Personal knowledge management and body of work

Got an email from my old colleague, Chunka Mui, this week that read

Here’s an out of the past question — do either of your archives have a copy of that HBS case study that included the Andersen AI work on FSA and Eloise? I got a message from an old colleague looking for it.

It wasn’t an unreasonable request.

Surprisingly, I was able to track down a copy. How that played out, however, is worth reviewing for what it might reveal about personal knowledge management in the real world.

The case study is a piece of my body of work. I wrote it in 1986, early in my doctoral studies. It’s out of print as near as I can discover. A digital copy, if it still existed, would be in Microsoft Word 2.0 format, predating Windows.

I was pretty sure that I had kept a physical copy. But that was five addresses and two downsizings ago. It wasn’t in the one file cabinet that was on the third floor. Eventually, I did find a hard copy in a stack of articles and file folders up on a shelf in a closet in my office. All of my old case studies are in that pile. At some point, I pulled them together with every intention of scanning them.

So. Is this a win, loss, or draw for my personal knowledge management system? It’s a win for Chunka. He got his answer in less than a day.

I got him that answer in less than a day. But it doesn’t feel like a win to me. Is that an indictment of my system or of my assessment? What I feel should have taken a few minutes turned into an hour or more spread throughout the day. All dependent more on my memory than on anything systematic in my practices. Plenty of room for improvement.


Review: Brief History of a Perfect Future

A Brief History of a Perfect Future: Inventing the World We Can Proudly Leave Our Kids by 2050. Chunka Mui, Paul Carroll, and Tim Andrews. 

Full disclosure, I’ve known and worked with the authors for over twenty five years. I’m biased. Largely because I have first hand knowledge of how smart they are and how deeply they think. 

In A Brief History of a Perfect Future, they take an old observation of Alan Kay’s–“the best way to predict the future is to invent it”–and turn that into a planning process you can apply yourself. 

When the goal is invention not prediction, you want to understand what you have to work with. You’d like to work with what is cheap and readily available. Today, computing power qualifies. Thirty years ago that wasn’t the case. Fortunes were made by those who recognized that trend and acted accordingly.  We’re still learning how to think about what’s possible with essentially free computation. 

Mui, Carroll, and Andrews extend that line of thinking into seven “Laws of Zero,” These are factors of production on similar improvement paths; computation, communication, information, genomics, energy, water, and transportation. What becomes possible when you can anticipate thousand and million-fold improvements in price and performance in any one of these elements? When they improve in concert? 

The authors play out potential scenarios in the next section of the book. What futures might we invent for ourselves? None are foreordained. None are impossible. The scenarios provided here are provocative and plausible. 

The point, of course, is not whether these particular futures come to pass. We’re not running a race on a fixed track. We’re building new roads to new destinations. Much more fun and rewarding than chasing someone else on their road. These are traveling companions worth getting to know.

There is no physics of human systems

An occupational hazard of acquiring a Ph.D. is adding lots of fancy new words to your vocabulary. Too often, these new terms devolve into weapons in status games. Sometimes, however, they clarify assumptions you didn’t realize you were making. “Nomothetic” and “idiographic” are terms you don’t pick up in casual conversation. They represent two distinct perspectives on making sense out of the world that explain a journey I’ve been on for years.

Let’s start with how the terms are defined;

  • nomothetic – relating to the study or discovery of general scientific laws.
  • idiographic – relating to the study or discovery of particular scientific facts and processes, as distinct from general laws.

Physics is all about the nomothetic; E=mc2 is true regardless. Economics aspires to be nomothetic, experience with trickle-down economics notwithstanding. Understanding Amazon or Apple, however, depends very much on the particulars of Jeff Bezos or Steve Jobs.

My own intellectual journey has been from the nomothetic to the idiographic. I was comfortable with mathematics and the hard sciences. Identify the problem, select the appropriate equation, crank the numbers and pop out the one right answer. 

I thought I would study psychology as an undergraduate but opted for statistics instead. I told myself this was because I was more interested in studying people than pigeons or rats. In retrospect, I wasn’t ready to accept that the messiness and particularity of human behavior as something you could study as a scientist.

Computer programming and systems design were holding actions trying to cling to the general and the universal. Debits and credits worked for all organizations and all situations. I could treat the vagaries of individual executives as mere noise within a larger system of order and predictability. 

I might have held onto that illusion if I hadn’t garnered a spot in the MBA program at the  Harvard Business School. Now I was at the center of the case study universe; the particulars anchored every discussion. I still didn’t have the words in my lexicon but I had begun the journey from the nomothetic to the idiographic; from the general to the particular. 

My comfort with individual courses correlated closely with where they fell on the spectrum from conceptual to concrete. Economics and accounting made sense, so did operations and finance. Marketing was a puzzle. Organizational behavior was mystifying; why did people insist on doing irrational things? 

My personal life was largely irrational at this point. Courtesy of wise friends I was getting professional help for that, but I still clung to a fading vision of systems and order. I went back to consulting and systems design. There were right answers to be developed and deployed. Resistance to those right answers was an aberration (if not an abomination) to be suppressed and eliminated. 

I could have pursued a classic strategy of accumulating enough power to impose my answers over the objections of weaker players. I’ve watched “my way or the highway” work (for certain values of “work”). I chose another path. I went back to sit at the feet of people smarter and more knowledgeable than I. What more could I learn from those who started by accumulating stories and case studies? I’ve written about this before 

This was where I finally learned of the distinction we’ve been talking about between nomothetic and idiographic. It’s comforting to discover that you are not the only one engaged in an important struggle. 

How we come to know things isn’t a conversation that happens in the average business context. In a world dominated by knowledge work and innovation, however, it’s a conversation that needs to move center stage. 

Organizations push for standardization and uniformity – this solves problems that organizations have to deal with. In the realm of knowledge work we have been too quick to assume generality before we’ve gathered enough data and insight about the richness and variety of actual practice.

In Ph.D. speak, nomothetic claims about human systems have to be filtered through an idiographic lens. 

That isn’t a sentence I would want to utter in a boardroom. What I would advocate instead is to drop one common phrase and substitute another. 

Stop talking about “connecting the dots.” Buried in that phrase is an assumption that there is already a right answer hiding in plain sight. The only work to be done is correctly recognizing the picture that exists. The particulars of this situation, in this moment, are mere noise obscuring what should be readily obvious. 

If you’ve come to hold the particulars in higher regard, then you will be better served to frame your task as that of “solving for pattern.” I’ve written about this shift of perspective before

There are no equations that we can plug data into and expect a clear answer. But for all the particulars in any new case, there are regularities and patterns we can use in shaping effective responses.

Increasing Knowledge Work Discipline

I stumbled across an interesting opinion piece in the Washington Post recently; want kids to learn math? Level with them that it’s hard. Learning to do knowledge work, like learning to do math, takes disciplined work. Tiago Forte is one of the people putting in the necessary work and sharing what he’s learning with the rest of us. In this piece, Forte turns to the kitchen and the realm of professional chefs for insight. Particularly, their obsession with order and preparation; mise-en-place.

Knowledge work lessons from the kitchen

Forte explores the following mise-en-place practices for relevant parallels to knowledge work 

1. Sequence

2. Placeholders

3. Immersive vs. process time

4. Finishing mindset

5. Small, precise movements

6. Arrangement

“Sequencing,” for example, looks at how decisions about the layout of tools and ingredients aids or hinders the flow of execution. You don’t want to be rummaging around for a whisk when you need one. Thinking through the process before executing will surface opportunities to sequence work to your advantage. 

Forte’s riff on “immersive vs. process time” may be the most interesting. Forte maps immersive time to Cal Newport’s deep work distinction. Process time on the other hand encompasses settings with a time component that runs independently of your attention and focus. In cooking, process time might be the 15 minutes it will take for rice to cook regardless of when you start the process. It may only take a minute of your attention to launch that cooking process but you want to take that minute of focus when it will do the most good. Start the rice now and it will be ready when you need it later. Wait, and you delay the entire process while you watch the rice cook. 

There are times when the back burner is the right place for the rice to be cooking while you pay attention to something more important. In knowledge work, the example that springs to mind is Stephen King’s advice to stick a freshly finished manuscript in a drawer for six weeks while you attend to something else. Doing that is taking advantage of your subconscious mind’s capacity to work on creative tasks without your input or attention. 

A sticky sequencing problem

Forte has been developing and sharing his system for managing knowledge work for several years now. His Building a Second Brain training course is highly regarded. Although I haven’t sprung for the course, I do pay for access to his other materials and have read his books. I’m picking on Forte here in much the same way that Alan Kay once described the Mac as the “first personal computer worth criticizing.” Whatever it’s flaws, his analysis sparks responses that move the entire argument forward. 

There’s a bit of a sequencing problem in his own analysis, here, because he’s invested so much time in his own systems. His analysis of mise-en-place is unavoidably influenced by the design decisions he’s already made and internalized. 

Here’s an example of where Forte’s current thinking distorts his analysis of mise-en-place. Talking about the role of sequencing tasks, Forte asserts;

Once we realize the importance of sequence, it becomes apparent that not all moments are created equal: the first tasks matter much more than the later ones.

I will grant the first part of his claim; all moments are not equal. It does not follow, however, that “first tasks matter much more than the later ones.” Forte wants to make a point about the importance of first tasks; that’s pretty conventional wisdom in multiple traditions and settings. But I think it is a hill too far to make a blanket assertion that first tasks always matter more than later ones. In fact, Forte contradicts his own point a bit later when he talks of the importance of a “finishing mindset.” 

You encounter ideas in the order you encounter them. The best you can hope for is to be cognizant of that and temper your assessments and enthusiasms for related ideas as you discover them.

A more troubling flaw in the argument

There is a lot to like in Forte’s exploration of mise-en-place. There’s a critical step in his argument, however, that I object to. Forte turns to the world of professional chefs because

It’s time that we had a working system for the mass production of high-quality knowledge work

“Mass production of high-quality knowledge work” is a nonsensical proposition. As a diner in a restaurant, my serving of Coquilles St. Jacques Provencale had better well be identical in all important respects with my dinner companion’s. But don’t try to hand me the marketing strategy you just prepared for my chief rival. Balancing Uniqueness and Uniformity in Knowledge Work has to drive the design of knowledge work practices and methods.

There are things to be learned about doing better knowledge work from contemplating mise-en-place. Knowledge work that makes an impact is more about inventing new recipes than about being more efficient and productive in churning out known recipes for tonight’s diners. You do want a solid mise-en-place set up before you embark on experimenting with recipes. You want your environment set up and tuned so that you can focus your attention and creativity on inventing some new and wonderful delectable. But if the lunch rush is at the door, you’re not doing knowledge work, you’re doing production work. And you need to be clear about the difference. 

Whining or Learning

In the early weeks and months of getting Diamond off the ground, we were intentional about the kind of organizational culture we wanted to create. Within the founding group we had decades of cumulative experience, mostly in organizations known for the strength of their culture together with a sprinkling of less pleasant experiences. 

In those early days, we spoke of “getting the band back together.” We were based in Chicago after all, so the Blues Brothers reference was a natural. 

Of course, what we were trying was more complicated. We weren’t getting _a_ band back together so much as we were trying to create a super group combining and mixing the talents of people who had been stars in their own groups. 

Consulting egos are rarely small; possibly never. Everyone had an opinion, often several. Whatever status or authority we might have had in prior organizations was politely acknowledged and promptly ignored. How things were done at McKinsey or Accenture or Booz or wherever was evaluated on the merits not the pedigree. 

More often than not, the differences were more cosmetic than substantive. Over time we were cobbling together a creole consulting language of our own. 

Sometimes the noise levels got out of hand. After one incident, one of my partners ordered up and distributed “No Whining” buttons to all of us. 

It was good for a laugh and a boost in morale. And it did help lower the temperature on some conversations. While I had a fraught relationship with this partner, I had also learned that their instincts routinely beat my analyses. Disrupting a conversational cycle before it degenerated into a shouting match helped. 

Unfortunately, “no whining” too often defaulted to no conversation at all. Labeling inquiry or pushback as “whining” was equivalent to “shut up and do it my way” or “if you can’t handle this, we’ll find someone who can.”

These sentiments might be marginally acceptable in some circumstances. If the work is sufficiently routine, problem solving can devolve into selecting a workable answer from a menu of known responses (I was about to suggest that you Google “the bedbug letter” but has a more interesting take—FACT CHECK: The Bedbug Letter.

I don’t live in that world. Neither did Diamond. And, if your work is knowledge work, neither do you. You get the problems that don’t have clear solutions. Which means you have to listen more carefully; to those you work with and to yourself. 

You must learn to distinguish between “whining”— a vague complaint that something isn’t fair—and the spark of discomfort signaling learning that needs to happen. Forbidding “whining” makes it harder to recognize the learning signals embedded in discomfort.

Habitual Struggles

One of my bad habits is obsessing about my deficiency of good habits. I know that good habits are important because the nuns and monks were relentless in telling me so; decades later their ambivalence about my academic success without suitable discipline still lingers.

I’m not devoid of good habits. Lately, I’ve had some success with Jerry Seinfeld’s keep the streak going/don’t break the chain trick (Jerry Seinfeld’s Productivity Secret), although Friday’s NY Times Crossword stumped me, breaking a 355-day streak. Two and a quarter years ago, I managed to reboot a journaling habit by starting a Morning Pages effort. 

Habits are on my mind lately for several reasons. Back in March, I finished Wendy Wood‘s Good Habits, Bad Habits: The Science of Making Positive Changes That Stick and thought it worth processing my thoughts into a post here. A bit slower than I would like, but not unusually so. 

More recently, I came across an interesting blog post from Eleanor Konik, The Difficulties of Teaching Notetaking » Eleanor Konik. I’ve been chipping away at improving my own note-taking practices and was toying with how and whether to incorporate any of that into my teaching. Konik’s argument is that no one implements good techniques simply because they’re good techniques. We’re all adept at cost/benefit calculations and the payoff structures for most “good” habits are hugely dependent on how heavily you weight the suasion of authority figures. For most of us, most of the time, the payoff for following objectively good practices doesn’t rise to the effort. 

This is where the accumulating research on habit formation comes in. Converting an activity from an invocation of willpower to automatic pilot effectively eliminates the effort and energy part of the equation. This is why Alfred North Whitehead’s observation continues to show up in discussions of habit:

 Civilization advances by extending the number of important operations which we can perform without thinking about them. Operations of thought are like cavalry charges in a battle- they are strictly limited in number, they require fresh horses, and must only be made at decisive moments. 

We know that thinking is expensive. As knowledge workers, thinking is a definitional component of your work. One of the highest returns on IQ points is reducing the demand for IQ points wherever possible. 

Woods offers an observation that I’ll close with; 

we repeatedly do the things we love doing. But we also grow to love the things that we repeatedly do.

The task then is to be more intentional about the practices whose repetition will lay down the habits that will free up the capacity for better “operations of thought” when those decisive moments arrive.

Don’t ask what is best

Dad was an engineer by training and temperament. His handwriting was as neat and precise as mine is not. He worked hard to express his thoughts clearly. 

My processes look and are much messier than his ever were. One trait I did inherit, however, was to worry about using language precisely. And, I was fortunate to have mentors who cared equally about precision. It isn’t something that seems a high priority in many current settings. 

There’s a question form guaranteed to set me off; “what is the best way to…?” I’ll grant that such questions are generally well-intentioned but are weighted down with so many unidentified and unarticulated assumptions as to be devoid of meaning. 

While such questions come from a good place, pushing back on their fuzziness won’t win you many friends. I’ve learned to be very cautious and selective about raising them in public. What I have tried to do is become more aware of when I am asking those forms of questions for myself. With that comes more effort to be more precise about the question I am trying to answer and to search for those assumptions. 

This is the underlying driver for my skepticism about exploring productivity in the context of knowledge work. Productivity is anchored in comparisons of “better” and “best” and I’m not sure those terms carry as much meaning as we might think. Instead, I’ve been working to shift my thinking toward the notion of “effectiveness.” It’s still too slippery a term but it does encourage me to widen my perspective to find those assumptions, understand the context, and be more precise about what I am hoping to accomplish. 

PKM Isn’t About Apps

I characterized my most recent post (Putting Personal Knowledge Management in Context) as potentially a grumpy old man rant. The triggering event was a side-by-side software review, Obsidian vs. Roam: Which PKM App is Right For You? – The Sweet Setup. My instant reaction was “PKM isn’t about apps” and thinking that it was will get you in trouble. I thought it might be helpful to explore that reaction and see where it might lead.

The review itself is a very well done comparison between the two apps in question. But it is based on an implicit assumption that is also flawed. Because the assumption is implicit, people aren’t likely to catch it and wonder later why they become dissatisfied with whatever choice they make. There is a clue lurking in one of the opening paragraphs:

It’s impossible to say “just use this one” when it comes to picking the right connected note-taking app for you. On the surface they may seem similar, but there are several important differences that stem from fundamentally different approaches to how your notes are stored and managed.

The author is acknowledging that they cannot, in fact, answer the question they have set before us. They then proceed to explore the question anyway because it’s an easy question to ask and answer and because we are accustomed to expecting questions to have neat answers. Worse, we only ask questions that we expect have answers. 

I expect this bothers me, in part, because the underlying topic is knowledge management. The underlying motivation for knowledge management, whether in organizations or for individuals, is dealing with questions that don’t have obvious answers. Or questions that provoke deeper questions. 

Asking questions has always been a high-risk behavior. Where will we end up if we continue to explore the PKM space as a practice for asking questions that don’t yet have answers?

Putting Personal Knowledge Management in Context

The notion of Personal Knowledge Management (PKM) is experiencing something of a renaissance. This rebirth is being driven by a combination of new apps, new ideas, and new thinkers promoting their wares. The apps, ideas, and thinkers are all worth paying attention to. At the same time. the brightness of shiny new things is obscuring important history and context.

At the risk of sounding like a grumpy old man yelling at people to get off of his lawn, I thought it would be helpful to look at some of that context in the hopes that it might make it less likely that we would repeat old mistakes. We should at least strive to make interesting new mistakes. 

If you set aside the notion that knowledge management could arguably be considered a synonym for library science, what we label knowledge management in organizations today was born in the late 1980s/early 1990s in the efforts of a number of knowledge intensive organizations (HP, IBM, Accenture, McKinsey, Toyota, etc.) to systematically leverage the things inside their workers’ heads. Chief Knowledge Officers were appointed (a hat I once wore), taxonomies were defined, religious debates were held over the relative merits of Lotus Notes and Microsoft Sharepoint. Today, knowledge management is a reasonably well-defined function within most large organizations. 

Enterprise knowledge management was built on the premise that the number of knowledge workers whose knowledge mattered enough to manage was a small and easily identified subset of the workforce as a whole. Knowledge management was a hedge against having the knowledge in those smart heads walk out the door. 

The knowledge management problem changes when the proportion of the workforce classified as knowledge workers represents a significant fraction of the workforce. When everyone is a knowledge worker, knowledge management becomes personal not organizational.

A classic enterprise knowledge management problem is that of persuading the knowledgeable to share their knowledge with the rest of the enterprise. The naive hypothesis was that knowledge workers hoarded knowledge to preserve and protect their organizational status and position. A slightly less cynical take was that knowledge workers needed help to unpack and externalize their expertise so that it could be shared. 

My take is that the average knowledge worker has no clue about what it would mean to manage their knowledge and no useful models to emulate. You see individual executives and knowledge workers using email as their primary knowledge storage structure. You see arguments that enterprise search engines will bring Google inside the organization and prove sufficient to find and reuse knowledge assets when the time comes. You see elaborate folder and directory structures trying to impose order on proliferating documents and deliverables. 

The rise of personal computing and the web encouraged some knowledge workers to experiment with better ideas. Wikis, blogs, and bookmark managers were pressed into service. Ideas that were ahead of the technology curve (Memex, Augmentation, Dynabook, Mundaneum, and, recently, Zettelkasten) have been dusted off and revisited. 

The latest round of innovation and experimentation holds great promise. As an individual knowledge worker, you have several choices. One is to do nothing and wait for the dust to settle. A second is to place your bet on one of the current players and hope your support contributes to that player becoming the winner. 

A third strategy—and the one I am pursuing—is to recall an observation Peter Drucker once made about the productivity gains made during the early years of the 20th Century;

Whenever we have looked at any job – no matter how many thousands of years it has been performed – we have found that the traditional tools are wrong for the task.

Peter Drucker

Those extraordinary gains flowed from examining tools and task and rethinking the combination in parallel. Changing tools without changing the task is a recipe for speeding up the mess. Changing tasks without rethinking tools will make the current mess a morass.

Personal knowledge management has to be one component of a personal quest to become a more effective knowledge worker.