Subject in the Almanac Index
Decision Making
Judgment, assumptions, evidence, tradeoffs, and choices under uncertainty.
What should healthcare remember?
A systems architect asks what care must carry forward: decisions, uncertainty, sources, and follow-up. A working hypothesis about memory, AI, and human judgment.
The work may be necessary. Do I need to do it?
I used to look at a task and think I could do it better than the other leaders around me. Sometimes I may have been right.
What scary questions reveal
Scary questions create fear. The useful part is what that fear reveals. A good scary question makes you feel a gap in your gut.
Path-Goal Theory in AI-Native World
Path-goal theory for AI-native teams: when agents can build, human leadership shifts from tasking work to clearing the path.
When building becomes cheap
AI is lowering the cost of execution. It is not lowering the cost of choosing well. That changes what founders must become good at.
Freedom Is Not Finding the Right Meaning
What is the most intelligent meaning to give to a life? Perhaps the question contains its own trap. “Intelligent” is already a meaning we have assigned.
The Atomic Thinking Protocol
The Atomic Thinking Protocol makes consequential judgment visible, testable, and correctable: a reality-contact practice for high-stakes decisions.
Beautiful Is Not the Same as Useful
In April 2013, an email from a friend led me back to his website. He had built it in 2005. It had been a beautiful design.
Ship an Imperfect Product
The 2012 note It is not perfect. It can still be improved. That is the wrong question. The better question is: when can I ship it?
The Project I Could Not Believe In
In 2010, I was asked to deliver a project I did not believe in. That created a conflict I could not solve with a project plan.
Does Your Project System Deliver Status or Evidence?
A status tells you what happened. Evidence helps you decide what should happen next.
Forward-looking first
Logic is essential. But logic has a limitation. It is exceptionally good at reasoning from what already exists.
Someone else is thinking about it
The bystander effect names something narrow.
The KNOLt Framework
The KNOLt Framework uses six movements to examine a knowledge object, its evidence, limits, and context before it becomes advice, principle, or operating rule.
Pull the Problem, Not Just the Task
Pull changed my relationship with Scrum because it moved responsibility closer to the person doing the work.
Grounded judgment
Feel the signal. Test the story. Choose in reality. From the Architect, 13 August 2026.
If it cannot make you wrong, it is not a decision
Clarity about priorities does not make the next decision easy. It makes the exposure visible. That is where motion becomes attractive.
Indispensable is a diagnosis, not a compliment
I have held a version of this as a rule for some time. If an operator cannot leave for ninety days, they may have built a demanding job for themselves.
Busy is often a priority problem
"Busy" might be the most socially acceptable answer we have invented. How are things? Busy. Why didn't this move? Too much going on. Why haven't you started?
When Not to Use Scrum
Scrum is useful when the work is uncertain and a capable team can turn short cycles into evidence.
Scrum Exposes Problems. It Does Not Solve Them.
My Scrum blog then said, "Scrum exposes problems. It does not solve them." I still think this is one of the most useful things I wrote about the framework.
The Backlog Is a Map of Uncertainty, Not a Feature List
A backlog can make a team look certain long before the team has earned certainty. The list is long. The items are estimated. Priorities are numbered.
What Does Done Mean in a Startup?
Four people can look at the same work and declare it done for four different reasons. The developer says it works. Operations says it has been released.
What Scrum Still Gets Right
I no longer see Scrum as the answer to how work should be done.
When Scrum Mattered
Scrum was once an important part of my public identity. I read it, practiced it, blogged about it, and eventually wrote a book about it.
Before the fundraise, find the evidence
Before the fundraise, find the evidence A version was published on LinkedIn on July 20, 2026.
The trade-off attached to capital
The trade-off attached to capital A version was published on LinkedIn on July 24, 2026. The export does not establish the timezone or first-publication venue.
Count barrels, not bodies
Count barrels, not bodies A version was published on LinkedIn on April 13, 2026. An earlier Journal origin is claimed in the source but remains unverified.
Pursue the business truth
Pursue the business truth A version was published as a LinkedIn article on December 13, 2019.
Your calendar is a confession
A completely full calendar looks like commitment. Sometimes, it is. Sometimes, it is the opposite.