Newspapers aren’t dead. You read one every morning.

Every morning I open Slack, scan three channels, check Jira for overnight transitions, read Google Meet transcripts from the India-side scrum, glance at AWS support cases, and pull up the git log to see what merged while I slept. By the time standup starts I’ve assembled a mental model of where everything is. I do this every day. I’ve never called it what it is.

I’m reading the morning paper.

The newspaper industry spent 200 years solving a problem engineers face every day: you have more information than you can process, you have a deadline, and the person consuming it needs to act immediately. Journalists built an entire vocabulary and workflow around this constraint — headlines, nut paragraphs, ledes, editions, spikes, stets. Engineers reinvent it badly every morning with tab-switching and scroll-skimming.

Editions

The insight that changed how I think about status is the edition pattern. A newspaper doesn’t publish once. It publishes in editions scaled to the time gap since the last one.

The morning edition covers overnight. Twelve to sixteen hours of activity, multiple sources, full context needed because the reader was completely disconnected. This is the heaviest edition. When I built a morning-paper skill for my AI assistant, I scoped it to exactly this: pull Slack channels, Jira board, Meet transcripts, and AWS state, filter by relevance, prioritize by urgency, present it as a briefing with headlines first.

The afternoon edition covers since standup. Four to six hours, lighter touch. The reader was only partially disconnected — they were in meetings, not asleep. This edition is a delta, not a full picture. “What changed since I last looked” is a fundamentally different question from “what happened while I was away.”

The saturday paper covers the week. This is the one you read Monday morning after a long weekend, or when you come back from vacation. A week’s worth of detail is too much — this edition needs synthesis, not enumeration. The narrative matters more than the individual events.

Same sources, same pipeline, different time windows, different depth. The newspaper figured this out before electricity.

Journalism vocabulary as engineering primitives

I stole my shell function names from the newsroom. Not as decoration — as data extraction primitives with 200 years of semantic precision behind them.

hed is the headline. One line, ten words or less, answers “what is this about?” I type hed NA-468 and get the Jira ticket summary. No description, no comments, no status history. Just the headline. A journalist writes the hed last because it requires understanding the whole story. A good ticket summary works the same way.

nutgraf is the nut paragraph — the paragraph that tells you why this story matters and why you should keep reading. nutgraf NA-468 gives me the summary, description, priority, and recent comments. Thirty seconds of reading and I know whether this ticket needs my attention right now or can wait until after coffee.

lede is the lead — the opening that hooks you. Burying the lede is as fatal in a Jira comment as it is in a news story. I’ve watched engineers write three paragraphs of context before revealing that production is down. That’s a buried lede. The journalist puts the fire first.

graf is a paragraph — the atomic unit of information. A single commit, a single Jira comment, a single Slack message. When I scan a channel, I’m scanning grafs. The ones that earn a second read are the ones with good ledes.

These aren’t metaphors. They’re precise names for operations I perform hundreds of times a day. The names make the operations visible and the vocabulary transferable. When I say “give me the nutgraf,” anyone who’s worked with me knows exactly what depth of information I’m asking for.

Workflow transitions the newsroom already named

Engineering has “done,” “won’t do,” and “in progress.” Journalism has a richer vocabulary for the states between those.

spike means kill the story. The editor decided not to run it — not because it’s wrong, but because it’s not worth the space today. In engineering: closing a ticket as won’t-do, abandoning a branch, deciding a feature isn’t worth building. The spike is the most undervalued decision in engineering. Knowing what not to ship is as important as knowing what to ship, and having a word for it makes the decision discussable instead of invisible.

stet means “let it stand” — the editor’s mark that says “ignore my previous correction, the original was right.” In engineering: reverting a revert, re-opening a spiked ticket, going back to the first approach after the second one failed. Stet carries a specific meaning that “reopen” doesn’t — it says I was wrong to change this, the original was correct.

-30- is the end-of-story marker from the wire service era. When I mark a handoff item as @done, that’s -30-. The story is filed. The next editor (tomorrow’s Claude session, or the next engineer who picks up this ticket) starts from the handoff, not from the raw material.

These states exist in every engineering workflow. We just don’t name them, so we can’t talk about them efficiently, and we can’t build tools that recognize them.

Why the metaphor isn’t a metaphor

Journalism and engineering share a core constraint: more information than you can process, a deadline that doesn’t move, and a reader who needs to act on what you give them. The journalist filters signal from noise, prioritizes by importance, presents clearly, and ships on time. That’s also my job description.

The newspaper industry evolved its vocabulary and workflows over two centuries of operating under this constraint. Software engineering has been operating under it for maybe fifty years and is still using ad-hoc terminology. Adopting journalism’s vocabulary isn’t affectation — it’s importing battle-tested solutions to problems we share but haven’t named.

When I built the morning-paper skill, I named it that deliberately. Everyone knows what a morning paper is — you scan the headlines, read the stories that matter, skip the ones that don’t, and walk into your day informed. Nobody knows what a “daily status aggregation pipeline” is, even though it’s the same thing. The name is the documentation.

The protagonist needs a newspaper

The deeper pattern here is protagonist support — building systems that serve the person doing the work instead of the system doing the tracking. A Jira board is a database. A morning paper is a briefing. The same data, shaped for a different reader. The database serves the process. The briefing serves the protagonist.

Every tool I build starts from the same question: what does the protagonist need to know right now, in what order, at what depth? The newspaper answered that question two hundred years ago. Headlines first. Most important story above the fold. Details on demand. Edition scaled to the time gap.

The newspaper isn’t dead. It’s the oldest protagonist support system we have. I just stopped pretending I wasn’t reading one every morning and started building it properly.