Favoring substance over soundbites

imageThe problem with communication ... is the illusion that it has been accomplished.
—George Bernard

An associate once suggested to me that in his experience, nearly all problems he encountered while engineering complex systems were the result of communications breakdowns. This assertion reminds me of one of my high school teachers, who had a sign which read: Eschew Obfuscation on her desk. For those who don't get the irony immediately, this message directs us to avoid confusing your audience, yet rarely is understood, because the audience doesn't understand the message.

Don Reinertsen describes some of the mistakes teams make in trying to address these communications breakdowns:

Almost every company would like to improve communications on their development teams. Too often they think this requires getting more information to flow faster inside the organization. More people are invited to meetings. More decisions are documented. More memos are sent to more people. Do communications improve? No Ironically, the difference between companies that claim to have good communications and companies that have bad communications is never the quantity of information that is exchanged. Companies that have bad communications can still be buried in information. The key driver of good communications is the amount of usable information communicated compared to what is needed to make good communications.

Communications channels come in many different flavors, and each has benefits and challenges. However, one of the things common to nearly all of them is that they are imperfect. Messages can be lost or altered in transit, and will have widely varying levels of effectiveness with respect to communicating what was originally intended. Different transmission mediums also have varying levels of intrinsic delay, and thus may produce results that are difficult to reconcile with each other at any particular time. Different communications channels also vary widely in their tolerance to noise. For example, when presenting concepts to a large audience, oral presentations, visual graphics, and written narratives can each be quite effective for defining notional intent, while they are quite inadequate in communicating specific details. You may see everything as sunny and rosy, whereas from your audience's entrenched position (often juggling many factors you don't even know about), it might be raining cats and dogs.

Unless approaches are used which will close this 'communications gap' with those you are communicating with, so your intended meanings are understood (and even accepted), you may increase your audience's confusion more than their support for your ideas. This can frustrate everyone, especially when the situation you are describing is novel or complicated from your audience's perspective, if the messages you strive to deliver build upon each other, or when the behaviors you are advocating is controversial, or departs significantly from traditional practices.

The communications researcher Osmo A. Wilo published a collection of rules of thumb about communications a number of years ago, while he was a member of the Finnish Parliament. Wilo's rules include (and elaborate on) the following points:

  • The probability of communications success is inversely proportional to the square of the number of paths traveled between the transmitters and receivers. This is why most communications is more likely to fail than it is to succeed.
  • The more information we attempt to communicate before we elicit feedback, the less likely it will be for the information to be successfully transferred and understood
  • Perceptions matter as much or more than facts do. If a message can be interpreted in several ways, it will likely be interpreted in the most detrimental way possible
  • There will always be someone who thinks they know what a message is intended to mean

To complicate things further, the problem often is with the message itself, rather than with the communications channels. Ambiguities may be the key tools of selected roles in a given situation. For example, if you are a politician, or evangelist seeking to sell a new idea, you likely will favor abstract messages over concrete ones. "We’ll invest all our considerable resources to satisfying the needs of our many constituents" sounds much better than "I’ll spend $10 million of the new taxes you're going to have to pay so I can build a highway that will only benefit my biggest campaign contributor." Finally, when your communications medium is conversational, it does not scale.

The first step in addressing these challenges is to tailor the message to the audience and purpose. Until that is done, it doesn't matter what communications channel you use. Then, realize that there are many reasons why people want (and may even need) to embrace abstract, rather than concrete, concepts:

  1. Preferences of communications style

    Part of the reason we often try to use abstract terms is because we are wired to use them to share lessons we have experienced ourselves. Narratives and associated storytelling have become comfortable ways of finding a common history within different cultures, and are used to begin to form agreements with one another for cooperative work. We often work hard for the lessons we've learned in our lives, and we're usually proud of what those experiences have taught us, and enjoy passing those experiences on to others. Abstract terms help us describe things in general ways, rather than relating them to specific events which might not be meaningful to those who have not had our specific experiences.

    Perhaps a friend promises to pick us up at four o'clock, and then leaves us standing there until six. Perhaps your partner promised you a night out on the town, then cancels to do something with someone else. From such experiences, we might realize we can't always trust people. Do we go into each of these specific stories with someone we are mentoring who is in a related situation? More probably, we just tell them "Some people will let you down." It may have taken many concrete, specific experiences to teach you such a lesson, but you try to pass it on to another person with a few general words. Your mind has captured the pattern, and likely has associated it with your perceived risk about unfortunate future circumstances.

    By using abstractions, you may think you will be able to more quickly communicate with your mentor, transferring your lesson without their pain. Unfortunately, the old adage, "No pain, no gain", applies here; the emotional impacts are what originally enabled you to learn this lesson, and the abstract concept which you categorized it as only occurred after the fact. "Some people will let you down" may be a fine title for an essay - but if you want to transfer such lessons reliably, you'll have to communicate the concrete, specific details, and provide an opportunity for dialog with the individuals who have to put such principles into action (perhaps answering their hard questions about when you should trust people, and when you shouldn't). Otherwise, the lessons you are seeking to transfer may not be exchanged successfully.

  2. Desire to hide disagreements in order to promote morale

    Fuzzy language is often used as a way of pretending that concrete disagreements don't exist. Unfortunately, this often results in use of 'business speak' (or BS) - generalities that no one may be able to achieve with demonstrable results, but everyone can collectively engage in the pursuit of.

  3. Gaps in subject matter expertise

    Participants may not have a detailed understanding about the nature of subject matter. This typically manifests itself in an inability to express exactly what would be required to address gaps in this knowledge for a given situation (often expressed as "I'll know it when I see it").

  4. Underestimating the future costs of ambiguity

    As long as things are kept at an abstract level, no tradeoffs have to be made. Yet when push comes to shove, and specific, concrete actions must be implemented, constraints suddenly emerge - usually over time, money,  throughput, or critical resources. An optimist believes that it will be easier to address such constraints and tradeoffs in the future than it will be at present. Determining how to respond to these constraints typically requires a level of common understanding that abstractions will not provide. Decision-makers must have access to meaningful, concrete facts and data to make their decisions, yet such information may not be available if ambiguity has not been isolated and progressively eliminated. The quality of their decisions will directly depend upon the timeliness and integrity of this information, because in such situations, decisions cannot wait for better data to be produced. This is a classic example of one of the hidden costs of abstractions.

  5. Insufficient foundation for meaningful communications to occur

    A sufficient foundation may not yet have been established for effective communications. Such a foundation typically includes a standardized terminology upon which other concepts are based, and that all users can then derive their own conclusions and principles from. There also may be a forum for the disambiguation of terminology as it is being used, and according to the context in which these terms occur. This is especially true when different people in the communications channel have different contexts or domains that they are operating from.

The terminology which we use may be concrete, yet still may be general, rather than specific. General terms and specific terms are not opposites, as abstract and concrete terms are; instead, they are the different ends of a range. General terms refer to groups, while specific terms refer to individuals, with room in between. Let's look at an example. Furniture is a general term; it includes within it many different possibilities. If I ask you to form an image of furniture, it won't be easy to do. Do you see a department store display room? a dining room? an office? Even if you can produce a distinct image in your mind, how likely is it that another person will form a very similar image in their own mind? While furniture is a concrete term (it refers to something we can see and feel), its meaning is still hard to pin down, because the group is so large. is it possible to form positive or negative feelings toward furniture? Again, it's hard to develop much of a response, because the group represented by this general term is just too large.

If we shrink the categorization space by adopting a more specific term, chair, it's easier to conceptualize - a chair gives us a clearer picture than the term furniture. But let's shift our thinking to an even more specific term, like a rocking chair. Now the image is likely to cause us to associate meaning with the name of this class of thing, because of past associations. The images we recall in response to this specific term are indeed more likely to have similar functionality (though may still differ in size, materials, design, and fabrication). We are therefore much more likely to have similar emotional associations (comfort, relaxation, calm). The less general or more specific a term is, the more clearly it may help us communicate desirable emotional responses, than the more general or less specific terms we might have used before such details could be recalled form our memory.

We can become even more specific. Our focus can become a lime green velvet La-Z-Boy rocker recliner with a cigarette burn on the left arm and a crushed jelly doughnut pressed into the back edge of the seat cushion. By the time we get to this last description, we have become very specific, and it is much easier to visualize what we are looking for, which is the first step in charting the course towards this destination. Since our minds are pattern recognizers, when we encounter something new, we try to plug that thing into our minds by relating it to something from our past. If our history included an incident in which we were punished for damaging such a chair, our response would likely be negative rather than positive. This is the risk and the reward inherent in making things more concrete.

Typically, most development efforts are well served by producing a hierarchy of abstractions which serve as themes which may scope the activities by which lower-level, more detailed requirements are captured. Some guidelines will help to frame the characteristics of such requirements. The following steps can help to begin addressing project ambiguities by establishing an appropriate concrete requirements baseline for your project:

  1. Begin with the end in mind, by defining the intentions, scope and deliverables required to change things, and who those changes are expected to benefit.
  2. Identify the reference domain knowledge and terms of reference which are necessary for understanding the focus and context for the endeavor.
  3. Adopt a standard format for work products, an appropriate checklist for assuring the presence of quality attributes, and a repository for managing information over time
  4. Actively manage the scope of your endeavor - explicitly define what is and is not 'in' that scope, in terms of organizations, customers, products, roles, and resources. Do this for a well-defined period of performance, while preserving the ability to tweak this in later iterations.
  5. Minimize the number of intermediary hand-offs between transmitters and receivers, as each introduces noise and the potential for misunderstandings. The greater the number of connections and volume of information that is being transferred, the greater the frequency that problems will occur
  6. Solicit and elicit inputs from a broad cross-section of stakeholders to identify the key abstractions that require elaboration.
  7. Synthesize responses received into a composite work product
  8. Apply George Orwell's six rules from Politics and the English Language to this work product.
  9. Establish a review and feedback process so work products are reviewed by knowledgeable subject matter experts
  10. Provide sufficient development and review cycles and resources so that  issues identified in this process are given sufficient attention to be resolved in a timely fashion
  11. Establish a change management plan for requesting changes to these work products while in use, making sure that consensus decisions about the disposition of those changes are communicated

As we lead a development team, these collective objectives should not cause us to cloud our intentions, misrepresent what we are building, or pass on ambiguity to others downstream, thinking that somehow things will automatically become clearer with time. Instead, everyone's goal should be to identify the actions required to collect and communicate information to the stakeholders who need that information, so they can perform their work effectively. While there often is  a role for abstractions to play early in these communications exchanges, we should generally seek to base as many of our exchanges as possible upon concrete specifics as soon as possible.