He Wrote 0.5. The Book Printed 0.05. Merrick Osgood-Talabi Dies At 79.
His field notebook, given to this newspaper by his daughter, records the tolerance twice in his own hand. Forty books, and one of our own wire items, print a figure ten times smaller. He wrote three times asking for a correction.
SHEFFIELD, England — The notebook says 0.5.
It says it on the right-hand page of 3 September 2002, and again on the facing page of the 11th, in pencil, in the same hand, with the instrument and block number beside it both times.
The book says 0.05.
Merrick Osgood-Talabi, who took the measurement and paid for the book, died on 14 March. He was 79. The cause, according to the South Yorkshire coroner's officer, was cardiac arrest. He was on the pavement outside his flat when it happened, with a 600-millimetre builder's level in his hands, measuring the fall on a paving slab the council had relaid in February.
Three weeks and one block
He was a water engineer for thirty-four years: mains, then sewerage design, then flood risk. He retired in 2001 and flew to Bolivia in 2002 because he had watched a documentary and did not believe the arithmetic in it.
He took the level, a set of feeler gauges, a straightedge he had made himself, and a permit. He measured one H-block — number 14 of the western group — because the permit allowed him one, and because he thought a single number obtained properly was worth more than forty obtained quickly.
What he got was half a millimetre of deviation across the channel floor. It is roughly what a careful man with a spirit level should get, and it is roughly right.
What the book printed
In 2003 he had the report printed by a shop in Rotherham, in an edition of 400, and sold it at cost.
Table 4, page 61, gives the figure as 0.05mm.
That is the number quoted in forty books, in fourteen films, and — until this morning — in this newspaper. It is the number that means machined. Half a millimetre means a man with sand and patience. Five hundredths means something else, and five hundredths is what got into circulation.
Where the point went
Kenneth Ivo Ashworth-Bello set the 2003 edition and is 84. He does not remember the job and said so within a minute of picking up the telephone.
He was useful anyway. Tables, he said, were the thing you re-keyed by hand: a scanner made a mess of columns, so somebody typed them again.
"If you are asking whether a man could move a point one place in a column of forty figures," he said, "the answer is yes, and so could you, and so could the author when he checked it, which is why proofs exist."
The typescript went in the bin in 2003 when the finished copies came back. No marked proof survives. We cannot say who did it.
Three letters
In 2009 he wrote to the publisher of the book that had made the figure famous. In 2014 he wrote to a production company. In 2019 he wrote to this desk.
All three letters are published with this obituary, with his certificates of posting. None of them was answered.
They are not angry letters. Four paragraphs each, a photocopy of the notebook page enclosed, and an offer to pay for the amendment slip himself.
The block, measured again
In May this newspaper paid a metrologist $5,200 to laser-scan block 14, alongside a replica cut with period tools, for a story about how long the cutting takes.
Ingrid Saavedra-Cheung's certificate gives the original channel floor as 0.6 millimetres across 610 millimetres.
That is not 0.05. It is within a tenth of a millimetre of what a retired engineer wrote in pencil in 2002, and it took a laser scanner and a certificate to say so.
He had been dead eight weeks when that scan was taken. The story it was taken for ran on 9 August and does not mention him, because when it was written we did not know his name.
What his daughter has
Ndidi Osgood-Talabi has the notebook, which she has given us, and 240 unsold copies of the book in a wardrobe in Sheffield.
"People assume he was cross about the aliens," she said. "He wasn't. He thought that was daft and none of his business. He was cross about a decimal point. Those are two different men and only one of them gets invited on television."
What this desk records
He died on a pavement, of a heart attack, measuring a paving slab, because water was standing where the fall should have carried it away. He had complained to the council in January; the file shows the complaint, and shows the slab relaid on 20 February.
His last notebook — a smaller one, in his coat pocket — records a fall of 1 in 190 where the specification asks for 1 in 80.
He was right about that as well. Nobody has printed it before now.
He is survived by his daughter, two grandsons, 240 books in a wardrobe, and one digit that this newspaper has finally got round to moving back.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
His daughter wrote to this desk in June, while clearing her father's flat, enclosing a photocopy of a letter he had sent us in 2019 and asking whether anybody here had answered it. Nobody had. We photographed all 96 pages and had the two relevant pages read by a second person who was not told what we expected them to say. We then traced the 0.05 figure through every citation we could find in four languages, and checked it against a laser scan of the same block that this newspaper had already commissioned, in May, for something else entirely.
Who we spoke to
- Ndidi Osgood-Talabi, His daughter. Interviewed twice at her father's flat; gave this newspaper the 2002 field notebook June and July 2026 Read this piece before publication and asked for one change: that the word 'amateur' come out. It is out.
- The 2002 field notebook, 96 pages, pencil and ink, one block, three weeks. Given to this newspaper; photographed page by page and published in full with this obituary June 2026 The two relevant pages were read independently by a second person who was not told what we expected them to say.
- Kenneth Ivo Ashworth-Bello, Compositor, retired; set the 2003 edition at a shop in Rotherham. Interviewed by telephone 17 July 2026 Does not remember this job and says so plainly. Explains how a table gets re-keyed, which is the part that matters.
- Ingrid Saavedra-Cheung, Metrologist. Hired by this newspaper in May 2026 to scan an original H-block and a replica for a different story; her certificate is published May 2026 Gives block 14's channel floor as 0.6mm across 610mm.
- South Yorkshire coroner's officer, Public record. Written record request 20 July 2026 Cause of death recorded as cardiac arrest.
Documents
- PX-1974 — Replication of a Puma Punku H-block — 411 logged hours, footage, scans and calibration accepted
What we could not confirm
- Who moved the decimal point. The typescript was thrown out in 2003, no marked proof survives, and the compositor is 84 and does not remember the job. Author, typist, compositor and proofreader are all live possibilities and we cannot choose between them.
- Whether forty is the right number of books. We counted forty in English, French, German and Spanish by searching the figure and the site name together. We did not search Russian, Japanese or Chinese and have no idea what is in them.
- Whether his 2009 and 2014 letters were read by anybody. We hold his certificates of posting for both. One firm has since been wound up; the other tells us it has no record of receiving anything, and we have no way to test that.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Hollow PostTHE MAN THEY WOULDN'T LET CORRECT PUMA PUNKU
Ran his three letters as evidence that a correction was suppressed.
Had that we did not
The letters, which are real, and the nil replies, which are also real.
Left out
That his correction makes the site less impressive rather than more, which is the whole of why nobody wanted it; and that indifference explains a silence better than instruction does.
-
Signal & LedgerErrata And The Short-Run Technical Report
A trade note on proofing practice in self-published field reports, and a good one.
Had that we did not
The mechanism, explained more clearly than we have explained it.
Left out
His name. Fourteen hundred words on the error without once naming the man who spent fifteen years trying to get it fixed.
-
They Buriedthis newspaperHe Wrote 0.5. The Book Printed 0.05. Merrick Osgood-Talabi Dies At 79.
Read the notebook, traced every citation we could find, and checked the figure against a laser scan we had already paid for.
Had that we did not
The two notebook pages at full size, the three letters, the certificates of posting, and the metrologist's certificate.
Left out
We printed the wrong figure in 2021, eighteen months after he wrote to this desk and got no answer. That belongs in the first paragraph of this story and it is in the tenth. — V. Ashcombe-Doyle, standards editor
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.