DEV Community: Extra Long Division

When to Use an LLM

Extra Long Division — Wed, 23 Sep 2026 04:20:53 +0000

This article was original posted on extralongdivision.com

Introduction

The following is a rule I’m placing on myself. I’m posting it publicly partly to stay accountable, but also because I suspect other people might benefit from following this rule. You, of course, are free to ignore my self-imposed restrictions or modify them however you like :).

“Oranges are Poisonous”

You’re immediate reaction should’ve been “no they’re not.” Because they aren’t. I first heard this counter example many years ago from a foraging guide: “If someone told you an orange was poisonous, you’d say ‘No, they’re not!’ That’s how confident you should be before eating wild plants.”

I have forgotten nearly everything I learned from that wise guide. Somehow, my mind summoned this idiom while reading a blog post that made it to the front page of Hacker News last week titled How To Write With An LLM.¹ A premise of the post is to not trust what AI² says at face value. Which lead me to ask: would I believe a chatbot if it told me an orange is poisonous? If amaranth is poisonous? Chickweed? Oleander?

One of those plants will actually kill you³. If you’re like me, you wouldn’t know which one without research. Hopefully, you’d verify an LLM’s response for something life or death like poison. But you may unwittingly accept a false and harmful response to an uninformed question.

This led me to another thought: what would I believe without questioning? I will shorten extensive thought experiments about what is truth and simply say: I’d believe many things I don’t have some expertise in.

I’ve never used the foraging guide’s advice in nature, but I will now with LLMs. I’ve already stated my hesitation regarding AI use, but I’ll add another rule. I will not use AI on a subject I can’t form confident rebuttals for. In nature, the foraging guide’s rule would prevent me from accidentally eating something poisonous; my corollary will prevent me from believing hallucinated misinformation.

This will make me a lot slower than people without this restriction. My handwritten code will ship much slower than a vibe-coded app. AI-generated blog posts will publish much faster than mine. I don’t care. The speed and ease of AI output does not outweigh the risk of believing an orange is poisonous.

Why are there footnotes?

large language model ↩︎
artificial intelligence ↩︎
oleander is poisonous ↩︎

Keep Your Skills Sharp

Extra Long Division — Mon, 14 Sep 2026 16:00:00 +0000

This article was originally posted on extralongdivision.com.

The title is not a command, but a desperate plea. Because soon, those AI¹ tokens you’re maxxing will be too expensive to rely on.

If you use AI to brainstorm, program, copy edit, whatever it may be, I am begging you: please keep your skills sharp.

This is a selfish request. Odds are, if you’re reading this, you work or maintain something I rely on. Whether you’re the sole writer of a newsletter or one of many engineers maintaining infrastructure. When token prices increase, it is imperative that you can continue your craft without the use of AI.

Historically, it is not wise to predict technology’s future. But a recent Jim Nielson blog post put it perfectly: sometimes blogging is stating the obvious.

I’d assume it obvious that tech services get more expensive over time, but it’s worth saying anyway.

The Loss Leader business model is ubiquitous. Netflix started with dirt cheap prices to convince people to cut the cord. Now streamers cost more than cable . Social media used to be free. Now verification is a credit card tap away. Anthropic used to offer API discounts but no longer does . ChatGPT striped features from its free plan and plans to add advertisements even to paid tiers . It will only get worse. Unsurprising since Open AI spends $1.61 for every dollar it earns .

AI degrading your skills may be less obvious, but it is a clear conclusion if you look at the recent past. I.e, GPS making us worse at spatial navigation . Some of you reading this might not know a world without a map in your pocket. As someone who does, I can confidently say: AI will make you worse at whatever you use it for. Research already shows this . But not just our skills. Chatbots degrade our very ability to think critically .

I cannot tell you when AI tokens will become prohibitively expensive. But I can guarantee, eventually you will be priced out of the AI model you currently use. When that day comes, for all our sake, I hope you kept you skills sharp.

Why are there footnotes?

artificial intelligence ↩︎

Please Define Your Jargon

Extra Long Division — Fri, 04 Sep 2026 02:00:00 +0000

This post was originally published on extralongdivision.com.

Split.

If you're a dancer, you probably thought of someone, hips to the ground, with their legs parallel to each other. If you're a runner, you might have thought of your pace at each mile in a marathon. If you're a bowler, you might have thought of the dreaded 7-10 split.

Same word. Different meanings.

Language is like that. Its imprecision can leave things ambiguous. Artists can use this to great effect in novels, poems, plays, etc.

Technical Writing Cannot be Ambiguous

This is rarely intentional, but ambiguity manifests when we assume jargon we use everyday is obvious. Engineers are notorious for this. I.e, this post about making a chrome extension uses the term "MV3"¹ four times but never defines it.

The meaning of MV3 is obvious in a post about chrome extensions.

You may know exactly what MV3 means, but an article titled "5 mistakes I made building an AI chrome extension" will attract beginners with little domain space knowledge. If everyone assumes the reader knows what MV3 means, how will a novice ever learn the definition?

Just Google it.

Let me remind you that search engines personalize your results. Searching "MV3" on a seasoned developer's machine may return docs on migrating a service worker. Someone else may get the 2004 St. Louis Cardinals.

Context makes MV3's meaning clear.

The author controls the article's context, but they do not control the readers' context. When I wrote "split" as the first word of this post, I had no idea how you would perceive it. I do not know if you are a dancer or a bowler or a runner. I have no idea what experiences you have when you read my writing.

I can, however, reduce any ambiguity by defining all the jargon I use. There's a way to do this without being cumbersome.

Footnotes

Write your post as you normally would, then, for any acronym, add a footnote. Your writing tools likely have built in support.

Seasoned web developers didn't need to read the footnote to know what MV3 meant. Everyone else has already scrolled down, read the definition, and continued reading. Footnotes are friction-less to knowledge area experts and immensely helpful to everyone else.

Do us all a favor — please define your jargon.

Google Manifest Version 3 ↩