~dr|z3d
@RN
@RN_
@StormyCloud
@T3s|4
@eyedeekay
@not_bob_afk
@orignal
@postman
@zzz
%Liorar
%cumlord
%snex
+FreefallHeavens
+H20
+Onn4l7h
+Onn4|7h
+Over
+bak83
+bpb
+poriori
+profetikla
+qend-irc2p
+r00tobo
+uop23ip
+xHarr
Arch
BubbRubb
Danny
DeltaOreo
Irc2PGuest2981
Irc2PGuest4354
Meow
Stormycloud_
T3s|4__
ac9f_
acetone_
anontor
duck
floatyfloatilla
halloy13412
mahlay
makoto
mareki2p_
nZDoYBkF_
nilbog
ntty
null911
r00tobo[2]
shiver_
simprelay
solidx66
thetia
u5657
vivid_reader56
zer0bitz_
StormyCloud
I accept Schoko Bons as payment
dr|z3d
you're handling donations, I'm sure you can arrange something :)
snex
shoko bons is when you have to move boxes around to a correct arrangement, right?
cumlord
site is giving sex appeal
cumlord
I think i2p should do merch
cumlord
Itoopie thigh high socks £30
dr|z3d
congratulation, cumlord, you will dildo of the day award. you're the dildo. :)
dr|z3d
*win
not_bob_afk
cumlord: I'd buy that!
dr|z3d
don't encourage him, not_bob_afk
dr|z3d
I worry about you.
dr|z3d
*** chuckles. ***
not_bob_afk
;)
not_bob_afk
Unrelated, I'm making progress with my sniffer. Getting some useful stats. Here at some point in the near future I'm going to run Shannon Entropy scores on every address I've ever seen.
not_bob_afk
This will allow me to easily detect custom hostnames, and also graph the distribution/curves of normal addreses.
dr|z3d
harry: what's this shannon entropy stuff not_bob_afk's talking about, or is he just making things up?
not_bob_afk
I'm nearing a point where I can release stuff that's useful.
H20
Shannon entropy is Clausius’ thermodynamics but applied to information—it measures unpredictability in bits: *H = −Σ p * log₂*p*.
H20
Not_bob_afk isn’t wrong—just poetic.
H20
Entropy reaches its maximum when all outcomes are equally likely and a minimum value of zero if one outcome dominates.
H20
It is useful for cryptography, compression algorithms, as well explaining why your life seems like such chaos at times!
not_bob_afk
dr|z3d: It allows me to score for randomness.
dr|z3d
of course it does, not_bob_afk :)
not_bob_afk
;)
snex
wouldnt it be easier to just check against dictionaries and look for consecutive strings
not_bob_afk
Anyway, it will be a fun thing to run on millions of addresses that have been in use.
cumlord
It’s truly an honor to accept this dildo
not_bob_afk
snex: That is also an option.
not_bob_afk
cumlord: My goal is to take any hosts that hit the threshold run through a dictionary.
not_bob_afk
Creating the score is much cheaper computer power wise than a brute force.
not_bob_afk
And when dealing with this many addresses, that really matters.
snex
i dont see how you can assign a score to a b32 without making assumptions about vanity addys
snex
00000...000.b32 is just as random as anything else
not_bob_afk
I'll see when I do the math.
not_bob_afk
It's not.
not_bob_afk
I mean, it is.
not_bob_afk
But, the score will not be the same.
not_bob_afk
There will be a curve with repeating letters/numbers.
not_bob_afk
Or, I could be totally wrong. I'll find out!
not_bob_afk
But, more than that, it will let me see how all the addresses that I've ever seen plot!
snex
i mean this kinda sounds like what "intelligent design" people have been claiming will revolutionize biology for 30 years
not_bob_afk
lol
cumlord
I think I tried something like that on a small subset after snex noticed something on the first few chars not looking random, or something, I don’t remember
not_bob_afk
I'm breaking it up. Calculating a score for the first 8, first 12 and whole address.
not_bob_afk
It's not just the score that I care about, but the deviation on the first 8/12/whole.
snex
if you look at the dataset as a whole you will see some chars appear more than others but i dont think you can score an individual b32
not_bob_afk
Why not?
not_bob_afk
And, I can. I've done it.
snex
because every b32 was generated using the same random process, even vanity ones. you have to make assumptions about what data outside the b32s to make these scores
snex
what data outside the b32s looks like*
cumlord
duhh that’s why, vanity addresses
snex
the same reason we cant look at DNA and say "a designer made this." you can only tell design by looking at how other known-to-be-designed things came about
not_bob_afk
Nah. The scores are not the same. the score for my addres (first 12 charecters is 3.2516, the score for a vanity adddress is 1.5850
not_bob_afk
Another totally random address scores out at 3.4183
not_bob_afk
The vanity address is very much an outlier.
not_bob_afk
Though, another one that has a vanity name scores 3.2516
not_bob_afk
But, only the first few charecters are vanity.
snex
you are scoring how "random" the word they grinded is, most likely
snex
bookkeeper - low score
snex
l33tsp34k - maybe higher score
not_bob_afk
Yeah, lower score means likely selected word.
not_bob_afk
Right, and it just means possible, not for sure.
snex
i bet a lot of b32s out there lack entropy, esp for first 8/12
not_bob_afk
I add two more non-random charecters to that address with only the first four vanity and the score score goes down to 2.9183
not_bob_afk
And it becomes an outlier.
not_bob_afk
I've only tested a small sample so far by hand. But, soon enough I'll do the entire list and then I'll find out the whole picture.
not_bob_afk
Basicly checking to see if what I'm about to try is even useful. And, it appears to be so.
snex
another thing you might want to do is normalcy test - ie test the liklihood of every combo of digits for every combo size
not_bob_afk
It's wroth trying.
not_bob_afk
But, I don't think I need to. What I'm doing will give me mor than enough information.
snex
tho im not sure if b32s would actually be expected to pass that. b64/b32 are weird
snex
*** makes a vanity grinder that lets you intersperse your desired digits ***
not_bob_afk
It's a thing you could do.
not_bob_afk
vain also lets you add the vanity part of the address at the start, end or anywhere.