Hi Richard, since this suggestions section has been open for so long, I thought I’d ask something. You’ve written a few times about Goodstein’s theorem, a straightforward proposition in number theory that’s provable in second order arithmetic but independent of PA.
What I’m wondering is whether there’s a proof of that theorem in classical analytic number theory, i.e. preferably not a reverse mathematics construction followed by Goodstein’s proof with ordinals, but rather, something with natural motivation like Hardy and Wright might have done, using contour integrals or whatever.
And a meta question about the above question (whether there’s a straightforward answer to it or not): is the question itself interesting, mathematically speaking? I just remember finding it surprising to hear that theorems about integers could be proved with analysis, and wondered if the use of analysis was unavoidable (this was before I knew any logic, not that I know much now).
dear Richard,
i have an idea about how to prove that every “almost perfect number” is a power of two:
gauss proved that a Constructible polygon is always a Fermat prime. if we could show that 2^ap +1 (ap= almost perfect number) is also always Constructible we actually proved that every “almost perfect number” is a power of two. if you could give me ideas about how to porove that i would be grateful.
Tuesday 2008 February 26 at 3:37 am
Hi Richard, since this suggestions section has been open for so long, I thought I’d ask something. You’ve written a few times about Goodstein’s theorem, a straightforward proposition in number theory that’s provable in second order arithmetic but independent of PA.
What I’m wondering is whether there’s a proof of that theorem in classical analytic number theory, i.e. preferably not a reverse mathematics construction followed by Goodstein’s proof with ordinals, but rather, something with natural motivation like Hardy and Wright might have done, using contour integrals or whatever.
And a meta question about the above question (whether there’s a straightforward answer to it or not): is the question itself interesting, mathematically speaking? I just remember finding it surprising to hear that theorems about integers could be proved with analysis, and wondered if the use of analysis was unavoidable (this was before I knew any logic, not that I know much now).
Thanks for any thoughts.
Saturday 2008 December 20 at 5:33 pm
dear Richard,
i have an idea about how to prove that every “almost perfect number” is a power of two:
gauss proved that a Constructible polygon is always a Fermat prime. if we could show that 2^ap +1 (ap= almost perfect number) is also always Constructible we actually proved that every “almost perfect number” is a power of two. if you could give me ideas about how to porove that i would be grateful.
Tuesday 2009 March 3 at 9:41 am
Hello webmaster
I would like to share with you a link to your site
write me here preonrelt@mail.ru