Marcel van Oosterwijk

Category: Knowledge & AI

How knowledge and understanding are created, represented, preserved, synthesized, transferred, and developed, including the changing role of AI in knowledge work.

  • Practice Note: The Death of the Document?

    Organizations rarely suffer from a complete absence of information. Especially around long-lived products, systems and capabilities, the opposite is usually true. Years of work leave behind requirements, architecture descriptions, source code, test cases, assessment reports, incident records, change requests, presentations, meeting notes, operational procedures, manuals, roadmaps and countless other artifacts. Some are current, some are obsolete, some contradict one another, and some look like little more than old junk.

    Yet hidden among them may still be a surprisingly large part of the organizational understanding that people assume has already disappeared. The problem is that nobody is going to read all of it.

    For a long time, that limitation shaped how we thought about organizational knowledge. If knowledge was going to be useful, somebody first had to turn it into something another person could reasonably consume. Knowledge had to become a document. The document had to be found. The document had to be read. Only then could the knowledge become useful again.

    AI may be changing that sequence more fundamentally than we realize.

    Knowledge Prepared for Reading

    Traditional knowledge management makes considerable sense when the eventual consumer is a human reader. Suppose the useful understanding of a twenty-year-old product is scattered across thousands of requirements, test cases, architecture descriptions, incident reports, change records and pieces of source code. In theory, much of the knowledge is still there. In practice, it is inaccessible because no person can economically reconstruct it whenever a question arises.

    The natural response is therefore to consolidate. Someone reads the sources, identifies what matters, resolves contradictions, creates structure and publishes the result as a specification, architecture description, product reference, handbook or some other authoritative document. That document reduces the amount of material the next person has to process. Instead of reading thousands of artifacts, the reader consumes one carefully prepared representation.

    The model is essentially:

    Knowledge → Publication → Reading → Understanding

    This is not a bad model. It solved a real problem.

    But it also created another one. Every publication became another representation that had to remain synchronized with the knowledge from which it was derived.

    The Cost of Publication

    The work does not stop simply because somebody has written the document. Requirements continue to change. Architecture evolves. Code changes. Incidents reveal new information. Assessments produce findings. Decisions are made in meetings. New constraints appear. Old assumptions become invalid. The publication therefore begins ageing almost immediately.

    Keeping it useful requires somebody to notice relevant changes, determine how they affect the document, update the appropriate sections and ensure that the resulting representation remains internally consistent.

    If several publications depend on the same underlying knowledge, the problem multiplies. The same new insight may need to appear in a product reference, an architecture description, a methodology, training material and assessment guidance.

    The problem is no longer simply preserving knowledge. It is maintaining all the representations through which people are expected to consume it. That cost was difficult to avoid when publication was the necessary bridge between accumulated information and human understanding.

    But what happens when the bridge is no longer always necessary?

    From Reading to Asking

    Imagine someone encounters a strange piece of code in a twenty-year-old system. Traditionally, the hope would be that somewhere there is documentation explaining it. Perhaps an architecture description contains the rationale. Perhaps an old requirement explains the business rule. Perhaps somebody wrote a decision record. If the right document exists, is current, can be found and contains the answer, the problem is solved.

    But the explanation may never have existed in one place. The relevant evidence might instead be scattered across an old requirement, a source-control change, two regression tests, an incident report and a presentation written three years later.

    A human is unlikely to discover all of those relationships. An AI-enabled knowledge environment potentially can.

    The person no longer needs to ask:

    Which document should I read?

    They can ask:

    Why does this code exist?

    The system can retrieve relevant evidence from several sources and synthesize an answer for that particular question.

    That is a different model of knowledge consumption. The answer did not have to be published before the question existed. The representation was created because the question existed.

    The Question Becomes the Interface

    This changes the role of the document. For much of organizational history, documents have served as interfaces to knowledge. Someone had to anticipate what future readers might need, decide how to organize it and create a representation that could later be consumed.

    The document therefore bundled many possible future questions into one predefined structure. A person looking for one answer might have to read twenty pages because the author had to write for many readers and many possible purposes.

    When knowledge can be interrogated directly, the question itself can define the representation. A Product Owner may ask what business outcomes and constraints apply to a proposed change. An architect may ask which assumptions shaped a particular design decision. An assessor may ask which quality expectations apply and what evidence supports them. A developer may ask why a seemingly unnecessary rule exists. Each question creates a different view over the same underlying body of evidence.

    The prompt becomes, in effect, a temporary specification for the representation required at that moment. That does not make documents obsolete. It changes their status.

    Documents Become Views

    There are many situations in which a stable document remains useful. A contractual context may require a frozen specification. An assessment may need a citable baseline. Governance may require a reviewed reference. A publication may deliberately capture the state of understanding at a particular point in time. In those situations, generating and preserving a stable document makes perfect sense.

    But the document no longer has to be treated as the knowledge itself. It can be understood as one representation of a larger body of knowledge, produced for a particular purpose at a particular moment.

    Tomorrow, another audience may need another representation. And sometimes no publication is needed at all. Someone asks a question, receives a synthesis supported by relevant evidence, makes a decision and continues working. The organizational knowledge has been consumed without first being packaged into a document.

    Reachability Becomes the Problem

    Once publication is no longer the mandatory interface, the knowledge-management problem changes.

    The first question is no longer:

    Where should we put all our knowledge?

    It becomes:

    Can we reach the relevant evidence when somebody needs to understand something?

    That distinction has architectural consequences. Requirements do not necessarily need to be copied into a central knowledge repository. Architecture information does not need to be moved simply because other people may need it. Code does not need to stop living in source control. Operational evidence does not need to become prose before it can contribute to organizational understanding. The sources can remain where they belong.

    What matters is that the knowledge environment can reach across them. Depending on the technology, that may involve indexing, linking, retrieval, ingestion or some combination of them. The implementation can vary.

    The principle remains simple:

    Do not move knowledge merely to centralize it. Make it reachable.

    Old Information Changes Role

    This also changes how we should think about accumulated historical material. When the objective is to create one clean authoritative document, old and contradictory information looks like a problem to eliminate. Before publication, someone has to decide what is current, what is obsolete and what can safely be discarded.

    A knowledge environment built around interrogation has different needs. An obsolete requirement may no longer describe what the product should do today, but it may explain why a piece of code was introduced fifteen years ago. A superseded architecture decision may reveal which trade-off originally shaped a subsystem. An old incident may explain an otherwise mysterious constraint. A test case may preserve a business rule whose original requirement has disappeared. Obsolete is not the same as worthless. Historical evidence can remain useful precisely because it is historical.

    The challenge is therefore not to make every source appear equally true. It is to preserve enough provenance to interpret each source correctly. When was this recorded? What did it apply to? Was it superseded? What is authoritative today? Which other sources support or contradict it?

    Without those distinctions, broad access produces confusion. With them, history becomes evidence.

    Connect First, Curate Where It Matters

    This raises another possibility. Traditional knowledge initiatives often begin with extensive cleanup. Thousands of artifacts must be reviewed, classified and curated before they are considered suitable for inclusion in the knowledge base. That made sense when the objective was publication. Someone had to decide what deserved to appear in the finished representation.

    But if AI can inspect sources at question time, asking humans to inspect everything first may recreate exactly the economic problem AI was supposed to help solve.

    It may be more useful to begin broadly. Connect the old requirements. Connect the architecture documentation. Connect the test repositories, assessment reports, change histories and incident records. Make the surviving evidence reachable, subject to appropriate security and governance.

    Then allow actual questions to reveal where human curation is valuable. Some sources will rarely matter. Others will repeatedly appear in important answers. Some contradictions will prove significant enough to resolve. Some missing rationale will become visible because the available evidence cannot explain what exists.

    Curation does not disappear. It becomes more targeted. Instead of deciding in advance what every future reader might need, human effort can concentrate on the areas where real questions expose uncertainty, conflict or importance.

    The Knowledge Environment Must Be Able to Say “I Don’t Know”

    There is an obvious danger in all of this. A system capable of synthesizing across thousands of sources can produce an explanation even when the surviving evidence does not justify one. Reachability without provenance can therefore create the illusion of understanding.

    A useful knowledge environment must preserve the difference between what is explicitly recorded, what is supported by several sources, what has been reconstructed from indirect evidence, what is disputed and what remains unknown.

    A plausible explanation should not quietly become historical fact merely because it sounds convincing. Sometimes the correct answer to a prompt will be that the available evidence does not explain why something exists. That is not a failure. It is knowledge about the state of organizational understanding. And it tells us where human attention may still be required.

    From Knowledge Prepared for Reading to Knowledge Prepared for Questioning

    This may be the deeper change. Historically, organizational knowledge had to be prepared for reading. Someone had to decide what mattered, assemble it into a coherent representation and publish it in anticipation of future need.

    AI makes another model increasingly practical:

    • Preserve the sources.
    • Preserve their provenance.
    • Preserve deliberately what cannot reliably be reconstructed.
    • Make the rest reachable.
    • Then allow future users to ask the questions that actually matter when they matter.

    The knowledge environment no longer has to anticipate everything somebody may eventually need to read. It has to preserve enough reachable evidence that somebody can ask what they need to know.

    That changes the economics of Understanding Debt as well. A twenty-year-old system may appear poorly understood not because all of its understanding has disappeared, but because the surviving pieces are fragmented across thousands of artifacts that no human could economically reconstruct.

    Making those fragments reachable does not magically restore everything. Some rationale will genuinely have disappeared. Some decisions will remain ambiguous. Some evidence will contradict itself. But fragmented knowledge is no longer necessarily inaccessible knowledge.

    Reflection

    Perhaps we have spent too much time asking where organizational knowledge should live. That question naturally leads toward repositories, migrations, canonical documents and the difficult work of keeping them synchronized.

    The more interesting question may now be how organizational knowledge should be consumed.

    For a long time, the path was largely:

    Knowledge → Document → Read

    Increasingly, another path becomes possible:

    Knowledge → Question → Synthesis

    Sometimes that synthesis will become a document. Sometimes it will become a decision, an explanation, an assessment or another question.

    The important change is that publication no longer has to come first. Knowledge does not necessarily need to be assembled into the representation somebody might one day need.

    It needs to remain sufficiently connected, contextualized and trustworthy that the representation can be created when the need appears.

    The document is no longer always the interface. Sometimes the question is.

  • Practice Note: When Practices Become Rituals

    I once worked in an organization where “DoD” had become a verb. People would ask, “Has the Product Owner DoD’ed it yet?” What they meant was whether the Product Owner had performed acceptance testing.

    Nobody seemed particularly bothered by the wording. It had simply become part of the local language. At first I treated it as harmless shorthand, but the more I thought about it, the more revealing it became. Definition of Done had not merely been interpreted loosely. Its meaning had drifted far enough that the name of one practice had become attached to a different activity altogether.

    The activity had survived, in name only. The original reasoning had not. That made me wonder whether this is how practices become rituals.

    Practices Perform Functions

    Professional practices rarely exist for their own sake. Code reviews, architecture reviews, retrospectives, readiness checks, governance forums and countless other mechanisms exist because someone believed they performed an important function within a larger system. The meeting, checklist, template or ceremony is therefore not the purpose. It is one way of realizing the purpose.

    A code review may exist to reduce defects, spread knowledge, challenge design decisions or strengthen collective ownership. An architecture review may exist to expose risks before they become expensive to reverse. A retrospective may exist to create structured learning from recent experience. The visible activity matters because of what it enables.

    The difficulty is that organizations often preserve activities more successfully than they preserve the reasoning behind them.

    When the Why Fades

    Recurring meetings remain in calendars. Templates are copied forward. Checklists stay embedded in workflows. Governance gates survive reorganizations. New employees quickly learn which steps must be followed and which approvals must be collected.

    The rationale is more fragile. Over time, people may become very good at performing a practice while becoming increasingly uncertain about what the practice is supposed to achieve. Eventually the organization no longer asks what outcome the activity is intended to create. It asks whether the activity has been completed. That is when a practice begins to resemble a ritual.

    This does not necessarily mean that the ritual has become useless. It may still perform its original function reasonably well. A weekly meeting may continue to coordinate work. A review may still expose important risks. A checklist may still prevent mistakes. The problem is subtler: the organization no longer knows with confidence why the practice works, what part of it matters, or what would happen if it changed.

    Rituals Can Look Healthy

    This can make ritualization surprisingly difficult to detect. A ritualized practice may look perfectly healthy from the outside. The meeting happens every week. The checklist is completed. The approval is recorded. The template is filled in correctly. Compliance can remain high even while understanding becomes weak. The organization becomes good at doing something it can no longer explain.

    And once that happens, the continued existence of the practice can begin to justify itself. We do this because this is what we do. The realization becomes its own reference. Eventually frustration appears, and the discussion often moves directly to removal.

    Do we still need this meeting?

    Can we simplify this checklist?

    Can we eliminate this approval?

    Can we automate this gate?

    Those may all be reasonable questions, but they begin one step too late.

    Recover the Function

    Before deciding what to do with a practice, we need to understand what function it was intended to perform.

    Perhaps the function remains important and the current practice still performs it well. Perhaps the function remains important but the practice has become a poor way of realizing it. Perhaps several practices now duplicate the same function. Perhaps another mechanism has made the original practice unnecessary. Or perhaps the function itself no longer matters.

    These are very different situations. They cannot be distinguished by looking only at the visible activity.

    This matters because organizational evolution does not require preserving practices unchanged. Quite the opposite. A practice may need to change considerably in order for its underlying function to remain healthy.

    Evolution Does Not Mean Preservation

    A review meeting might become asynchronous. A checklist might become automated evidence. A central governance forum might be replaced by distributed decision-making. A formal approval might disappear because the confidence it once created is now produced continuously through better controls.

    If the important function survives, the organization may have evolved successfully even though the original practice has disappeared. Preserving the activity is therefore not the same as preserving what matters.

    The opposite problem also occurs. Organizations sometimes retain practices long after their function has ceased to be relevant. The meeting remains because cancelling a recurring meeting feels more consequential than letting it continue. The report survives because someone might still need it. The approval stays in place because nobody is quite certain what would happen without it.

    Practices accumulate much like old systems. Each one may once have had a perfectly good reason to exist. Over time, the artifact survives more easily than the understanding behind it. Eventually the organization inherits a landscape of ceremonies whose origins are difficult to reconstruct.

    But the loss of rationale can also push in the other direction. A practice whose function is no longer understood can appear pointless precisely because the organization has forgotten what value it once created.

    People see the meeting but not the coordination it enabled. They see the review but not the risk it was intended to expose. They see the documentation but not the decisions it helped preserve. A practice may therefore be preserved for the wrong reason or removed for the wrong reason. In both cases, the deeper problem is the same: the organization no longer understands the function well enough to make a deliberate choice.

    From Practice to Realization

    This suggests a simple principle. Before changing a practice, recover the function.

    What problem was it intended to solve? What risk was it meant to control? What understanding was it supposed to create? What decision did it support? What evidence did it produce? What coordination did it make possible?

    Once those questions can be answered, the organization has choices.

    The practice may remain.

    It may change.

    It may be replaced.

    It may disappear.

    What matters is that the decision concerns the function rather than merely the artifact.

    Seen this way, a practice is one realization of an organizational need. The need describes what the organization must somehow accomplish. The practice describes one way of accomplishing it.

    That distinction makes deliberate evolution possible. If the practice is confused with the function, any change to the practice can appear threatening because it seems to endanger the need itself. If the two remain distinct, the organization can experiment with different realizations while preserving what actually matters.

    A Continuity Problem

    This is also why the problem belongs naturally to Continuity.

    When a practice becomes ritualized, the organization has preserved the visible artifact of an earlier decision while losing part of the understanding that made the decision meaningful.

    The meeting survived.

    The checklist survived.

    The approval survived.

    The process survived.

    The why did not.

    Once that happens, the organization may still change the practice, but change becomes guesswork. It may still retain the practice, but retention becomes habit. It may still remove the practice, but removal becomes an experiment whose consequences are difficult to predict.

    Continuity therefore does not mean preserving practices. It means preserving enough understanding to know what may safely change.

    Reflection

    Perhaps practices do not become rituals because people stop following them. Perhaps they become rituals when people continue following them after the reasoning behind them has faded. The activity remains visible, familiar and repeatable. The function becomes harder to articulate. Eventually someone asks the deceptively simple question:

    Why are we doing this?

    That question should not automatically be treated as evidence that the practice is useless. It may instead be evidence that something more important has already been lost.

    The practice can always be changed. The harder task is preserving enough understanding to know what the change must not accidentally destroy.

  • Practice Note: The Collection As a Subject

    I have been thinking about watches. This has proved unexpectedly dangerous, not because choosing a watch is particularly difficult, but because somewhere along the way I stopped thinking about the watches themselves and started thinking about the collection.

    At first that seemed like a small shift in perspective. A watch is obviously something that can be understood and assessed. It has characteristics, functions, limitations and qualities of its own. A collection, by contrast, can easily look like little more than several watches placed next to one another.

    But that is not quite true. The moment we draw a boundary around several individual watches and consider them together, new characteristics appear. We can ask whether the collection provides useful variety, whether important situations are poorly supported, whether several watches contribute almost the same thing, whether one member carries too many important roles, or whether removing a particular watch would materially weaken the whole.

    None of those questions belongs to any one watch. They arise from the composition. The collection has become a subject in its own right.

    Changing the Boundary

    Suppose four watches are placed on a table. Each can be assessed independently. One may be unusually robust, another elegant, another highly functional and another particularly enjoyable to wear. Those qualities belong to the individual objects.

    Now imagine drawing a conceptual boundary around all four.

    The questions change. Do these watches complement one another? Does the collection provide meaningful choice? Are some situations represented strongly while others are not represented at all? Is there deliberate overlap, or merely repetition? Does the collection depend heavily on one member for several important situations?

    Nothing about the watches themselves has changed. Only the boundary has changed. Yet changing the boundary changes the subject.

    This is important because it means that a portfolio cannot always be understood by aggregating what we know about its members. Some of its most relevant characteristics exist only at the level of the whole.

    Individual Quality Does Not Determine Portfolio Quality

    This creates a distinction between the quality of a member and the quality of the portfolio it belongs to. A collection may consist entirely of excellent watches and still be poorly composed.

    Imagine four outstanding watches that all serve roughly the same role. There may be nothing wrong with any of them, and there may be perfectly good personal reasons for owning all four. But their individual excellence tells us surprisingly little about whether the collection as a whole is suitable for what its owner wants from it.

    The reverse can also be true. A comparatively ordinary watch may make a disproportionately important contribution because it provides something the rest of the collection does not.

    The relevant question is therefore no longer simply whether a watch is good. It is what the watch contributes to the collection.

    That is a different assessment problem.

    The Whole Needs Its Own Intent

    Once the collection is treated as a subject, another question becomes unavoidable.

    What is the collection actually for?

    That purpose cannot be inferred safely from whatever happens to be in the watch box today.

    One collection might be intended to cover the realistic situations in which its owner wears watches. Another might explore historically important designs. A third might focus on mechanical movements, a particular manufacturer or watches associated with personal memories.

    The same set of watches could therefore make sense against one intention and make very little sense against another.This means that the collection needs an intent of its own.

    Without that intent, composition becomes difficult to judge. More watches may look like progress simply because there are more of them. Greater variety may look desirable simply because variety is conventionally associated with collecting. Missing categories may appear to be gaps even when they have no relevance to the owner. Once intent becomes explicit, those assumptions can be challenged.

    A Portfolio Reference Model

    From there, the idea of a Portfolio Reference Model follows quite naturally. Such a reference would not prescribe which watches should be owned. Instead, it would describe what the collection as a whole should enable.

    It might express the situations that matter, the forms of variety that are valuable, the level of overlap that is useful, the dependencies that should be avoided, the constraints that matter and the areas the collection deliberately does not need to support.

    The distinction between requirement and realization is important here. A conventional watch category such as dive watch, dress watch or chronograph may be one way of satisfying part of the reference, but the category itself is not automatically the requirement. If I begin by assuming that a complete collection must contain one of each recognized type, then the classification system has already started dictating the answer.

    The reference should come from the purpose of the collection. The watches come later.

    Composition Is the Realization

    For an individual product, realization is usually the product itself. For a portfolio, realization is largely its composition.

    The watches are individual objects, but the collection exists through which watches are included, what roles they play and how they relate to one another. This means that the portfolio can change even when none of its members changes.

    A watch may be added. Another may be removed. A previously central watch may become secondary because the owner’s circumstances have changed. A new watch may make an existing one redundant. An inherited watch may enter the collection for reasons that have nothing to do with functional coverage but still become highly significant.

    Portfolio development is therefore not the same as developing the individual subjects within it. It is the intentional evolution of composition.

    That evolution can involve adding, retaining, repositioning, replacing or retiring members. The important question is not whether the collection is changing, but whether the change makes it a better realization of its intent.

    Assessment Changes Too

    If composition is the realization, portfolio assessment must also operate at the level of composition. Assessing the individual watches tells us whether those watches satisfy the expectations placed upon them. Assessing the collection tells us whether we have the right combination.

    The questions therefore become different. Where does the portfolio provide strong support for its intended purposes? Where is that support weak? Which contributions are unique? Which are heavily duplicated? Does apparent variety represent meaningful variety, or are several members different mainly in appearance while performing essentially the same role? Which important purposes depend on a single member? Which members remain even though nobody can clearly explain what they contribute?

    These conclusions cannot be produced simply by averaging individual assessments. Every watch could assess well on its own while the collection still assesses poorly. The whole must be assessed as the whole.

    Emergent Portfolio Characteristics

    This is where the idea becomes more interesting. A portfolio has characteristics that cannot sensibly be assigned to any one member.

    Coverage is one example. It describes how much of the intended domain the portfolio collectively supports.

    Redundancy is another. It arises when several members provide similar contributions.

    Concentration describes how dependent the portfolio is on a small number of members.

    Complementarity concerns the extent to which members strengthen the usefulness of one another.

    Resilience concerns whether the portfolio can continue to satisfy important needs when one member disappears.

    Diversity concerns whether differences between members are actually meaningful against the portfolio’s intent.

    These characteristics emerge from relationships and composition. They are not hidden properties of the individual members waiting to be discovered. That is why individual assessment is necessary but insufficient.

    Portfolio Risk

    The same applies to risk. Each watch can have risks of its own. It may be fragile, difficult to service, unsuitable for water or expensive to replace.

    The collection introduces other risks. An important situation may depend entirely on one watch. Resources may be concentrated in alternatives that add little additional value. Several apparently different watches may depend on the same underlying technology or share the same weakness. The collection may continue to reflect needs that were important years ago but no longer matter.

    There is also a more subtle risk. The owner may gradually forget why particular members are there at all. At that point, the existing composition begins to acquire authority simply because it exists.

    When the Portfolio Becomes Its Own Reference

    This is a familiar danger in another form. Once a collection has existed for a while, it becomes tempting to develop it relative to itself.

    I already have this kind of watch, so the next one should be different.

    I do not own that category, so perhaps I should.

    Serious collectors seem to have one of those, so maybe mine is incomplete without one.

    The reasoning has quietly reversed. Instead of asking what collection is appropriate for the owner’s life and interests, the owner starts asking how the current collection compares with a conventional idea of what a collection ought to contain. The realization has started replacing the reference.

    A Portfolio Reference Model provides a way out of that loop. It creates an independent basis against which both the existing collection and possible changes can be judged.

    The Portfolio Can Outlive Its Reasoning

    This is also where continuity enters the picture. Portfolios rarely appear fully formed. They develop over time.

    A product is launched for a market opportunity. Another arrives through acquisition. A system survives a migration. A capability is created for a strategic initiative. A watch is inherited. Another is bought because its owner simply loves it. The members remain, but the reasoning behind their presence can gradually disappear.

    Eventually the organization, or the collector, can describe what is in the portfolio without being able to explain why the portfolio has that shape. That loss of understanding matters because composition decisions depend on context.

    Without preserved reasoning, something may remain because nobody is confident enough to remove it. Something new may be rejected because it does not resemble what is already there. A historical accident may become a requirement. A member may continue occupying an important place long after the purpose that justified it has disappeared.

    Portfolio Continuity would preserve the understanding needed to avoid that. Not merely what the portfolio contains, but why it contains those things, what they are expected to contribute and under which circumstances that composition should change.

    Beyond Watches

    Watches make this easy to see because the stakes are low. The same structure appears in much more consequential domains.

    A product portfolio is not simply a collection of individually successful products. Two excellent products may compete unnecessarily with one another, while an apparently minor product may occupy a strategically important niche.

    A capability portfolio is not simply a collection of mature capabilities. An organization may be exceptionally strong in several areas and still be constrained by one capability it does not possess.

    A systems portfolio can contain individually stable systems while being collectively expensive, redundant, fragile or strategically incoherent.

    In each case, the members matter. But composition matters too.

    Portfolio Continuity

    This suggests a possible specialization of the broader Continuity idea. If Product Continuity preserves the understanding needed to develop and assess a product, and Capability Continuity preserves the understanding needed to develop and assess an organizational capability, Portfolio Continuity may preserve the understanding needed to intentionally develop and assess the composition of a portfolio over time.

    The structure would be familiar. Portfolio intent establishes what the portfolio is meant to accomplish. A Portfolio Reference Model describes the significant characteristics expected of the portfolio as a whole. Portfolio Development changes its composition. The realized portfolio consists of the current members and the relationships between them. Portfolio Assessment determines how well that realized composition satisfies the reference. Learning then informs what should be added, retained, repositioned, replaced or retired.

    Portfolio Management may therefore be understood, at least in part, as maintaining continuity through that cycle.

    Reflection

    There is a deceptively simple lesson in all of this. When several things are assembled to serve a larger purpose, it is natural to keep looking at the things themselves.

    Are these good products?

    Are these strong capabilities?

    Are these reliable systems?

    Are these good watches?

    Those are all useful questions.

    But once the boundary expands around them, another subject appears. That subject has characteristics the individual members do not have, risks that do not exist at member level and development actions that can change the whole without changing any of its parts.

    At that point, the portfolio question is no longer:

    Are these good things?

    It is:

    Do we have the right ones?

    And answering that requires a reference for the whole.

  • Practice Note: From Experience to Pattern

    Professional experience accumulates as cases. A difficult transformation. A successful assessment. A process improvement that changed the process without improving the underlying capability. A framework that gradually became more important than the purpose it was introduced to serve. Over a long career, practitioners encounter hundreds of situations like these.

    The value of that experience, however, does not lie only in remembering what happened. Its greater value may lie in recognizing what different situations have in common.

    Experience Is More Than a Collection of Stories

    Individual cases are always contextual. The organization was different, the people were different, the technology was different, the constraints were different, and the intervention was different. This makes experience difficult to transfer directly.

    A practitioner may say, “We tried that twenty years ago and it didn’t work.” That may be a useful warning, but it is not yet reusable professional knowledge. The more interesting questions are why it did not work, which conditions mattered, which mechanisms were involved, and whether those mechanisms are present in the situation being considered now.

    Professional experience becomes more transferable when practitioners move from remembering cases to recognizing patterns.

    Consider three apparently different situations. One organization assesses its testing capability against its existing testing process. Another evaluates a transformation according to how completely an external framework has been implemented. A third replaces a legacy product but treats the existing product as the implicit definition of what its successor should become.

    The subjects and circumstances are different, yet the same underlying mechanism may be present. An explicit reference is missing or insufficiently developed, so an available object gradually becomes the reference.

    The individual cases are useful. The recurring pattern is more powerful because it may allow us to recognize the same mechanism somewhere we have never encountered before.

    Pattern Recognition Is a Form of Expertise

    Experience changes professional reasoning. A less experienced practitioner may encounter each situation primarily as a new problem, while an experienced practitioner may recognize similarities with situations encountered previously.

    That does not mean the experienced practitioner already knows the answer. It means experience can provide better hypotheses.

    Someone who has repeatedly encountered reference substitution may become suspicious when framework compliance is discussed as though it were synonymous with organizational effectiveness. Someone who has seen organizational rituals outlive their purpose may ask why a governance mechanism exists before recommending that it be preserved. Someone familiar with the Local Hero pattern may investigate whether apparently strong organizational performance actually depends disproportionately on a few exceptional individuals.

    Pattern recognition directs attention. It suggests questions and identifies potentially important evidence. It can tell us where to look, but it cannot tell us what we will find.

    Recognition Is Not Proof

    This distinction becomes increasingly important with experience. A familiar situation can tempt a practitioner to conclude, “I have seen this before. I know what the problem is.” But similarity does not establish equivalence. The current situation may involve different causes, relationships, constraints or consequences.

    A more defensible conclusion is: “I have seen this pattern before. It gives me a hypothesis worth investigating.”

    Experience therefore complements evidence rather than replacing it. In fact, the more patterns a practitioner recognizes, the more disciplined the distinction between recognition and judgment needs to become.

    Perhaps this is one of the less obvious characteristics of professional maturity. Experience allows us to form hypotheses earlier. Maturity reminds us that we still have to test them.

    Different Cases Reveal the Pattern

    One example can illustrate an idea. Several substantially different examples can do something more: they can help reveal the abstraction.

    If the same proposed mechanism appears in a testing organization, a transformation programme and a product replacement, the differences between those cases help separate what is incidental from what may be fundamental. Apparently redundant examples can therefore be valuable when they demonstrate the same proposition under meaningfully different conditions.

    Examples from other practitioners are particularly useful. Someone may recognize a pattern and respond, “I have seen this too,” then contribute a case from a completely different environment. Another practitioner may respond that their situation looked similar but was caused by something else. Both responses improve the idea.

    The objective is not to accumulate anecdotes that appear to prove a theory. New cases should be capable of corroborating a pattern, revealing its boundaries or challenging it altogether. A useful professional pattern must remain open to all three.

    This also makes examples more than illustrations of theory. They become tests of it. Does the proposed mechanism explain substantially different situations without forcing them to fit? Does it help distinguish superficially similar situations that actually have different causes? Does recognizing the pattern suggest better questions or useful evidence? Where does the explanation stop working?

    A pattern becomes more interesting when it continues to explain different observations while remaining capable of being challenged by observations that do not fit.

    Naming What We Repeatedly See

    Some recurring patterns are particularly useful because they describe ways in which apparently reasonable organizational behaviour produces undesirable consequences.

    The process becomes the capability. The framework becomes the target. Everyone is responsible, so coherence becomes ownerless. The assessment becomes the roadmap. The Local Hero compensates for missing organizational capability. The current realization becomes the reference for its own assessment.

    Giving these patterns names makes them easier to recognize and discuss. The name does not establish that the pattern is present. It gives practitioners a shared concept with which to investigate whether it might be.

    This is an important distinction. A named pattern should improve our questions, not replace them.

    From Personal Experience to Shared Professional Knowledge

    This suggests a deliberate way of treating accumulated professional experience. Preserve enough of an interesting case to understand its context and reasoning, then ask what mechanism it may reveal. Compare it with other cases, abstract cautiously and expose the proposed pattern to other practitioners.

    Their experiences can then challenge the abstraction. Some cases may strengthen it. Others may reveal its boundaries. Counterexamples may force it to be revised or abandoned. What began as personal experience can gradually become shared professional knowledge without pretending that experience itself constitutes proof.

    Over time, something that began as “something interesting happened on one of my assignments” may become “we repeatedly observe this mechanism under these conditions.”

    That is a substantial transformation.

    A useful working principle is:

    Experience provides cases. Reflection reveals patterns. Evidence challenges them. Shared experience makes them transferable.

    The objective is not to turn every professional story into a universal rule. It is to prevent valuable experience from remaining trapped inside the people who accumulated it.

    When practitioners can preserve, recognize, articulate and challenge the patterns behind their experience, thirty years of professional work becomes more than thirty years of memories.

    It becomes something the next practitioner can build upon.

  • Practice Note: Learning to See Stories

    I want to become better at videography, and the obvious way to approach that would be to start learning video: camera settings, frame rates, codecs, Log profiles, color grading, stabilization, camera movement and editing. There is certainly no shortage of technical subjects to explore.

    The more I think about it, however, the more I suspect that starting there would teach me the wrong thing first. Before learning how to make better video, I need to learn how to see a story.

    Modern cameras already make technically good images remarkably easy. My Canon EOS R can produce beautiful photographs and perfectly respectable video. My iPhone can record excellent footage with almost no preparation, while a GoPro can capture perspectives that would have required specialised equipment not very long ago. I therefore don’t think the technology is currently my biggest limitation. The more difficult question is much simpler: what should I point the camera at, and why?

    That question is equally relevant to photography. It is quite easy to return from a trip with dozens of attractive photographs of buildings, landscapes, food and interesting objects. Each image may be perfectly good in isolation, yet the collection does not necessarily communicate what happened, what we experienced or why a particular moment mattered. A beautiful photograph records something worth seeing. A visual story has to do more: it needs to help someone understand an experience.

    For now, I have reduced that problem to three simple questions.

    Where are we? What are we doing? What makes this moment special?

    The first question establishes the environment. A wider image gives the viewer enough context to understand where the story takes place. The second moves closer and shows the activity or interaction that gives the scene its purpose. The third searches for the detail that gives the moment its character: an expression, a gesture, a texture, steam rising from a cup, a hand adjusting a camera, or something else that might easily have gone unnoticed.

    Conveniently, these three questions also correspond reasonably well with the way I already think about focal lengths. A wide-angle view can establish the context, a normal focal length can show the activity, and a telephoto or macro view can isolate the meaningful detail. It is certainly not a complete theory of visual storytelling, nor is it intended to be. Its value is precisely that it is simple enough to remember while actually taking photographs.

    Imagine, for example, that my partner and I arrive somewhere with a beautiful landscape. My natural photographic instinct might be to find the strongest composition and make the best landscape photograph I can. There is nothing wrong with that, but if I want to communicate the experience of being there, the landscape itself is only the beginning. One photograph could establish the vista. Another could show Elena enjoying the view. A third might show her photographing it, perhaps followed by a detail of the resulting photograph on the back of her camera. None of those additional photographs needs to be as visually impressive as the landscape itself. Together, however, they communicate something the landscape photograph cannot: the experience of being there.

    The same idea works on a much smaller scale. In a café, I might begin with an image that establishes the place, then photograph the preparation of the coffee, and finally notice the steam rising from the cup or the texture of the foam. Instead of collecting several unrelated attractive images, I am deliberately looking for images that have a relationship with one another.

    This led me to a slightly unusual conclusion. Perhaps the best way for me to begin learning videography is not with video at all, but with photography.

    Photography slows the problem down. I can stop and think about what establishes the scene, what activity matters and which detail completes the sequence. I do not yet have to worry about camera movement, clip duration, stabilisation, sound or editing rhythm. I can concentrate entirely on recognising the story and expressing it through a small sequence of images. My initial exercise can therefore be very simple: whenever something seems worth recording, try to tell its story in at least three photographs—introduction, activity and detail.

    Once that way of seeing becomes natural, the transition to video should be much less dramatic. Instead of taking the photograph that establishes the location, I can record a short establishing clip. Instead of photographing an activity, I can allow it to unfold for several seconds. Instead of isolating a detail in a still photograph, I can give the viewer time to observe it. The medium changes, and with it many technical possibilities, but the basic storytelling structure remains remarkably similar.

    Thinking about videography this way also changes how I look at my equipment. My EOS R is clearly the more capable photographic tool when I want maximum image quality, control over depth of field, different lenses or simply the deliberate experience of using a dedicated camera. That does not automatically make it the best tool for learning video. Using it for video introduces additional decisions about lenses, the 4K crop, resolution, support and eventually perhaps a larger rig. None of those questions is unreasonable, but none of them helps me recognise a story either.

    For the foreseeable future, the iPhone therefore makes more sense as my primary video camera. It removes much of that friction, is always available and quick to deploy, and its different focal lengths already provide something close to the wide, normal and telephoto perspectives I use when thinking about photographic stories. The GoPro adds another useful capability: viewpoints. It can go on a bicycle, be attached to something, sit very low to the ground or capture immersive walking and driving sequences without requiring a large support system.

    This means I no longer need to decide which of these cameras is somehow the best. They can simply have different roles. The Canon remains my deliberate photographic camera, particularly when image quality, lenses, portraits or the pleasure of photography itself matters. The iPhone becomes the everyday storytelling camera and, at least initially, my primary video camera. The GoPro provides movement, immersion and viewpoints that would otherwise be inconvenient or impossible. They complement one another rather than compete.

    There is also a broader lesson in this that I recognise from other areas. When learning something new, it is tempting to begin with the most visible tools and techniques of the discipline. Videography appears to be about cameras, codecs, camera movement and editing because those are the things we can observe videographers using. Yet those techniques are supporting a more fundamental capability. In my case, that is the ability to recognise a story and decide what someone needs to see in order to experience it.

    That is the capability I want to develop first. I want to become better at recognising the context that establishes a story, the activity that moves it forward and the detail that gives it meaning. Photography gives me a deliberately simpler environment in which to practise that skill. Once I can reliably tell a small story with three photographs, I can start replacing those photographs with moving images and gradually introduce everything else that videography makes possible.

    The technology can become more sophisticated as my needs become more sophisticated. The EOS R can always return to the video workflow if it eventually provides a creative advantage that I actually understand and need. There is no reason to introduce that complexity merely because the camera is capable of it.

    For now, I would much rather make a technically ordinary video that tells an engaging story than a technically beautiful video that says very little.

    Perhaps learning videography does not begin with learning video after all. Perhaps it begins with learning to see.