« first day (2961 days earlier)      last day (1699 days later) » 

1:42 AM
@Patriot Uh, are you sure about that?
You know that we consider him to be a bit... wrong, right?
I mean, his heart is in the right place, but if you promote his site, you'll be promoting pseudoscience and snakeoil. You shouldn't do that if your goal is to improve the state of cryptography knowledge.
 
 
2 hours later…
3:31 AM
1
Q: Efficient way of generating a random number of N (less than 64) bits with exactly M bits equal to one

VincentWould there be an efficient way to implement a function with the following signature: unsigned long long int random_word(size_t n, size_t m) that would generate a random machine word (64 bits here) such that exactly m bits over the n least significant ones at set to 1. For example: random_wor...

 
 
3 hours later…
6:14 AM
-1
A: Bitlocker/Filevault and VeraCrypt together?

Raphael M.Lets say you have an encryption method A, that maps your input to B; Applying another encryption method, lets say X, thtat maps B to C will not help, because there must exist a function Z that maps A direct to C. A(input) => B then X(B) => C or directly: Z(input) => C In other words, mathemat...

This answer makes my skin crawl.
 
6:46 AM
@forest I definitely support Paul's effort to get people to generate their own random numbers. It is one of the main things that Eve and Mallory don't want anyone else to do. Once you have generated a lot of random numbers, then you can at least have a private conversation if you air gap and move the ciphertext properly. That is the point.
 
6:57 AM
@forest When we disagree with others on this site, we can still talk and keep up the debate, even a vigorous disagreement--without forming a clique. This site is not the eighth grade, thank goodness. "We" means all members of the site, and that is a lot of people, many of whom have voted for, and obviously appreciated, his contributions. That said, whenever someone says something we really disagree with, then let them have it--politely.
 
@Patriot Eh, I can't think of one regular here who thinks he's correct.
And his ideas for generating random numbers are... not ideal, at best.
Promote his site only if you want to encourage people to do unsafe things.
He constantly shows that he has little to no understanding of the things he talks about. This isn't a new thing. It's been consistent for many years. There's no debate to be had.
 
@forest Is that a real criticism of anything? "Your _________ is not ideal."
 
I used that term simply because I didn't want to sound too harsh. :P
He considers himself a self-styled expert in TRNGs while showing a very basic understanding of EE at best and a very poor understanding of cryptography and information theory.
I mean, lately he's been improving a bit, to be honest.
But pretty much anything related to entropy, he'll be wrong about.
Shortly before you came here, he deleted a majority of his downvoted answers. Before that, a very good portion of his answers were heavily downvoted (conspiracy theories, bad or dangerous advice).
 
Well, try to talk to him about it. You are both smart guys.
 
Oh I have, as has everyone else here. :P
But he's Dunning-Kruger if I've ever seen it. That's not an attack, just an observation.
 
7:02 AM
Maybe the time will come for a rapprochement.
Do you have a website for me to promote?
 
What kind of website? Are you in need of a website to promote?
 
I am trying to spread knowledge about cryptography.
Or anything closely related.
 
DJB's website (cr.yp.to and the various subdomains) are good for more complex topics.
 
Right
 
There's also uh, what's it called, Cryptopals?
 
7:05 AM
I saw that.
Veracrypt
Tails
libsodium
NaCl
this one
Schneier's blog
 
Schneier's blog is good for political stuff, but it's light on detailed cryptography.
 
Dirk Rijmenants
Right
But when he says something crypto-related, it really matters.
he = Schneier
 
yup
 
Did you ever play with
Dirk Rijmenants's Enigma simulator?
or M-209 simulator?
I thought it was a blast.
 
No. I'm not particularly familiar with classic ciphers.
I should read up more on them, though. They seem interesting.
 
7:15 AM
The simulator he made for the M-209 will dazzle you.
But I wish he had a certificate so I could play with it on https
and no verification of any downloads
Man, what gives?
I wrote him.
I want to know why he does not support https on his website.
I am not happy about it.
 
Did he not reply?
 
I am on vacation and have not checked.
I hope he has.
I am going to step back from this site pretty soon because of my work.
But recently I learned so much.
Rabin's cryptosystem
There is Rijmenants's M-209 simulator
Here in Seoul a bottle of Bek Se Ju is calling. :)
Later...
One, more thing, if you use the M-209 simulator, then make sure to click on it and open the cover so you can see the wheels, etc., inside.
 
 
4 hours later…
11:12 AM
Wow, some real crypto question on StackOverflow, for the experts:
Lenstra elliptic-curve factorization algorithim
1
Q: Lenstras Elliptic curve problem with attribute error

Dead_Ling0Writing some code, for tne Lenstra elliptic-curve factorization algorithim. I'm fairly happy with the code, however it only works some of the time. The function that adds points should return either an integer or a point. In the case it returns an integer d the lenstra block of code should simply...

Any EC persons that are able to help? I'd really have to dig into it to help, and there's not enough time for that.
 
12:13 PM
@MaartenBodewes It might be a programming problem, based on "I've found that if I add if type(Q) == int return Q to the start of the function point_add() then it seems to work as intended though this isn't ideal i feel."
 
12:53 PM
Given the state of the programming I presume it is one. However, within a good feel of the intent of the statements, it is hard for me to say what's wrong and what's not.
What I mean is that I think the programmer clearly understood the problem and the programming reflects that. So it is likely something to do with an erroneous statement in there.
 
 
1 hour later…
2:08 PM
@Patriot Bek Se Ju is pretty nice.
I've spent zero time in South Korea, but people in Japan had that occasionally.
 
@bdegnan Thanks. Greetings from the Land of the Morning Calm. :)
Chosin
You gotta spend some time in South Korea. It's positive and I think you will find it fascinating.
You already know so much about Japan and China.
 
I don't speak Korean. :P
 
I don't either.
Their language is syllabic.
 
Between Chinese and Japanese, I probably could just learn Korean.
 
It is similar to hiragana and katakana.
I knew a dude who was half Chinese and half Japanese.
 
2:15 PM
My Japanese dialect shares 20% of the words with Korean. It's complete unintelligible to those who only speak Tokyo standard Japanese.
 
He said that he wanted to learn Korean.
:D
I am not joking.
 
Seems like a good path
 
Which dialect do you speak?
 
I speak the Tokyo Standard and Kanazawa Dialect from the ISHIKAWA region.
There's something like only 50k speakers left.
 
so cool
I like the dialect up north around Akita.
O-ban desu.
so charming
I hope you get a chance to visit S. Korea for bit.
Very positive. I find that travel throws a light on the other places we have been.
 
2:20 PM
I've been to the North a few times.
South Korea is about the only place I haven't spent time in Asia.
 
I understand.
That is good.
It will be very meaningful because of your background.
 
When you speak Chinese, you can get pretty far.
 
So many Chinese here.
 
Japanese/Chinese/Russian. That pretty much gets you around Asia.
 
I come often.
I understand that the RU folks are in Harbin and Xiamen.
Xiamen... port city
 
2:23 PM
It's pretty useful in the *stans.
 
Of course
 
Uzbekistan was basically me improving my Russian as Uzbek is a bit of a mystery. :P
I was totally unprepared for that trip.
 
They say that Tashkent is a party city...
Better than the Mog :D
I have only been in Bishkek
 
When the Chinese were shoot the Ughirs, I had to get out by going West. White people make a good target, and I they were just spraying bullets. It was a strange time to be there. That was almost 20 years ago at this point.
 
I see
 
2:26 PM
"do you have a visa?" "I have cash" "You have a visa" "awesome"
 
That was a very different world.
pre-9/11
 
Yeah. It was.
 
Are you doing any interesting work at the moment?
 
Through Tajikistan, to Uzbekistan, north to Kazakstan, and then I caught the trans-Siberian railroad back to Vladivostok. I then took a container boat to Toyama Japan. I was like 2 weeks late to class.
I only do interesting work.
 
You are an adventurer!
 
2:31 PM
Everyone is at 22. :P
 
Educational :)
I was in the 101st ABN and went to Egypt when I was 23... so I know how you must have felt.
Much sticks in the memory.
Are you doing any especially interesting work at the moment?
@bdegnan By the way, do you like Edward Seidensticker?
 
I only do interesting work. :P
 
He edited Kawabata Yasunari's "Yukiguni"
 
I've never read any of his translations, as I generally read the original texts.
 
translated
This is very good.
I see.
My Japanese was never good enough.
 
2:47 PM
I still speak Japanese daily, and I try to read a few pages of a book a day. I'm not nearly as good as I was when I lived there. I need to do the same with my Chinese. It's falling apart on the reading side. :/
 
But you have never touched Murasaki Shikibu, right?
 
my kids speak Japanese to me.
 
Or have you?
I got it.
 
no, I don't do the classics.
Sadly, a lot of my Japanese reading is semiconductor physics
 
May I recommend to you Murasaki Shikibu?
 
2:50 PM
Yeah, I'll check it out.
 
Kanai ...
she's Thai
The patter of little feet is in the distance.
I think we will start with Vernam Cipher and modulo addition...
Then we can move on to how to not trust anyone... admit nothing, deny everything, make counter-accusations, etc. You know, education.
From the cradle...
Mukden Incident
The Three Alls Policy
"We will not speak of it because of our tact"
etc.
EDUCATION
:)
@bdegnan If you don't mind, I would like to recommend one work to you, as a break from your efforts in semiconductor physics:
Kawabata Yasunari's "The Izu Dancer"
GAME OVER
:)
cheers
 
 
1 hour later…
4:08 PM
@Patriot "Well, try to talk to him about it. You are both smart guys." Well that generated one unexpected chuckle. Believe us. That's the first thing we tried. This is a website for communicating knowledge, so...
He is a persistent problem. He's not someone you successfully reason with. The end result is never correction or mutual understanding. His responses are always weasely. He does sort of take in what we say to him, but he thinks everything is a matter of opinion and only uses criticism to rewrite posts to not sound so outright ridiculous.
2
He's said that he finds math mostly useless.
He thinks you can judge things just by eyeballing. He thinks cryptography is a craft, not an academic field. He thinks his own intuition is more valuable than any formula. He thinks that people are conspiring against him to censor him, when in reality it's damage control. Or dismay and shock over how wrong he gets things with 100% confidence.
2
 
4:24 PM
Adding to that, first principles physics are not welcome either.
 
Oh ya. Pseudoscience and pseudomath.
There's not anything we can do besides correct him, though. He's gamed the site's reputation system pretty well. (One up vote = 5 down votes, plus users upvote him just because he sounds like he knows what he's talking about and has high rep to back it up.) But he's not really breaking any site rules either.
 
4:46 PM
Yeah, it seems that you can do pretty well on SE just by answering, regardless of it is actually factual. It took me forever to make 2k; however, there's not really many questions that I feel qualified to answer. Interestingly, I have higher rep here than electronics.se.
 
 
3 hours later…
7:24 PM
I tried answering this interesting question on Ethereum.stackexchange about a potential for a partial collision on the trailing 160-bits of a Keccak digest, which could cause two different private keys on elliptic curve secp256k1 to have access to the same ethereum account (because the account address is based on only the last 40 hex of the hash of the public key): ethereum.stackexchange.com/questions/10055/…
I suggested the collision-related part of the question be better suited for crypto SE here.
 
8:20 PM
I meant to cut out the "You are both smart" sentence because that wasn't really the funny part. However, what he is is something worse than "not smart". He's one of those people that see math and science and conflates it with opinion.
Like the kind of person that watches a TV show about space on the Discovery Channel who then starts saying they proved Einstein wrong. "I don't get general relativity. It clearly doesn't make sense. It must not make sense to any physicist. All physicists must be bullshitting the public. What I'm doing has just as much validity as what they do - but I'm not the one lying!"
 
science is inherently revisionist. Facts change, opinions do not.
 
I'm not sure. I think in some fields we're in "definitely mostly correct" territory. We'll refine things but not fundamentally change them.
When we put science to practice, through engineering for example, we are constantly using the science to make predictions and those predictions turn out to be accurate. Kind of a prerequisite to us making tiny machines that can do billions of calculations per second very very little chance of making errors.
Ideally we use science to figure out what ideas are wrong. We can't prove ideas to be correct, just increase are confidence for them. If science is "revisionist" it's only because we're adding things to our knowledge, not undoing any claims in the hard sciences. We are never going to find out that the planets and stars are actually lights on the inside of a crystal orb rotating over our heads.
 
9:23 PM
@FutureSecurity Science constantly improves itself. Just like with software, there's the "bleeding edge" branches where paradigm shifts are the norm and strongly-supported theories are toppled overnight, and there's the stable branches where any contradiction in evidence is a big deal.
Of course, the average non-scientist only sees all this bleeding edge information being disproved again and again, and it poisons their understanding and trust behind the process.
tl;dr I totally agree. In many fields, we are quite certain that what we know is correct.
@StevenHatzakis Yeah, that seems like a good fit for this site.
 
10:09 PM
@StevenHatzakis The accepted answer is correct. There is nothing magic about Keccak or formatting that exempts it from the pigeonhole principle.
Well. Mostly correct; the sentence ‘They will create the same public key, which the ECDSA verification algorithm is performed against’ is wrong (here ‘they’ = distinct public keys sharing a common address), but the conclusion is correct.
(Two distinct private keys necessarily have distinct public keys, but what matters here is that they can share a common address.)
 
10:27 PM
Thanks @forest and @SqueamishOssifrage (nice to hear from you) and thanks for the feedback here, I am reading your comment after having just posted this follow-up in the hopes that someone could dive deeper and perhaps show some proof or calculation, it still seems a little ambiguous to me in terms of whether there can be a way to be sure there are potential address collission and what those chances are:
0
Q: What is the possibility of collision of trailing 160 bits of Keccak_256, for any two differing public-keys as pre-images?

Steven HatzakisEarlier today I was answering a question on the ethereum SE site that analyzed the potential for more than one private key on curve secp256k1 (which maps to a distinct public key) to control the same ethereum address (which are derived by hashing the public-key with Keccak256 as a byte array, and...

 
10:38 PM
The pigeonhole principle tells you that there are guaranteed to be collisions. The lack of any structure in the hash function related to secp256k1 suggests that on average the number of distinct private keys corresponding to any particular address is about 2^256/2^160 = 2^96.
 
11:05 PM
> Indistinguishable from pseudorandom
Someone needs to make a book containing quotes from Paul.
 

« first day (2961 days earlier)      last day (1699 days later) »