Lemdro.id

3 points

1 year ago

hmac(passphrase, “one”) -> seed used to create the private key.

Its so god damn simple man. Passphrase is the key. Standardize the solution so that when Hmac(passphrase, ‘one’) emits the same private key on two separate devices, we know that their code is legitimate. Run tests on commercial solutions to make sure they emit the standard answer to a set of publicly known private-keys (as well as a few personal tests to ensure it works on your end) and bam, problem solved.

You’re telling me that all the best cryptocoin wallet peeps can’t come up with a college-textbook answer like that?

As other keys are needed, use hmac(passphrase, “two”) and hmac(passphrase, “three”), etc. etc.

No. The answer is that no one is really trying to solve the info-sec issue with regards to cryptocoin. Its just a money game to them. There’s elementary / college-level solutions that are in the front of any textbook (or maybe left as an exercise, that you’d find at the back of the book / answer key).

When the level of cryptography knowledge is this primitive, I know that the entire damn community ain’t serious about it and I shouldn’t waste my time with them.

report

manitcor@lemmy.intai.tech

[ - ]

0 points

1 year ago

thats how it works, im not sure what you are getting at?

report

[ - ]

3 points

1 year ago

Oh really, wallet hardware companies are publishing the hmac and algorithm used to go from passphrase to private key?

Care to post one?

report

manitcor@lemmy.intai.tech

[ - ]

3 points

1 year ago

there are opensource wallets, the standard is called BIP39

im not sure if any of the hardware providers are doing it though

im still not sure what you are getting at, if you are suggesting I somehow trust hardware cold wallet providers, I dont, does not mean Im not stuck using the tools.

report

[ - ]

2 points

1 year ago

if you are suggesting I somehow trust hardware cold wallet providers, I dont

I’m saying there’s an obvious solution to anyone who has passed a cryptography 101 course here.

PKI private keys are randomly generated prime numbers and/or ellipitcal curve numbers (depending on algorithm). Either way, that random number generator needs a seed, and that seed can be based off of the passphrase. BIP39 isn’t the whole solution, that’s just a way to turn long-strings of alpha-numeric characters into binary data.

My overall point is that there’s a blatantly obvious, simple solution to the hardware wallet problem. I brought it up because its not a hard crypto-problem to solve. The fact that there’s no adequate solution in 15 years is a failure of the cryptocoin community. Not due to a failure of basic cryptography problems.

The cryptocoin community, despite using “cryptography” is a joke. They barely know how to use cryptography even at its most elementary levels. It takes 15 years to come up with crappy, untrusted hardware wallets and they still can’t open a basic textbook to come up with a better solution that’s already written down.

report

manitcor@lemmy.intai.tech

[ - ]

3 points

1 year ago

the issue with the hardware wallet is not a “simple math” problem but a “trust” issue. in reality you simply can’t trust any hardware you didn’t make yourself, in practical use we usually pick vendors we like and decide to trust them.

for example. many people considered ledger trustworthy until they introduced firmware that indicates a capability to exfiltrate the keys.

I think the problem you are speaking to was some older hardware keys (and maybe some strange off brands) that encode keys at the factory, to my knowledge no major product does anything like that and they take pains to show you are generating the key. the big back and forth there has been with hardware providers using methods that are potentially reversible or other types of vulnerabilities.

Yes pretty much all devices will allow you to import a key you have generated by whatever means you prefer, however once you put it on the device you are signing up for the other issues that come with hardware still.

report

[ - ]

1 point

1 year ago

I don’t think you recognize how easy it is to generate trust with the methodology I laid out.

Buy a standard-compliant offline wallet.
Buy a second, standard-compliant, offline wallet that you know uses a different codebase, as much as possible.
Generate a passphrase. Use it on #1 and #2 to generate your wallet/private keys.
Is it the same private key? Success. Unless the wallets have fallen prey to the same flaw (unlikely, as they were manufactured from two separate companies and running two separate code paths), you’re probably good.

“The Standard-compliant” method is any algorithm that goes from hmac(passphrase) into seed -> generating the random numbers needed to build a wallet. (Prime numbers or whatever). As far as I can tell, this “standard method” doesn’t exist, not yet anyway.

report