|
JP Aumasson
@
veorq
Switzerland
|
|
author of nostarch.com/seriouscrypto — founder/CEO @teserakt_io — security @taurusgroup_ch — VP @kudelskisec
|
|
|
20.775
Tweetovi
|
502
Pratim
|
14.982
Osobe koje vas prate
|
| Tweetovi |
| JP Aumasson proslijedio/la je tweet | ||
|
Teserakt
@teserakt_io
|
6 h |
|
new post about E4's cryptography, answering questions such as:
- why AES SIV and not GCM?
- why SHA-3 and not BLAKE2?
- what about side channels?
- how are keys stored?
etc.
blog.teserakt.io/2020/02/03/iot…
|
||
|
|
||
|
JP Aumasson
@veorq
|
11 h |
|
someone claims to have "cracked" RSA, should I tell him that Time AI already did? pic.twitter.com/PaCQmDRPOS
|
||
|
|
||
|
JP Aumasson
@veorq
|
1. velj |
|
🧂 := randomSalt()
if !IsGoodRandomSalt(🧂) {
return errors.New("bad bad salt!")
}
😜 pic.twitter.com/W9Tf6MzNT3
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
yeah but I find witness confusing cos it's also used in the contest of zero-knowledge proof and NP languages for example, something more specific to hash trees may be better imho
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
authentication path, Merkle path, Merkle proof, inclusion proof.. looks like we need a standard term for this, any other candidates?
/cc @GEndignoux @cr_yp_to
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
before our post on E4's crypto, this writeup by @diagprov starts by talking of threat modelling, functional constraints, trust assumptions, business needs, platform support, etc. the opposite or typical twitter/HN is-it-secure-or-insecure discussions blog.teserakt.io/2020/01/27/iot…
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
BLAKE4 will clean dishes, cut vegetables, and iron clothes
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
only works if you only do very basic cooking and always use the same utensils
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
yeah reading Baudelaire while watching my rotting food, meditating on the ephemeral nature of things, and love
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
doesnt scale to random dishwasher runs every other day, and it's only a 10min job; only AI and blockchain can help here
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
I live in fscking Switzerland, "poor background" here is being French,; I pay my maid 30 CHF per hour (she comes 2h per week, does more/better than I would in 2 days)
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
my maid is doing this once every week so that's fine; dishwasher runs more frequently so that's a problem; we need to make the world a better place with autonomous AI-based dish bots that are better than this: reddit.com/r/shittyrobots…
joint venture with @diagprov and @thegrugq?
|
||
|
|
||
|
JP Aumasson
@veorq
|
31. sij |
|
billion-dollar business idea: a product that empties a dishwasher automatically and stacks clean dishes at the right place; I'll be the first customer
|
||
|
|
||
| JP Aumasson proslijedio/la je tweet | ||
|
Matt's Grad Students
@MDGGradStudents
|
9. sij |
|
We put together a site to look for applied crypto research. Eprint is great, but wading through 1500 papers/yr for 100 practical papers is 💩. Join us and help out! acrab.org #RealWorldCrypto
|
||
|
|
||
|
JP Aumasson
@veorq
|
30. sij |
|
you don't always have the luxury to choose the best/coolest/fastest primitives when you depend on many other components :)
|
||
|
|
||
|
JP Aumasson
@veorq
|
30. sij |
|
key takes the whole first block, so yeah it's padded, and message data starts in the second block
|
||
|
|
||
|
JP Aumasson
@veorq
|
30. sij |
|
I think you're right, so if you hash 2^64-1 bytes with B2s and a key then there would be a counter overflow in the very last block, correct?
|
||
|
|
||
| JP Aumasson proslijedio/la je tweet | ||
|
peter purgathofer
@peterpur
|
29. sij |
|
|
||
|
JP Aumasson
@veorq
|
29. sij |
|
won't be surprised if this is a major news in the Swiss newspapers
|
||
|
|
||