Two Groups Showed Their Working. Two Would Not.
Four computations of the same orbit disagree by 214 years. We asked all four to publish their inputs. The two that did both reproduced on our correspondent's own machine. The other two are not wrong. They are worse than wrong.
Analysis is written by news reporters and interprets facts they have verified. It is not opinion, and it takes no position on what should be done.
ITHACA, N.Y. — Four groups compute a solution. Their answers for the next perihelion passage differ by 214 years.
This piece is not about which of them is right. It is about a much smaller and much more answerable question: which of them will show their working.
The spread
The four published dates span 214 years. That is not, by itself, remarkable — solutions to poorly constrained orbits diverge, and any orbit constrained mainly by a broken tablet is poorly constrained.
What is remarkable is that the four groups all cite PX-1207, all cite substantially the same secondary literature, and still land 214 years apart. Divergence of that size, from inputs that similar, has to come from the computation rather than the evidence.
Which means the computation is the interesting object.
The question we asked
We wrote to all four with one question: will you publish the complete input set to your solution — initial elements, epoch, integration method, step size, and perturbing bodies included?
The Grupo Efemérides Austral published theirs by return, in full, with a note apologising for not having done so already. The Hattingen Circle published theirs after a second request.
The other two declined. One cited "ongoing refinement." The other gave no reason and did not reply to a follow-up.
What happened when we re-ran them
Dr. Vantongeren took both published input sets, implemented them independently, and integrated them on her own machine without reference to either group's code.
Both reproduced. Austral's to the day; Hattingen's to within four days, a difference she traced to a rounding convention in their epoch and which the Hattingen Circle acknowledged within an hour of being asked.
That is what publishing inputs is for. Two groups can now be argued with. Two cannot.
The distinction this piece is drawing
There is a habit in every field of treating the confidence of a claim as evidence for it, and a corresponding habit of treating disagreement between claimants as evidence that all of them are unserious.
Both are lazy. The four solutions disagree by 214 years, and that fact tells you nothing at all about whether any of them is right. What tells you something is which groups will hand you their arithmetic.
Austral and Hattingen have. Their answers still differ by 61 years from each other, and both of them are now improvable, because a third party can find the error.
The other two solutions are, for the moment, not wrong. They are worse than wrong. They are not checkable.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
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.