Skip to main content
23 events
when toggle format what by license comment
Mar 12, 2023 at 18:21 comment added Ofir Gorodetsky I see, you guessed the filename from the EUDML broken link, cool. It could be that these two scans are all there is on-line (I am not sure I've seen the remaining 1965 paper with my own eyes, could be I became aware of it only through zbMath/MathSciNet, I can't recall exactly, these sites have a short review of it).
Mar 12, 2023 at 13:31 comment added darij grinberg Ah, thank you for that one! Then hdml.di.ionio.gr/pdfs/journals/994.pdf is the second one. The scans are horrible and I can only download them with "wget --no-check-certificate", but it's a start...
Mar 12, 2023 at 13:27 comment added Ofir Gorodetsky @darijgrinberg No, sorry. Google does find the last one: "On congruences in number-theory", see hdml.di.ionio.gr/pdfs/journals/1014.pdf (note the link gives a security warning).
Mar 12, 2023 at 3:07 comment added darij grinberg Do you have the Kazandzidis papers? The EUDML links are going nowhere right now...
Oct 27, 2022 at 14:20 vote accept Vlad Matei
Oct 26, 2022 at 18:18 comment added Vlad Matei Great! At least this shows that what I was trying to attempt for another problem clearly does not work. Thank you for looking into it
Oct 26, 2022 at 18:16 history edited Ofir Gorodetsky CC BY-SA 4.0
added 35 characters in body
Oct 26, 2022 at 18:11 history edited Ofir Gorodetsky CC BY-SA 4.0
added 35 characters in body
Oct 26, 2022 at 18:03 history edited Ofir Gorodetsky CC BY-SA 4.0
added 35 characters in body
Oct 26, 2022 at 18:00 comment added Ofir Gorodetsky @VladMatei I managed to prove above, using existing results, that your congruence fails for $n=4$, $a=2$ and $b=1$ and most primes up to $10^9$ (only 2 exceptions). For $n \le 3$ it holds by existing congruences you are familiar with.
Oct 26, 2022 at 17:54 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1792 characters in body
Oct 26, 2022 at 17:47 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1792 characters in body
Oct 26, 2022 at 17:40 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1792 characters in body
Oct 26, 2022 at 17:35 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1792 characters in body
Oct 26, 2022 at 17:27 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1792 characters in body
Oct 26, 2022 at 16:33 comment added Ofir Gorodetsky @DavidESpeyer I've updated the answer, thanks.
Oct 26, 2022 at 16:32 history edited Ofir Gorodetsky CC BY-SA 4.0
added 126 characters in body
Oct 26, 2022 at 16:27 comment added Vlad Matei Thank you Ofir! I was aware of all those results but they do not help for this
Oct 26, 2022 at 16:24 comment added Vlad Matei Thank you for the remark! This does not exclude the $p^n$. I would be happy with an example showing that is also false; in the example you had $n=2$ so mod 5^2 it works out
Oct 26, 2022 at 16:19 history edited Ofir Gorodetsky CC BY-SA 4.0
added 245 characters in body
Oct 26, 2022 at 16:01 comment added David E Speyer Note that the congruence as stated in the OP is false: We have $\binom{ap^n}{bp^n} \equiv \binom{a p^{n-1}}{b p^{n-1}} \bmod p^{3n}$, but we only have $\binom{ap^n}{bp^n} \equiv \binom{a}{b} \bmod p^{3}$. Indeed, I checked $p=5$, and we have $\binom{50}{25} \equiv \binom{10}{5} \bmod 5^6$, but $\binom{50}{25} \equiv \binom{2}{1}$ only modulo $5^3$.
Oct 26, 2022 at 15:37 history edited Ofir Gorodetsky CC BY-SA 4.0
added 1432 characters in body
Oct 26, 2022 at 15:23 history answered Ofir Gorodetsky CC BY-SA 4.0