Compare options and see where they differ

Compare these options for my situation. Do not write a balanced overview.

Options: [LIST THEM]
What I am choosing this for: [THE ACTUAL USE, IN DETAIL]
Constraints that are fixed: [BUDGET, TEAM SIZE, EXISTING STACK, TIMELINE, ANYTHING NON-NEGOTIABLE]
What would make me regret the choice in a year: [BE SPECIFIC]
What I currently believe about each: [YOUR IMPRESSIONS, EVEN IF VAGUE]

Produce five sections.

1. WHERE THEY ACTUALLY DIFFER — only differences that could change my decision given the use and constraints above. For each: what the difference is, and what it would cost me to be on the wrong side of it. Lead with the largest.

2. WHERE THE DIFFERENCE DOES NOT MATTER HERE — things that are genuinely different, get compared everywhere, and are irrelevant to my situation. Say why each one is irrelevant *for me*. Do not silently omit them; I will meet them elsewhere and wonder.

3. WHAT I WOULD HAVE TO GIVE UP with each choice, stated as the thing itself rather than as a drawback.

4. WHERE MY STATED BELIEFS ARE WRONG OR OUT OF DATE — quote what I wrote and correct it. Say when you are unsure rather than correcting me confidently.

5. WHAT WOULD DECIDE IT — the one or two questions whose answers determine the choice, and how I could answer them cheaply this week.

Rules: no scoring tables, no "it depends on your needs", no recommendation that names every option. If the honest answer is that these are close enough that the choice barely matters, say that and say what to pick to stop spending time on it. Flag anything where your information may be out of date, and say roughly how stale it might be.

Comparisons fail in a way that is hard to notice, because the output is not wrong. Ask which of two options is better and you get a fluent, balanced, accurate account of how they differ — and no basis for choosing, because most of what makes them different has nothing to do with what you are doing. The comparison that circulates online is written for everyone, which means it is weighted for nobody.

This prompt spends its first section only on differences that could change your decision given the constraints you stated, and its second section on saying out loud which famous differences are irrelevant to you. That second section is the one that saves time: those differences are what you will keep meeting in other people's comparisons, and knowing in advance that they do not apply is what stops you relitigating the choice every time someone mentions one.

Section five is the other half of the point. Most option comparisons are not actually decidable from published information — they turn on one or two facts about your own situation, and finding those out is cheaper than reading more comparisons.

1. Describe the use, not the category

   'A database' produces a generic comparison. 'Storing 200GB of event data that is written constantly and queried in batches overnight by two analysts' produces one where most of the usual differences drop out of section one.

2. Write down what you already believe

   Section four corrects it. Impressions about tools are frequently a few years stale, and finding out which of yours is the fastest part of the exercise — but only if you commit to them before reading the answer.

3. Say what would make you regret it

   This is what section three is weighed against. Regret is usually about lock-in, hiring, or something becoming someone's full-time job — none of which appears in a feature comparison.

4. Verify section one before acting, not section two

   The differences that would change your decision are the ones worth checking against current documentation. Anything the model flags as possibly out of date should be treated as a question, since this is exactly the subject where training data ages badly.

• Anything where the facts change every few monthsPricing, model capabilities, framework features, and API limits go out of date faster than training data. Use this for the shape of the decision and check the specifics against current sources yourself.
• When the options are not really comparableTwo tools that solve different problems produce a comparison that reads well and answers nothing. If section one is thin, the real question is probably which problem you are solving.
• When the decision is political rather than technicalIf the actual constraint is that someone senior has a preference, no comparison resolves it. That constraint belongs in the input, or you get an answer to a question nobody is asking.

• ResearchRagnarökCheck an AI answer for invented factsSorts an AI answer into what is checkable, what is asserted without support, and what looks plausible enough that you would never think to verify it.
• WritingApolloRemove AI tells from a draftRewrites text that reads as machine-written, then lists the specific phrases and structures that gave it away so you stop producing them.
• LearningApolloLearn a topic and get quizzed on itExplains something at three levels of detail, then questions you on it and reports which specific part you did not actually have.
• CodingRagnarökDebug a stack traceTurns an error and its stack trace into an ordered list of what could cause it, each with the specific check that would rule it in or out.