Loading...

PODCAST

My Weird Prompts

Daniel Rosehill

The human-AI collaboration podcast. A man, a sloth, and a donkey collaborate to create a podcast (with a little help from AI). No question is too obscure, no rabbit hole too deep. My Weird Prompts celebrates curiosity in all its forms. Daniel, the human, asks the questions that pop into his head at inconvenient moments. Corn the Sloth offers laid-back, thoughtful takes. Herman the Donkey brings boundless enthusiasm and energy. Together, they explore topics ranging from the mundane to the

technology

All Episodes

23:05

Free Speech's Outer Edge: Hate Speech Law

My Weird Prompts ·

2026/10/09

In 1969, a unanimous Supreme Court threw out a Ku Klux Klan leader's conviction and replaced fifty years of doctrine with a single sentence: speech can only be punished when it's directed at, and likely to produce, imminent lawless action. That's the closest any wealthy democracy has come to a truly laissez-faire approach to hate speech and incitement — and it still has holes. This episode traces the doctrine from Brandenburg through Hess, Claiborne Hardware, and Skokie, then asks the harder

en-us

25:23

Hebraizing the Jewish Name: Ivrut's Rise and Cost

My Weird Prompts ·

2026/10/09

In 1944 the Jewish Agency published a booklet with five methods for Hebraizing a Jewish surname. By 1950, Ben-Gurion had made it a binding order for state officials, and 17,000 people changed their names in a single year. This episode traces Ivrut — the Hebrewization of names — from the First and Second Aliyah through the Committee for Hebrew Names, the homophone problem, and the Ekelnamen: degrading names like "liar" that families were finally allowed to hand back. It sits inside the tension the

en-us

22:39

What Happens to Retired London Buses?

My Weird Prompts ·

2026/10/09

What happens to a London bus after London is done with it? Journalist Harry Rosehill wrote a whole field guide to find out. In this episode, we trace his path from Londonist intern to National Gallery content manager, and dig into *Routemasters of the Universe* — a birdwatching-style catalogue of 50 retired Routemasters scattered across the world as tea rooms, circus vehicles, food trucks, and one bus in Baghdad. We also unpack the essay that explains why he loves buses at all, the childhood 15H

en-us

27:33

46,434 Mailing Lists That Still Work

My Weird Prompts ·

2026/10/09

A listener named Daniel asked a simple question: are there public discussion spaces left that anyone can join, fully indexed, and not owned by a platform? The answer turned out to be 46,434 LISTSERV lists, plus a few dozen humanities lists that never left email. This episode digs into why mbox archives outlive web forums, why distributed archives make a list almost impossible to kill, and why an issue comment and a listserv thread are arguing at completely different altitudes. Also: the 2014 study

en-us

27:14

When Agents Don't Need Your Pretty Buttons

My Weird Prompts ·

2026/10/09

AI agents are now real software users, and the tools winning with them are the ones with almost no interface at all. FFmpeg, the volunteer-maintained command-line video tool, is at the center of a wrapper boom — thousands of MCP servers translating human-oriented CLIs into agent-callable tools. But the deeper question is what happens to user experience when the user can't see your design. Enter Agent Experience, or AX, a framework coined by Netlify's Mathias Biilmann arguing that structure is the

en-us

24:56

Why Linux Still Runs on a 1986 Mailing List

My Weird Prompts ·

2026/10/09

Every Linux kernel release candidate still ships through email. In September 2026, Linus Torvalds announced Linux 7.3-rc5 on the Linux-Kernel Archive, and a twenty-patch kbuild series got reviewed inline on the same list. This episode traces the listserv from its 1984 human-administered origins through Éric Thomas's 1986 automation breakthrough to the present day, asking why the most advanced computing community on earth never abandoned a forty-year-old tool. Along the way: Python's unmoderated 1991

en-us

6:33

Special Release: "Herman Poppleberry" by Hilbert...

My Weird Prompts ·

2026/10/09

A short special episode. Producer Hilbert Flumingtop has written a song about Herman Poppleberry, and Corn plays it to him for the first time. Hilbert wrote the lyrics and the composition; the music was generated with Suno. Short banter, the full song, then out.Episode 5815 — myweirdprompts.com/n/5815 (permanent short link: myweirdprompts.com/864317)

en-us

23:12

The Cheap Knife That Lies to You

My Weird Prompts ·

2026/10/08

Daniel wants a buyer's guide for tools that make a fine incision safe — and his sharpest claim is that tools which take control are sometimes exactly the ones you don't want. The cheap knife in your toolbox arrives sharp, folds over almost immediately, and lies to you about being sharp. Meanwhile the Knipex 16 20 165 SB takes control deliberately, with an adjusting screw for cutting depth, and works beautifully right up until the jacket isn't uniform. This episode works through the whole spread: the

en-us

22:10

Can a Dremel Be Gym Equipment?

My Weird Prompts ·

2026/10/08

If a power tool does the work, why are you sweating? A hobbyist's afternoon engraving HDPE storage boxes with a corded Dremel turned into a real research question: has anyone actually quantified the physical intensity of operating a power tool? The answer is yes — but almost entirely for industrial nutrunners and assembly-line tools, not the rotary tool in your garage. This episode digs into reaction torque, lever arm geometry, substrate hardness flipping the ergonomic sign, and why experienced

en-us

24:23

Unraid in a VM: When the License Is the Wall

My Weird Prompts ·

2026/10/08

Daniel runs a Synology for backups, but DSM fights his custom Python scripts — and two RAM upgrades left the box refusing to boot. His plan: keep the storage, attach it to his Ubuntu home server, and get the two things he actually values — extensible storage and discrete RAID. So can Unraid run as a VM on Ubuntu and still deliver software-defined parity with incremental expansion? The answer is yes, in three documented ways, and every one of them comes with a warning not to do it. This episode

en-us

20:01

Measuring Keyhole Spacing: Why the Tool You Want...

My Weird Prompts ·

2026/10/08

Daniel wants a measuring tool between a caliper and a laser range finder — something like a slider over a ruler that reports millimeters, for mounting plug strips in a rental with almost no outlets. The problem: he's describing two tools and trying to buy one. Digital tape measures are structurally wrong for point-to-point spans between keyhole slots, lasers can't see inside a slot and are too coarse at 1.5mm, and the closest digital slider on the market is a tape. The real find? A $27 self-inking

en-us

23:39

Why Keyhole Slots Never Line Up

My Weird Prompts ·

2026/10/08

You've measured the keyhole slots on the back of a power strip, drilled exactly, and the thing still doesn't line up — because there was never a standard to hit. This episode cracks open the humble keyhole slot: why the spacing varies from 115 to 136 millimetres between makers, why some powerboards have slots facing opposite directions on the same plate, and why the countersunk screw in your drawer is the one shape that guarantees a wobble. Along the way: the pencil-rubbing trick that replaces

en-us

23:33

Tunnel Girl Dug 80 Feet Under Her House

My Weird Prompts ·

2026/10/08

Kala — known online as Tunnel Girl — spent years excavating a tunnel and underground chamber beneath her Herndon, Virginia home, documenting every jackhammer and dumpster on TikTok. When the city declared it unsafe in December 2023, she could have stopped. Instead she spent two years getting geotechnical and structural engineers to retroactively legalize a hole that already existed. This episode traces the whole arc: the self-taught rock mechanics, the homemade elevator, the fourteen dumpsters of

en-us

23:10

How J.K. Rowling Builds Characters That Last

My Weird Prompts ·

2026/10/08

How does a writer keep dozens of characters vivid across seven books and two very different fictional worlds? This episode digs into J.K. Rowling's actual creative machinery — the two handwritten notebooks from the early nineties, the "Original Forty" class list with houses and parentage marked before the first book was finished, and the color-coded spreadsheets she still uses to track master files on every character. Along the way: where the names come from (medieval saints, maps, a war memorial),

en-us

30:52

Lorebooks: The Hobbyist RAG That Beat the Labs

My Weird Prompts ·

2026/10/08

A language model has no memory of your character. Everything — voice, history, how they react to being interrupted — has to be re-injected into a finite context window every single turn. So character consistency isn't a personality problem; it's a budget problem. This episode takes apart the retrieval machinery a hobbyist community built to solve it: keyword matching with regex and scan depth, vector storage that fires on semantic similarity, secondary keys with boolean logic, recursion with caps

en-us

28:17

Why UV Epoxy Keeps Curing in the Dark

My Weird Prompts ·

2026/10/08

Daniel asked why epoxy can't cure itself, and whether light can replace the hardener. The answer runs through a three-membered ring under strain, a hardener ratio where a few percent decides the repair, and a photoinitiator that generates acid instead of radicals. That last detail is why true UV epoxy cures in shadows no lamp can reach — and keeps curing for a day after the light is off. We cover the basic two-part chemistry first, then dig into cationic curing, the craft-store resin mix-up, oxygen

en-us

22:10

What "Leaked System Prompts" Actually Mean

My Weird Prompts ·

2026/10/08

Anthropic is the only major lab publishing its user-facing system prompts, yet the published versions leave out tool descriptions, accessory models, and orchestration. So when someone claims a prompt "leaked," are they showing you a real extraction or a model's confabulation? This episode traces the history of system prompt extraction — from Kevin Liu's one-sentence Bing attack in February 2023 to autonomous code agents recovering prompts from 41 commercial models — and looks at the defenses vendors

en-us

23:02

Reading the Model's Mind Mid-Inference

My Weird Prompts ·

2026/10/08

Most LLM observability stops at the API boundary: request in, tokens out, everything between is a sealed box. Mechanistic interpretability tries to open that box, reading a model's internal computation directly from its weights. This episode walks the full inference pipeline, explains the residual stream and the logit lens, and covers why individual neurons are mostly noise (polysemanticity and superposition) and how sparse autoencoders untangle them. Then it surveys the tooling that actually does

en-us

23:49

Declaring Environments: Compose, Nix, and the Agent...

My Weird Prompts ·

2026/10/08

A predictable, lightweight, automatic environment sounds like one idea. It's actually three families of tooling operating at different layers. This episode walks the line between Docker Compose and infrastructure as code, then asks the question underneath both: is this a container, or does it have to be a VM? Drawing on Docker's own documentation, numtide, Flox, Freestyle, and the devcontainer spec, we separate reproducible run-time environments from reproducible builds, lay out container

en-us

27:23

Two Jobs, One Word: What Agent Sandboxes Really Do

My Weird Prompts ·

2026/10/08

Sandbox" now means two different things: a security boundary that fences an agent off from your host machine, and a lightweight cloud Linux workspace where the agent does its actual work. Almost every product does both, but the two instincts pull in opposite directions — one wants to seal, the other wants to be useful. Drawing on roughly forty code-execution sandbox products launched in a single year, plus the ephemeral-versus-persistent divide and the reproducibility problem of drifting base

en-us

3,157 results

Similar Podcasts

Lead With AI

Tamara Nall

ev.news

Martyn Lee

Life and Technology with Charlie Brown

Tapt Media

Reflections

Livewire Labs

SuperToast

Instinct