A short piece of fiction, written by API Evangelist. The reader is imagined; the post, and every number and quotation from it, is real: "Almost ready for business: the third OpenLint office hours," Kin Lane, October 15, 2026, https://openlint.org/blog/2026/10/15/ready-for-business/
Marguerite maintained the API linting setup for her team, which meant she was the person who got the message whenever a rule fired on a pull request and someone didn't understand why. She read the OpenLint post on a Friday afternoon, with a sandwich and a half-finished migration ticket open in the other window.
The first thing she noticed was the tone. The post said the focus was to "set the table before adding anything new", and she liked that. Most projects she followed did the opposite, adding features while the foundation sagged. A rename, passing tests, a release pipeline and a docs site were dull things to be proud of, and she trusted them more for it.
Then she got to the schedule and her shoulders went up. Version 0.1.0 within days, 0.2.0 by the following week, a 1.0 in a month or two. She had seen schedules like that before. Every one of them had slipped, and she was the one who had to explain the slip to her manager. The post was honest that these were plans, and she gave it credit for that, but plans were not packages. She decided she would believe the numbers when she could install them.
The compatibility promise was the part she kept returning to. A new core ruleset and an export named Spectral, so that JavaScript users could "switch packages and nothing else". She had about forty custom rules and a handful of custom functions, written over three years by people who had since left. If the promise held, her afternoon would be a find-and-replace. If it didn't, she would be spending a sprint on something nobody had asked her for. She noted, a little grudgingly, that the post drew a line: "Breaking changes wait for 2.0." That was a better promise than most, because it told her where the risk lived.
She felt less sure about the money. A first donation of $1,000 and a plan to chase European grants sounded small next to the list of unfinished work. Then she reread it and saw that the first things to be funded were small, well-scoped pieces like the release pipeline and the docs site. Small seemed to be the point. Still, she wondered who would be paying for the language server she wanted for her editor, and the post didn't say.
The last line, that the post was drafted with Claude from the recording and edited by the author, gave her a moment of unease she couldn't quite justify. She decided it was fine. They had said so, which was more than she could say for half of what landed in her feed.
She didn't migrate anything. She didn't star the repo or write an enthusiastic message to the team channel. What she did was open a new file in her tooling repo, name it for the migration, and list every custom rule and function she owned, one per line, with a column for whether anyone still used it. Half of them, she suspected, nobody did. If 0.1.0 turned out to be real, she would at least know what she was moving.
She saved the file, finished her sandwich, and went back to the ticket.