A method by Andrew Blount

TALR

Think. Act. Learn. Repeat.

A general method for solving problems, building projects, making better decisions, and improving anything that can learn from reality.

You do not need perfect knowledge before you begin. You need a clear objective, the best next action available, honest feedback, and the discipline to begin again with what you learned.

Apply TALRExplore The Plan
Think
Clearly
Act
Deliberately
Learn
Honestly
Repeat
Intelligently

01

The origin of TALR

The method came before the name.

I discovered the pattern behind TALR when I began writing computer software at six years old. A program would not work. I would think about why, change something, run it again, study what happened, and try again.

I applied the same pattern to problems, projects, businesses, inventions, relationships, decisions, and life. In 2012, I formalised it as Think, Act, Learn, Repeat.

Think clearly. Act deliberately. Learn honestly. Repeat intelligently.A simple idea with a demanding standard

02

The method

Four steps, run in order, then run again.

The power of TALR is not in any one step. The power is in the loop.

Think

Understand the current reality. Define the objective. Separate what you know from what you assume. Identify the constraints. Then choose the best next action available with the information you have now.

What is the best next action I can take now?

Thinking does not require certainty. It requires enough clarity to make a reasoned decision. A useful Think step should produce at least one of: a decision, a hypothesis, a priority, a test, or a next action.

Thinking without an output can become avoidance.

Act

Put the thought into reality. Run the test. Build the first version. Make the call. Have the conversation. Take the step.

What action will create progress, information, or both?

Action matters because reality contains information that thought alone cannot provide. Prefer the smallest meaningful action that can create progress or create useful information. A small action with clear learning value is often better than a large action built on untested assumptions.

Action without thought can create noise. Thought without action cannot test itself.

Learn

Observe what actually happened. Compare the result with what you expected. Measure the difference. Identify which assumption was supported, weakened, or disproved.

What do I know now that I did not know before?

Learning is not merely noticing success or failure. Learning requires an update: evidence plus honest comparison, producing a changed or strengthened model.

If the evidence changes but the model never changes, learning did not occur.

Repeat

Begin again using the improved model. Think again. Choose the next action. Act again. Learn again.

What is the next cycle?

Repeat is not doing the same thing indefinitely. Repeat means running the method again with better information. Continue until the objective is achieved, changed for a valid reason, or rejected because reality shows it should not be pursued.

The next Think should be better, because the previous cycle produced evidence.

The loop, stated formally

  • an = Think(Sn, O, Mn)the action chosen from the current state, the objective and the model
  • Sn+1 = Act(Sn, an)the action changes the state and produces evidence
  • Mn+1 = Learn(Mn, En, Sn+1)the evidence changes the model
  • n ← n + 1and the process repeats, from a better model

03

Why TALR works

It does not depend on the first decision being right.

Most meaningful problems cannot be solved perfectly in advance. Information is incomplete, reality is complex, conditions change, and actions create consequences that cannot always be predicted. TALR depends on five things instead.

  1. 1

    A current state can be observed.

  2. 2

    An objective or direction can be defined.

  3. 3

    An action can be taken.

  4. 4

    The result can be observed.

  5. 5

    What is learned can change the next action.

When those conditions hold, TALR can be applied to almost every solvable problem, buildable project, improvable system, and intentional part of life. TALR does not guarantee a particular result. It creates a disciplined way to improve the next decision using evidence from the last one.

What TALR is not

Not random trial and error. Each action should follow a reasoned model or hypothesis.
Not endless planning. Thinking must produce action.
Not blind persistence. Learning may require changing the method, the plan, or the objective.
Not defending the first idea. The model must remain open to correction.
Not a one-time sequence. Its value comes from repeated, evidence-informed cycles.

04

Apply TALR

Solve. Build. Live.

The same method operates at different scales. Start with one problem. Then apply it to a project. Then apply it to the largest and longest project you will ever have.

Solve

Applying TALR to a Problem

Reduce uncertainty, test causes, and close the gap between the current state and the desired state.

Build

Applying TALR to a Project

Hold the objective steady while evidence continuously improves the path.

Live

Applying TALR to Your Entire Life

Give TALR a life-level direction through The Plan: Believe. Read. Walk.

05

Solve · a problem

The smallest complete application.

A problem exists when the current state differs from the desired state. Problem = Desired State − Current State.

Think. Define the problem precisely. What is happening? What should be happening? What evidence do I have? What am I assuming? What are the most plausible causes? What is the smallest useful test?

Act. Run the test. Change one meaningful variable where you can. Take an action that can either solve the problem or reduce uncertainty.

Learn. Compare expected with observed. Did the action work? What cause became more or less likely? What new constraint appeared?

Repeat. Update the problem definition, select the next action, and continue until the gap is closed, accepted, or redefined for a valid reason.

Worked example

A computer cannot connect to the internet.

  1. Think The cause could be the computer, the wireless network, the router, or the internet service.
  2. Act Test a second device on the same network.
  3. Learn The second device connects. The network, router and service are now less likely to be the cause.
  4. Repeat Focus the next cycle on the original computer.

The first action did not solve the problem. It still created valuable information. Each cycle reduces uncertainty.

06

Build · a project

A stable objective, adaptive execution.

A project is a collection of connected problems organised around an objective. It needs both stability and adaptation. Before the first cycle, identify the intended outcome, the current state, the success measures, the constraints, the known risks, the most important assumptions, and the next decision that would reduce uncertainty.

Think. What are we trying to create? Where are we now? What matters most at this stage? What is blocking progress? Which assumption creates the greatest risk?

Act. Build the prototype. Run the experiment. Interview the customer. Write the proposal. Complete the next meaningful unit of work.

Learn. Did the action move the project toward the objective? What worked, what failed, what surprised us? Which assumption changed? What should be preserved, stopped, or tried next?

Repeat. Keep the objective while it remains valid. Change the path when the evidence requires it.

A strong project is not one that never changes. A strong project learns without losing its purpose.

Stopping conditions

A project cycle should continue until one of these is true:

  1. The objective is achieved.
  2. The objective is no longer worth pursuing.
  3. A constraint makes the objective infeasible.
  4. A better objective replaces it.
  5. The next cycle costs more than the progress or information it can produce.

07

Live · an entire life

A method needs a direction.

Applying TALR to a problem requires a desired state. Applying it to a project requires an objective. Applying it to an entire life requires something deeper. Before optimising a life, you must ask: what is true? What is good? What is worth pursuing? What kind of person should I become? How should I treat other people? What should remain fixed while my understanding and methods continue to improve?

TALR can help you move toward an objective. TALR cannot, by itself, decide the ultimate objective. A powerful method pointed in the wrong direction can produce the wrong result more efficiently.

The life application therefore requires both a direction and a method. The direction is called The Plan. The method remains TALR.

08

The Plan

Believe. Read. Walk.

The Plan establishes the direction. TALR provides the method for taking the next step.

Believe

Believe that truth is real and worth seeking. Believe that good is real and worth choosing. Believe in God. Believe in Jesus Christ. Believe in the Gospel. Believe enough to begin.

Belief is not the end of thought. It is the orientation that begins the search.

Read

Read the Catholic Bible. Read it honestly. Read it repeatedly. Read to understand rather than merely to confirm what you already think.

Allow what you read to test your beliefs, correct your errors, deepen your understanding, and change your actions.

Walk

Live what you have come to believe. Walk as Jesus walked. Turn belief into action. Practise truth, love, justice, mercy, humility, repentance, forgiveness, courage, and service.

Your walk is not merely what you say you believe. Your walk is how you live.

Believe Read [ Think → Act → Learn → Repeat ] ∞

Think: given what I believe, what I have read, and the situation in front of me, what is the best action I can take now? Act: take it. This is the Walk. Learn: was it truthful, loving, just, merciful, helpful? What fruit did it produce? Repeat: read again, think again, act again, walk again. Continue for as long as you live.

09

Supporting The Plan

A plan meant to guide a life should be open to examination.

The supporting resources test The Plan from three different directions.

Source

The Catholic Bible

Tests whether Believe, Read and Walk are supported by the complete Catholic Scriptures. Believe: Mark 1:15, Mark 10:18, John 20:29–31, Hebrews 11:1–6, Romans 12:2, 9, 21. Read: Deuteronomy 17:18–20, Revelation 1:3, 1 Timothy 4:13–16, Acts 17:11–12, Sirach 6:32–37, Wisdom 6:11–12. Walk: 1 John 2:3–6, James 1:21–25, John 13:12–17, 34–35, Micah 6:8, Ephesians 5:1–2, Sirach 15:14–17.

A compact rule of life assembled from a coherent scriptural sequence. The arrows represent prescribed progression, not automatic causation.

Translation test

The New International Version

Tests whether the structure survives a different English translation. Believe remains explicit in Mark 1:15. Read remains explicit in Revelation 1:3. Examine is reinforced in Acts 17:11. Act is explicit in James 1:22. Good and walk appear together in Micah 6:8. The NIV renders 1 John 2:6 as living as Jesus did rather than using the word walk, and the meaning holds.

Logical structure: pass. Translation robustness: pass. Same canon: no. The standard NIV has a different canon, so it tests robustness without replacing the complete Catholic source.

Formal test

The mathematical proof

Makes the structure explicit enough to test for consistency and missing assumptions. With B, R and W as believing, reading with understanding, and walking accordingly, completion is B ∧ R ∧ W. Each component is necessary by definition, and the sequence is procedural (B ▹ R ▹ W) rather than causal. It does not claim that B implies R, or that R implies W.

The Plan is logically consistent when its premises and terms are defined explicitly. Mathematics can prove internal consistency. It cannot prove the theological premises.

10

Apply TALR

The universal worksheet.

Use it for a problem, a project, a decision, an experiment, or a life question. Nothing is sent anywhere and no account is required. Fill it in, then export it.

TALR System

Turn any task into a learning loop.

Local, private, and deliberately simple. Your work never leaves this browser.

The loop is ready when you are.

Start with a real objective, not a perfect plan.

One-cycle worksheet

Need one pass, not a full system?

Use the original worksheet for a single decision or exportable snapshot.

0 of 9 answered

Nothing you type here is sent anywhere. It stays in this browser until you export it.

Start the next cycle

You do not need to know every step. You need to identify the next one.

Think clearly. Act deliberately. Learn honestly. Repeat intelligently.

Open the worksheetWork with Andrew