Jump to content

Talk:Nothing-up-my-sleeve number

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Digits of pi

[edit]

Using later digits of π, say starting at the 7,463,398th position, would not be considered as trustworthy. Perhaps whoever contributed that would care to enlighten us as to whether that position is a "Nothing up my sleeve number", or whether the digits at that point have some significance. MarkMLl (talk) 18:33, 18 July 2008 (UTC)[reply]

No special signifigance. I think that is the purpose of the word "say" in the sentence.--agr (talk) 19:19, 18 July 2008 (UTC)[reply]
Every finite sequence of digits can be found in π. So if you have a number which actually "has something up my sleeve", it is still in π...although, most likely, at a rather remote position. Foe eample, the number 12345 can be found in π, at position 49703. 11111 is at position 32789, 111111 at position 255946. Not exactly the most most trustworthy "security codes", eh? ;-) --Klaws (talk) 13:33, 18 February 2011 (UTC)[reply]
Your assertion is a common conjecture—indeed, it is widely believe that π is a normal number—but there is no mathematical proof. Ntsimp (talk) 15:45, 18 February 2011 (UTC)[reply]
Although, for practical intents and purposes, it might as well be since any chosen "nth" digits would have to be in the list of known digits, which have so far behaved this way.75.70.236.107 (talk) 01:51, 25 August 2013 (UTC)[reply]

Evidence that anyone notable has ever used this phrase?

[edit]

I understand the concept but I have never heard this name for it before, can anyone provide evidence that this name has notably been used? ciphergoth (talk) 11:48, 16 January 2009 (UTC)[reply]

Hardly seems to be used at all; almost all google hits for this phrase seem to be WP mirrors. --David-Sarah Hopwood ⚥ (talk) 23:11, 11 October 2009 (UTC)[reply]
I was hoping for the same thing, but I can't find anything. --B-Con (talk) 17:48, 26 June 2010 (UTC)[reply]

Slight flaw in reasoning - See birthday attack article

[edit]

1) Let's suppose that I want to make sure that certain bits in the constants are set to a certain pattern, but don't want it to be obvious.

2) I create a list of many seemingly innocent constants. Pi,1/Pi,sqrt(1/2), sqrt(7), e^Pi, Phi^7e, ln(7), etc. This includes putting 256 bits of said constants through MD5 or the like.

3) I cherry pick the constants with the relationships that I want between certain bits.

There is no trivial way to detect such shenanigans unless you already know what patterns would make the algorithm weak. However, it might take some not-incosiderable skill and imagination to create a believable set of constants. Thus, it's just like magicians' acts.  ;) Also, there might be legitimate reasons to want constants defined with more complicated formulas. 75.70.236.107 (talk) 02:03, 25 August 2013 (UTC)[reply]

Million dollar curve

[edit]

Is the million-dollar-curve a kind of "nothing up my sleeve" curve (set of number): https://cryptoexperts.github.io/million-dollar-curve/ --Oliverpoool (talk) 13:52, 3 May 2016 (UTC)[reply]

The backdoor in Dual_EC_DRBG had been exposed?

[edit]

Does it?

From my understanding, no backdoor about Dual_EC_DRBG had been exposed. It has been shown that it is it is possible to create a backdoor, and some suspect the NSA did just that. However, AFAIK there is no actual proof. By proof, I mean the backdoor itself, or a solution to a related problem that cannot be solved with state-of-the art cryptanalysis. — Preceding unsigned comment added by GuBx42 (talkcontribs) 16:52, 8 July 2019 (UTC)[reply]

and your point being? the article does not claim it has a backdoor. the article claims that 1, it does not adhere to the numsn principle and 2, some people suspect that it has a backdoor. Krisztián Pintér (talk) 17:18, 8 July 2019 (UTC)[reply]