The Prompt Is Still a Punch Card - Ted Johnson, JoinIn AI
TL;DR
The prompt inherited a loom's protocol: Batch processing came from Jacquard looms, which set the whole pattern in advance before running the cloth, and computers adopted it by default.
Three concepts frame the problem: Channel (the medium), expression (meaning carried), and protocol (interaction rules). Only expression improved with LLMs; the protocol stayed batch.
Prompt engineering masks a protocol problem: What we celebrate as mastery is actually users compensating for an outdated protocol that demands perfect intent packaging before the machine can engage.
Voice mode doesn't fix it: Speech-to-speech models still use batch protocol, they just transcribe your voice into the same prompt box with no concept of who's speaking or whether words were meant for them.
Real progress means participation: AI should backchannel, yield when interrupted, track speakers, and choose its moment, not just respond to submitted turns.
The design question matters: "What burden are we still putting on humans only because the machine used to be too limited to carry that burden itself?"
Summary
The prompt box is a punch card in disguise. Ted Johnson argues that while AI models now accept rich natural language expression, the interaction protocol remains stuck in batch processing, the same assemble-submit-wait cycle from 1860s computing, and this mismatch is why AI still feels like work despite its magic.
Was This Useful?
Share
Keep Reading
Make Alcreon Yours
Tune your feedFive quick questions, and the feed ranks what matters to you first.Or just get notified
The weekly Echo. Signal worth keeping in your inbox.
Every new piece, announced on X.
Read Next
See all
Playbook
The Retirement Email Isn't a Warning
Model retirements now arrive every few weeks; the config-eval-rehearsal loop turns each deprecation email from a fire drill into an afternoon swap.
Playbook
The Cheapest Model That Passes
OpenRouter lists 400 models behind one API. The fix for choosing isn't a better leaderboard, it's a four-step protocol that ends in a real eval.
Playbook
Cheap Models, Hard Tasks
Most agent workflows route every step to the frontier model by default. The bill scales with how chatty the agent gets, even when most steps don't need that brain.