product, beginning with the immovable pieces and the largest patterns, then nestles in the rest of the patterns as he is guided by affinity between them. It’s like sketching the outline of something, then slowly layering in the detail.

This, to me, is the starting point of managing through knowledge games, or designing a DesignShop…you have to know what you want, even if you don’t have an exact picture of it yet. As each piece comes together, the feeling of one component helps to guide what the next should be. But those outer boundaries, the overall idea and reason for cohesion, needs to be present (if this is to be imagined as an enterprise, as opposed to the chance games of unconnected people…a business vs. a conference).

That brings me to my model. Whereas an industrial process focuses on inputs and ouptuts in a very tangible sense, the design of a game has three dimensions of input and output, though the boundaries of “in” and “out” are not rigid. The model is still sketchy, but I was thinking of the flow as having three dimensions:

1. Tactical
2. Emotional
3. Conceptual

The tactical thread to a string of games is the more traditional stream of information as it flows from one segment to the next. This is the explicit “purpose” of the interactions – the information to be gathered, the model to be made, the document to be created. Progress in this stream can be thought of with three spectra; accurate to inaccurate, framework to detail, relevant to off target. The principle flow of progress between games can be measure along the “framework to detail” spectrum, whereas the other two spectra are more guides to be used to gauge effectiveness.

The emotional thread, I think, is a critical but often overlooked aspect of knowledge work. It could also be called the “personal” thread. This thread is what makes it a game – it is the journey through the game as experienced subjectively by the participants. This is a very non-linear thread between games, though the progression should end with “engaged” for a sufficient number of players to achieve critical mass behind the outcomes. The elements of the emotional thread are: frustration to satisfaction, connected to independent, bored to engaged.

The last dimension of inputs and outputs in knowledge games is the conceptual thread, which I see as the layer where “understanding” lives. Understanding is, I think obviously, a critical component of knowledge work. The conceptual thread involves the slow development of cohesion and growth of ideas within the group. The three spectra within this dimension are: scattered to associated, analogous to synchronous, and, molecular to holistic.

To look at the model as a whole, what you see in the design process – and especially the delivery process – is that the journey through a series of games will be a journey experienced in these three dimensions, which if properly engineered, will result in a group that has achieved the desired level of detail in their artefacts, feels connected and satisfied as a group, and which has developed an awareness, or holistic view of what they have achieved and need to do next. Each stage, though, will have different requirements. Creating frustration may be necessary to “unleash” the group at a subsequent step, and the move between framework and detail may be recursive, depending on the complexity of the task.

The main idea, though is that a group always experiences a game along these three dimensions (probably more).

This is where Michael Dila’s model comes in. In his video, he sketches out a model of knowledge work inspired by the TCP/IP stack. Very, cool, I thought. He puts visualization at the bottom of the stack, collaboration next up, innovation on top of that, then participation at the very top.

I stewed on that for a while, then started drawing links between that and my model which I’d scribbled on the same page. That’s when I thought that visualization and collaboration both address the emotional and conceptual level of my model. Visualization helps us to connect with ideas, see the big picture, and feel some affinity for the concept. Collaboration results from a sense of connectedness between people with a common sense of concept and purpose.

There is interplay, there. Innovation, I think, occurs when there is visualization and collaboration – in this sense, innovation is an interaction between people and ideas, which I believe is the heart of collaboration. Visualization enables shared understanding, shared understanding allows groups to work together, and it is this combination of human intelligences that allows innovation to spring forth. The whole stack, then, can only really reach its potential if we design for people, and accept their multi-dimensional nature.

Which brings me to the original question of how to thread games together. What would it be called? A tournament? Who knows. The idea, though, is that by setting an overall goal for the sequence, a series of games can then be assembled towards that end. Defining the tactical stream might be the easiest dimension to design for, but the knowledge that true emergence can only take place if the other two dimensions are considered is what makes this a game. How many process flows consider fun? Interactivity? Mood? Experience?

This is where I like Michael Dila’s model. Even if you were to design one-dimensionally, along the tactical thread, using his knowledge work stack would force you to, at the very least, draw in consideration of the conceptual dimension of the experience.

I’ll play with this more in another post…

Picking a tool for strategy development is, to me, an exercise in doctrine and dogma, and can easily miss out on something which I very firmly believe; the most elegant and beautifully conceived strategies can fail completely if they do not connect – in a deep and meaningful way – with the people who make up an organization.

So what I’m saying is that, in the end, I don’t really care which strategy tool is used, because I think that strategy – and especially a really clever strategy – can be held up as a false god, giving comfort to leaders and a vague sense of meaning to their teams.

And if strategy is a false god, then consultants are their prophets; we are incredibly good at creating and selling strategies which then are passed off to organizations with varying degrees of success. We ask that organizations and teams take it on faith that the development of a strategy is critical to their survival in a dog-eat-dog world; that by tricks and clever machinations we can devise “the plan” to guide them through troubled waters.

But for the sake of argument, what if I were to play the part of the strategic atheist – to say that I didn’t believe in strategy? What if I asserted that using this tool or that tool to create something which – by my eyes – does not exist would not help me?

What would I be left with? How could I steer my organization?

What I think I would be left with is my people. To mix metaphors a little, I would suddenly see that the boat is moving not because of the destination, but because of the people rowing it. I would see that it is not that we have a strategy that is important, it is actually more important that we believe in and are motivated by something common. The atheist view of strategy, then, would assert that it is the tactics of cohesion which drive success. Central to that is the “fit”; whether or not the aspiration and the group are compatible.

In that sense, then, I would say that the strategy itself could be considered secondary to the process of creation. Do people feel a sense of connection with the strategy? Did they help to create it? Is it an expression of their shared understanding and world view? Do they feel capable of being a part of it? What is more important; beliefs or belief itself?

It is not the god of strategy that gives me value, but my ritual of worship and observance. As my zen teacher once told me, “I’m not telling you to worship the Buddha. I’m just teaching you how to sit.”

Which is all to say, I don’t really know which method I prefer…I just care about the process.

Key takeaways from the talk, for me, was a reaffirmation of the importance of environment in support of knowledge-working teams. I don’t know how many times Christopher Alexander’s “A Pattern Language” popped into my head during this part of Dave’s talk…especially the parts regarding interconnected workshops.

Second, I’m anxious to get my hands on a copy of the book to see just what exactly is laid out regarding the use of “knowledge games” to spur groups into co-creation of products.

Using “play” as a way of solving complex problems is, of course, central to a lot of creative work processes, but I know from my experience as a process designer that the sequence is critical in bringing the best out of a group. Which is all to say that I agreed rather violently with everything Dave laid out tonight, but with the open question being this: if serious work in the knowledge economy looks like play, and is carried out in these “knowledge games”, how does he envision the role of the game designer?

The original sketch

I clipped out a few slides from a recent presentation on enterprise business architecture. The original concept was from a sketch I came up with when trying to think through the “process of change” in a government setting…I then got stuck with the sketch, as it seemed to capture what we were trying to say.

This was an interesting lesson for me in the visualization process…to me, visuals resonate most when they’ve been collaboratively created with clients. The trouble is, unless the client has specifically come to you for visualizations, and thus knows what to expect, it’s hard to say “we should create a visual of this” without being met with blanks stares. My solution to avoid puzzlement was to present a sketch…”we could present it like this”. Well, wouldn’t you know, then I got stuck with the sketch.

What I did really like about it – as primitive as it was – is that it enabled a bit of storytelling in the presentation. This was one of those unfortunate “documents” that gets presented on a screen and distributed as a document, and so ends up with heaps of text and extras. But with the pictures, we could at least walk through the illustration to tell the story of change as it trickles through the organization. The fun addition we made to the illustration was the business architecture representation, which connects in with, well, lasers. If I can get lasers in there, I’m happy.

I also think of there being a static component, which is what I think of more as a visualization of some sort; some attempt to capture and synthesize a thought process, then produce a visual to represent the concept. This could be a model on a powerpoint slide, or a more elaborate and immersive visual representation, a mural, or interactive piece.

Last month I had the good fortune to work with a couple of incredibly talented scribes at the Middle East Summit of the World Economic Forum over in Jordan. What was very cool about the experience (okay, one of the many things that was cool about the experience) was that each of us was scribing in  a separate room over the course of a number of days, and what I saw emerge were not just different styles of scribing, but actually different approaches to modelling conversations.

It really got me thinking, and has started to bring me to a more nuanced set of thinking about the above visual thinking categories. Under the heading of graphic facilitation/recording, what I saw last month were four sub-categories, which I’d like to explore here:

• Connective
• Categoric
• Synthetic
• Illustrative

A little caveat; I’m not saying these are the only sub-categories, only that these were three distinct approaches the three of us seemed to be taking to represent the conversations in our respective sessions. Nor are the approaches exclusive, in fact, I would argue that in a given conversation, there would likely be great benefit in employing the three. What’s of major interest to me is what the approaches mean about how to listen, and what synthesis is done as a result of the listening process.

Connective Mapping

Simple connective mapping (by me)

I’ll start with the connective approach, as this represents my initial understanding of what graphic facilitation was supposed to do. This approach is very much a derivative of mind mapping; showing the connections between interrelated ideas. I find I still use this as my default, especially when trying to capture things in a hurry, as it imposes a simple and flexible taxonomy on the concepts as they come.

The approach is largely hierarchical, capturing “topics” as a central node that radiates outward into connected sub-topics. The deeper into a sub-topic the conversation goes, the more connectors will radiate out from a given node. The advantage is that you don’t really need to know where the conversation is going to capture it, and it is easy to go back and file subsequent ideas where they belong, contextually.

There are, of course, a lot of ways of doing this, but in terms of the way of listening to and interpreting a conversation, this means paying attention to the structure of the conversation, identifying the unifying concepts and drawing connections between them.

(Really) Simple connective mapping with sub nodes (by me)

Categoric Mapping

This was something that popped out at me when looking at the scribing of Sita Magnuson during the Summit. I think, from a structural perspective, categoric mapping is virtually the same as connective mapping; it is the presentation that is completely different. The organization, again, is hierarchical, in that major topics form the basis for clusters, but instead of a radiant layout, text flows into ordered categories.

Using text layout to map categorically (Sita Magnuson)

The challenge with this approach, and why I find it very appealing, is that it requires strong skills in text layout to differentiate topic areas. As you can see in Sita’s example, different size, font and color give a sense of differentiation and emphasis. In many cases here, the text itself is used as a connector between topics.

Also evident is the mixing of styles; there is some categoric mapping, some connective mapping as well as the next kind…

Synthetic Modeling

To really conceptualize this style, it helps to take a look at some of the models that I’ve put together below which attempt to capture some of the different levels within a conversation. There is a strong connection between this approach and the previous two, in that it relates to the conversation on a structural level, but takes the approach of addressing the structure itself, showing the relationships of the various elements with each other on a conceptual level. This example is from Kelvy Bird’s scribing at the Summit.

This style was really interesting to me. It begins to make the structure of the concepts and the conversation visible, while at the same time divorcing itself from the temporal element of the interaction.

A synthetic approach to modeling a conversation (Kelvy Bird)

Illustrative Modeling

The illustrative style is the one that, predictably, gives graphic facilitation its name. This is, of course, a style that is interspersed within each of the other methods to approaching the conversation. Illustration can be a part of the structure, it can be an independent element, or it can be the structure itself. The idea of illustration, however, is to directly engage meaning. In the connective approach, for example, meaning is derived; here, the meaning is conveyed in a direct manner. This goes back to my earlier point that visuals can be employed where words are obscuring meaning.

Illustrative modeling (by me)

In the example here, I’ve thrown in images that directly (I hope) trigger understanding and, by extension, better recall than could have been done with text.

This, for me, is much easier when people are having a conceptual discussion, rather than a tactical, detail oriented one. I’ve included illustrative modeling because of how it relates to some of the conclusions that have come out of this comparison for me.

A Model of Conversation

After sitting down with this and mulling it over, I found the only way I could start to make sense of it was to come up with some kind of model for conversations, listening and recording. Kelvy got me started thinking about different ways of listening, and this is my first cut at answering that. There seemed to be four elements to an interaction: flow/time, structure, meaning and concept.

Flow is quite simply a byproduct of the fact that conversations take place in time, and are thus necessarily carried out within some sequential confines. This is important, since time dictates how we are presented with information.

The four layers in conversation.

Structure varies based on a lot of things, not least of them the number of participants in a conversation, but also the style of presenting an idea, the contentiousness of the topics, as well as the relative emphasis or interrelatedness of various elements. Topic, sub-topic, sub-topic, new topic, important point, side note…this is the general structure of the conversation. It’s worth noting that this is separate from flow; the relationships which make up the structure do not necessarily line up in time (in fact, they rarely do…people ramble, digress, come back to points they have more to say on…).

Through all of this we assemble meaning. We pull out snippets here and there, and from this we put together our understanding of what is being said. Again, this is an independent element, which is – in part – why we can all leave the same conversation with a different understanding of what’s been said.

The concept layer, I think, is independent of meaning, though it is deeply connected. People can exchange meaning without generating a concept, but a series of related elements of meaning can begin to create something which is emergent.

I’ve sketched this out in two different models here for the purpose of getting an idea of how the various approaches map.

Elements of meaning plotted out in a scatter, without a time dimension.

First is the conversation itself, which I’ve drawn here as a scattershot of ideas, some bigger than others, some that are outliers and others that are clustered together. I’ve deliberately drawn this one differently from the model above to try and escape from the time dimension and focus solely on the content of the conversation. Admittedly, the two axes are meaningless. Ah well.

Your standard linear notetaking approach would look something like the line in blue, which is forced to follow the meandering path of the conversation whereever it may go in time.

The clustering or categoric approaches to listening, while still connected with the time dimension, pays most attention to correlation and clustering within the various clouds of meaning, paying attention to emphasis and affinity.

Linear notetaking following the temporal flow of the conversation. It actually probably isn't even this connected looking, now that I look at it.

The synthetic approach, I think, goes back to the original model of the layers of conversation; the ‘concept’ is separate from the structure. This separate layer, from a visual perspective, can be plotted out and used as the structure within which clustered elements can be organized.

The entire concept, for me, is very reminiscent of the old Taoist ways of representing communication and knowledge. There’s a saying about words being like the finger pointing to the moon; stare at the finger, and you’ll miss all the heavenly glory.

I would be interested to get some ideas and feedback on this (not least from Sita and Kelvy!)…how do others conceptualize ways of conversation, listening and of how to model what we hear?

Capturing correlated pockets of meaning through the clustering approach.

On the other hand, involvement from the group – if it isn’t structured enough – can lead to pretty confused images, or can force you to resort to “recording” rather than synthesizing.

I did a quick sketch of what I like as a process for engaging a couple of groups to create a stakeholder map; a complex interaction diagram that would model the actors, interactions and influences within a given system.

I like this approach because it gives the chance to really get some in depth perspectives before too much structure gets imposed on it, but gives lots of opportunity to introduce some visual frameworks and metaphors through the exercise to keep it creative. The result is a cohesive body of information with a lot of nested complexity; perfect fodder for using as an input for the development of the final visualization.

Image courtesy of NYTimes.com

It is somewhat vexing to see the hounds jumping on Alan Greenspan in the rush to assign blame for the financial crisis to a single individual. That’s the way of politics, I suppose, that many feel the need to find someone to single out, someone towards whom we can all point an accusing finger so we can, ourselves, feel absolved.

But for those with an interest in collaboration and how groups of individuals make decisions and work together, this whole mess brings up a question of a favourite text on collective decsion making; Surwiecki’s The Wisdom of Crowds.

What made me think of this was a quote from Greenspan in his Congressional hearing; “Those of us who have looked to the self-interest of lending institutions to protect shareholders’ equity, myself included, are in a state of shocked disbelief.”

What he’s talking about is a fundamental belief that through aggregated or collectivized self-interest and rational decision making on a distributed basis, better decision will be made…also known as the wisdom of crowds, group genius, or whatever you want to call it.

This, for me, points to a caveat for those who believe in collaboration and collective action as the best way to get things done:

1. actors will likely only act rationally when rational micro-action can be expected to happen within a context of rational macro-outcomes
2. collective action is effective only when the environment supports self interest that is rational

There’s a lot of overlap between those two things, but some of the distinctions, I think, are important. Ultimately, point #1 should create the “environment” for point #2. The issue is that within an irrational environment, there is little incentive to behave rationally, and in the case of the stock market and the mortgage bubble, there was significant (short term, as it turns out) incentive to behave in an irrational way.

If I think of the implications of this model for large organizations, it would seem that major organizational change can only really happen effectively if people can be assured that making good decisions within their part of the world will be met with collectively desirable outcomes at a macro level.

The lesson for all of us, and Mr. Greenspan, is that – as he suggests – we must continue to rely on the rational self interest of individuals and institutions, but it is the responsibility of leaders (and government) to set a rational context for those decisions.