A short piece of fiction, written by API Evangelist. The reader is imagined; the post, and every number and quotation from it, is real: "The governance layer Europe's API standards are missing," Kin Lane, November 11, 2026, https://openlint.org/blog/2026/11/11/governance-beyond-openapi/
Marit coordinates data catalogues for a regional mapping office, and she opened the post on a Wednesday afternoon because someone had forwarded it with a single line: this sounds like your problem. She expected a pitch for yet another OpenAPI linter. Her team doesn't mostly publish OpenAPI. They publish geodata services and DCAT-AP records, and the table in the post listed both. She found it oddly pleasing to see her own working week laid out in a grid, with a validator named beside every row.
The validators were the part she trusted. The post says a validator answers one question: "does this document or service conform to the standard". She knew that question well. The INSPIRE Reference Validator was open in a tab most mornings, and it had never lied to her.
Then the post turned. Governance, it said, is "everything a public body decides on top of the standard." She stopped reading and thought about the wiki page her office kept on what every catalogue record must contain. It had been written by someone who now worked elsewhere. Half of it she followed from memory. The other half she followed because Jonas, two desks over, followed it, and she had copied his habits without ever seeing a rule written down. The post put it more bluntly than she would have: those decisions live in PDFs, wikis and people's heads. She felt a small, unwelcome flush of recognition. Her office's rules were in all three.
She did not agree with all of it. The idea of one way to write down rules for every standard struck her as the kind of tidy promise that falls apart at the first awkward profile. Her records had local quirks, and she doubted a shared baseline would survive contact with them. She also noticed that the examples were mostly hopeful. Spectral already lints other formats, the post said, but nobody had yet written rulesets for her kind of document. She read the line saying "None of this is decided" twice. It reassured her that nobody was selling her anything. It also meant nothing here was something she could use yet.
What kept her from closing the tab was the line that these rules would sit next to the existing validators, not instead of them. That was the version she could live with. Nobody was asking her to give up the INSPIRE tool or the people who maintained it. They were asking what she checked by hand.
She had never counted. So she opened a blank note and, over twenty minutes, listed what she actually checked before a record went out. Does the title follow the naming pattern. Is the contact field filled in. Is the licence stated in the form her office prefers. Which version of the profile did this record claim. It came to eleven items, and she could not point to a document for six of them.
She did not feel inspired, exactly. She felt mildly embarrassed, and slightly curious about whether other offices had the same six.
The post asked readers to say which standard they would want rulesets for first. Marit pasted the list into a new thread in the project's Discussions, wrote that her answer was DCAT-AP, and added a sentence saying she was sceptical that one format could hold all of it. Then she marked the following Thursday in her calendar, in case office hours turned out to be worth an hour.