Skip to main content
25 events
when toggle format what by license comment
Mar 24 at 13:53 comment added Alexey Ustinov @GHfromMO I've added my reconstruction of Jacobsthal's argument as a separate answer.
Mar 24 at 8:19 history edited GH from MO CC BY-SA 4.0
added 2 characters in body
Mar 24 at 5:53 history edited GH from MO CC BY-SA 4.0
added 952 characters in body
Mar 24 at 0:26 history edited GH from MO CC BY-SA 4.0
added 67 characters in body
Mar 23 at 23:44 history edited GH from MO CC BY-SA 4.0
added 1029 characters in body
Mar 12 at 21:30 history edited GH from MO CC BY-SA 4.0
added 3 characters in body
Mar 12 at 20:50 history edited GH from MO CC BY-SA 4.0
edited body
Mar 12 at 3:13 comment added GH from MO @RolandBacher I removed the reference to Barnes (1974), but added a detailed treatment in the "Added" section. I found the exposition in Jacobsthal (1907) too terse.
Mar 12 at 3:08 history edited GH from MO CC BY-SA 4.0
deleted 64 characters in body
Mar 12 at 2:45 history edited GH from MO CC BY-SA 4.0
deleted 8 characters in body
Mar 12 at 2:39 history edited GH from MO CC BY-SA 4.0
deleted 8 characters in body
Mar 12 at 2:21 history edited GH from MO CC BY-SA 4.0
added 1359 characters in body
Mar 12 at 2:14 history edited GH from MO CC BY-SA 4.0
added 1359 characters in body
Mar 11 at 13:09 comment added Roland Bacher Thanks for your great answer! (Concerning the Barnes proof : I remember having read it 30 years ago and there was a really stupid mistake with modular arithmetics (the type of mistake in all those marvellous proofs of the Fermat-Wiles Theorem).)
Mar 11 at 13:03 comment added GH from MO @RolandBacher I don't know that. The brief MathSciNet review about Barnes' paper was written by Paul Erdős, but that of course does not mean much. Jacobsthal's proof should be reliable (and there was an earlier proof by Cauchy, presumably hard to read for us). BTW, I expect that one can also see from the analysis for $p\equiv 1\pmod{16}$ why the pattern breaks down for $p\equiv 1\pmod{32}$. Thanks for accepting my answer officially!
Mar 11 at 13:01 vote accept Roland Bacher
Mar 11 at 12:47 comment added Roland Bacher The proof by Barnes is fatally flawed if my memory is correct.
Mar 11 at 1:56 history edited Gerry Myerson CC BY-SA 4.0
spelling
Mar 11 at 0:30 history edited GH from MO CC BY-SA 4.0
deleted 11 characters in body
Mar 11 at 0:23 history edited GH from MO CC BY-SA 4.0
added 703 characters in body
Mar 11 at 0:11 history edited GH from MO CC BY-SA 4.0
added 703 characters in body
Mar 10 at 22:36 history edited GH from MO CC BY-SA 4.0
deleted 1 character in body
Mar 10 at 22:28 history edited GH from MO CC BY-SA 4.0
added 1305 characters in body
Mar 9 at 18:26 comment added Roland Bacher Nice observation which shows a bit more: The odd number among $\lbrace A,B\rbrace$ is always a square modulo $p$. If $p\equiv 5\pmod 8$, then half of the even number (which is odd) is therefore also a square.
Mar 9 at 18:11 history answered GH from MO CC BY-SA 4.0