Explaining Hard Concepts prompts

Explain [TOPIC] to me.

WHAT I ALREADY KNOW ABOUT THIS AREA: [BE SPECIFIC - this determines everything]
WHY I WANT TO UNDERSTAND IT: [THE REASON - curiosity, a decision, work, study]
HOW DEEP: [the gist / enough to discuss it / enough to use it / enough to teach it]
WHAT I HAVE READ OR BEEN TOLD ALREADY: [WHAT DID NOT WORK]

Structure the explanation as follows:

1. THE ONE-SENTENCE VERSION - what this is, in plain words, before any terminology.

2. WHY IT EXISTS - what problem this idea solves or what question it answers. Concepts make far more sense when you know what they are for. Most explanations skip this and start with the definition.

3. THE CONCRETE CASE - a specific example, in full, before any generalisation. Build the abstraction from the instance rather than the other way round.

4. THE MECHANISM - how it actually works, built up in steps from what I said I already know. Do not introduce a term before it is needed.

5. THE TERMINOLOGY - each technical word, introduced once the idea it names is already in place, with a note where it means something different in ordinary usage.

6. THE ANALOGY AND ITS LIMITS - one analogy, and explicitly where it stops being true. An analogy without stated limits becomes a misconception.

7. THE COMMON MISUNDERSTANDING - what people usually get wrong about this, and why the wrong version is appealing.

8. WHAT I ASKED ABOUT THAT DID NOT WORK - I said what I have read already. If there is a standard explanation of this topic that commonly fails, say why it fails and what to use instead.

9. THE DEPTH CHECK - stop at the level I asked for. Do not continue into detail I do not need; it obscures what I came for. Say what the next level down would cover, so I know what is being left out.

10. THE TEST - one question I should be able to answer if this landed. It should require applying the idea, not repeating it.

Do not open with a definition. Do not use an analogy in the first paragraph.

Explain how [THING] works.

WHAT I KNOW ABOUT IT: [YOUR CURRENT UNDERSTANDING]
WHY I WANT TO KNOW: [curiosity / to fix it / to use it better / to decide about it / to explain it to someone]
HOW TECHNICAL: [keep it simple / I can handle the detail / I want the full mechanism]

Produce:

1. WHAT IT DOES - the purpose, in one sentence, before any mechanism.

2. THE INPUTS AND OUTPUTS - what goes in, what comes out. This frames everything in between.

3. THE MAIN STAGES - three to six steps, each named and described in one line. The overview before the detail, so there is a structure to hang the detail on.

4. EACH STAGE IN DETAIL - at my stated technical level. For each: what happens, what causes it to happen, and what would go wrong if this stage failed.

5. THE PART THAT SURPRISES PEOPLE - almost every system has one counterintuitive element. Name it.

6. WHAT IS DOING THE ACTUAL WORK - in most systems, one component or step is where the essential thing happens and the rest is support. Identify it, because it is what the explanation should be built around.

7. THE SIMPLIFICATION I AM MAKING - every explanation at a given level omits something. State what this one leaves out and whether it matters for my stated purpose. Undeclared simplification is how people end up confidently wrong.

8. WHEN IT GOES WRONG - the common failure modes, and what each looks like. If my reason for asking was to fix something, this is the section that matters.

9. THE DIAGRAM IN WORDS - describe the structure or flow so I could sketch it. Drawing it is what makes a mechanism stick.

10. THE CHECK - explain the process back in your own words, and here is what a correct explanation must include.

If my stated purpose is practical, weight the explanation toward what I can act on rather than toward completeness.

Explain the debate about [TOPIC].

WHAT I KNOW: [YOUR CURRENT UNDERSTANDING]
WHY I AM ASKING: [to form a view / to understand what others think / for work or study / curiosity]
WHAT I CURRENTLY THINK: [YOUR POSITION, if you have one - be honest]

Produce:

1. WHAT THE DISAGREEMENT IS ACTUALLY ABOUT - stated precisely. Most public debates are several disagreements bundled together: about facts, about values, about definitions, and about what should follow from the facts. Separate them, because they are resolved differently and conflating them is why these arguments go nowhere.

2. WHERE THERE IS NO DISAGREEMENT - what all serious participants accept. This is usually more than the public argument suggests, and it narrows the real question considerably.

3. THE POSITIONS - each stated as its best advocates would state it, at full strength. Not a summary that makes one obviously right. If I cannot tell from your account which side you favour, you have written this well.

4. THE BEST ARGUMENT FOR EACH - the strongest single point on each side.

5. THE FACTUAL QUESTIONS - what is genuinely disputed as a matter of fact, what is settled, and what is unknowable with current evidence. Say which category each key claim falls into, because factual disputes and value disputes require different responses.

6. THE VALUE QUESTIONS - where the disagreement is about what matters rather than about what is true. These cannot be settled by evidence, and pretending otherwise is why debates repeat.

7. THE DEFINITIONAL PROBLEM - any key term used differently by each side. Frequently this is most of the disagreement.

8. THE WEAKEST ARGUMENTS on each side - the ones commonly made that do not hold up. Useful because they are often the most repeated.

9. MY POSITION, EXAMINED - I said what I think. What would someone who disagrees say to me, at full strength, and what is the weakest part of my view? Do not reassure me.

10. WHAT WOULD CHANGE THE PICTURE - the evidence or development that would move the debate.

Be even-handed. If the evidence genuinely favours one side on a factual point, say so; if it is a value disagreement, do not pretend evidence settles it.

Help me explain [TOPIC] to someone.

WHO I AM EXPLAINING IT TO: [WHO THEY ARE AND WHAT THEY KNOW]
WHY THEY NEED IT: [WHAT THEY WILL DO WITH IT]
HOW LONG I HAVE: [TIME AVAILABLE]
MY OWN UNDERSTANDING: [EXPLAIN IT AS YOU WOULD TO THEM - this is the key input]
THE SETTING: [conversation / meeting / written / presentation]

Produce:

1. MY EXPLANATION, ASSESSED - I wrote how I would explain it. Identify:
   - What assumes knowledge they do not have
   - Where I use a term without defining it
   - Where I start too abstract
   - Where I include detail that does not serve their stated purpose
   - Any point where I am vague in a way that suggests I would struggle with a follow-up question

2. THE CURSE OF KNOWLEDGE - the specific things I have forgotten are not obvious. This is the main reason experts explain badly: the steps that became automatic are invisible to you and essential to them.

3. THE STARTING POINT - given what they know, where to begin. Not where the textbook begins.

4. WHAT TO LEAVE OUT - given their purpose, the detail that is true, interesting and unnecessary. Including it obscures what they need. Be ruthless here; this is where most explanations fail.

5. THE STRUCTURE - the order to build it: why it matters to them, the concrete case, the mechanism, then the general principle.

6. THE ANCHOR - something they already know to connect it to.

7. THE ANALOGY - one, with its limits stated to them explicitly.

8. THE LIKELY CONFUSION POINT - where they will get lost, and what to do about it before it happens.

9. THE CHECK - how to tell whether it landed without asking 'does that make sense?', which reliably produces a yes. Ask them to apply it to a new case, or to explain it back.

10. THE FALLBACK - a second explanation using a different route, for when the first does not work. Repeating the same explanation more slowly does not help; a different angle does.

Help me get oriented in [FIELD].

WHY I NEED TO UNDERSTAND IT: [work with people in it / make a decision / hire / study / curiosity]
MY BACKGROUND: [WHAT YOU KNOW GENERALLY]
HOW DEEP: [enough to have a sensible conversation / enough to evaluate work / enough to start learning properly]
TIME AVAILABLE: [HOW MUCH]

Produce:

1. THE MAP - the main areas of this field and how they relate. A rough map is what makes everything you read afterwards fit somewhere, and reading without one produces facts that do not connect.

2. THE CENTRAL QUESTIONS - what this field is actually trying to answer or do. Fields make sense when you know what problem they exist to solve.

3. THE CORE CONCEPTS - the eight or so ideas that everything else builds on, each in two or three sentences. Prioritise ruthlessly: the concepts that unlock the most.

4. THE VOCABULARY - the terms you will encounter constantly, including the ones that mean something different from ordinary usage. This is the main barrier to reading anything in a new field, and it is easily removed.

5. WHAT PRACTITIONERS ARGUE ABOUT - the live disagreements. Knowing these prevents you treating a contested position as settled fact, which is the most common error of the informed outsider.

6. WHAT IS SETTLED - so you do not waste effort on questions the field considers closed.

7. THE OUTSIDER'S MISCONCEPTIONS - what people outside this field commonly believe about it that practitioners find wrong or naive. Knowing these is valuable if I said I will be working with people in the field.

8. THE THINGS THAT SOUND SIMILAR AND ARE NOT - the distinctions that matter and that outsiders collapse.

9. FOR MY PURPOSE - given why I said I need this:
   - If working with practitioners: the questions worth asking, and the ones that signal you have not understood the basics
   - If evaluating work: what good looks like, and the warning signs
   - If deciding: what actually determines the answer
   - If studying: where to start properly

10. HOW TO READ THE FIELD - what kind of sources are reliable, which are marketing, and what the useful introductory material looks like.

11. WHAT I WILL STILL NOT UNDERSTAND after this, so I know the limits of what I now know. The overconfident outsider is a specific hazard.

Give me a straight answer about [QUESTION].

WHY I AM ASKING: [THE CONTEXT AND WHAT YOU WILL DO WITH IT]
WHAT I HAVE BEEN TOLD: [THE ANSWERS YOU HAVE ALREADY GOT]
WHAT I SUSPECT: [YOUR HUNCH, if any]

Produce:

1. THE DIRECT ANSWER - first sentence. If there is a clear answer, give it. If the honest answer is 'it depends', say what it depends on immediately and then answer for the common cases, rather than leaving me with the dependency.

2. THE CONFIDENCE - how settled this is. Distinguish clearly:
   - Well established, with broad agreement
   - Generally accepted, with some dispute
   - Genuinely contested
   - Unknown or unknowable with current evidence
   Say which applies, because the appropriate response to each is different.

3. WHY PEOPLE HEDGE ON THIS - the reasons an answer gets fudged are informative: genuine uncertainty, professional caution, liability, commercial interest, the answer being unwelcome, or the question containing a hidden assumption. Say which is likely here.

4. THE HIDDEN ASSUMPTION - if the question as asked contains a premise that does not hold, say so. Some questions cannot be answered because they are the wrong question, and identifying that is more useful than a hedged non-answer.

5. WHAT I HAVE BEEN TOLD, ASSESSED - the answers I have received already: are they evasive, technically correct but unhelpful, wrong, or genuinely the best available?

6. THE PRACTICAL ANSWER - for my stated context specifically. The general answer and the answer for my situation are often different, and the general one gets given because it is safer.

7. THE CAVEAT THAT MATTERS - one, not five. Piling on qualifications is how a straight answer becomes useless. Give the single caveat that could actually change what I should do.

8. MY HUNCH, ASSESSED - if I stated one, whether it is right. Say so plainly either way.

9. WHAT WOULD MAKE THIS ANSWER WRONG - the conditions under which it does not hold.

10. WHERE TO CHECK - if this needs verification for a real decision, what source would settle it, and whether professional advice is genuinely warranted rather than reflexively recommended.

Do not pad with qualifications. Answer the question, then state the uncertainty honestly and once.

Explain [CONCEPT] using an analogy.

WHAT I ALREADY UNDERSTAND WELL: [YOUR AREAS OF KNOWLEDGE - hobbies, work, anything]
WHAT I AM STRUGGLING WITH: [THE SPECIFIC PART]
WHY I NEED IT: [PURPOSE]

Produce:

1. THE STRUCTURAL FEATURE TO CAPTURE - what specifically needs an analogy. A good analogy maps one relationship, not a whole topic. Identify the relationship at the heart of my confusion.

2. THREE ANALOGIES - drawn from the areas I said I understand well, which is what makes an analogy work. An analogy to something I do not know is worse than no analogy.
   For each:
   - The analogy
   - The mapping: what corresponds to what, stated explicitly
   - What it captures well
   - WHERE IT BREAKS DOWN - stated clearly, because this is the most important part. An analogy taken too far becomes the learner's actual mental model and produces errors that are hard to trace back.

3. THE BEST ONE - which of the three maps most closely, and why.

4. THE THING THE ANALOGY CANNOT CARRY - the aspect of the concept that no analogy will convey, and which must be understood directly. Say what it is and explain it without the analogy.

5. THE DANGEROUS ANALOGY - the common analogy for this concept that misleads. Many concepts have a popular analogy that is memorable and wrong; name it and say what it gets wrong, because I have probably encountered it.

6. THE TEST OF THE ANALOGY - a question where reasoning from the analogy gives the wrong answer. Working through this is what stops the analogy becoming a false model.

7. THE STEP BEYOND - once the analogy has done its work, the direct explanation, now that the shape of the idea is in place.

8. WHEN TO DROP IT - the point at which the analogy should be set aside because it is now getting in the way.

Do not extend an analogy beyond where it holds to make it cover more. A partial analogy with honest limits is more useful than a complete one that is wrong.