/biz/ - Business & Finance

>>54612902
That's fucking retarded, it requires that the users trust the validators, and it requires that users trust that gold will be handed over when they cash out.
You might as well just issue a gold-backed IOUs in a federated manner.
KYS retard.

>>54552619
You can't use biometrics for cryptography because:

1. Biometrics usually don't have that much entropy in the first place.

2. Biometrics are "open source". You are constantly leaving fingerprints on everything you touch and face-scans on every camera you walk by

3. Biometrics work off of "soft" metrics like the relative position of visual patterns in fingerprints or eyeballs or faces. In other words, it's an analog secret. There's no such thing as an analog hash algorithm or because you have to deal with low-precision details (person's face and fingerprints change over time, sensor inaccuracy, lighting conditions, humidity, etc). If your hash/asymmetric key algorithm accepts values within a certain range, that breaks the premise which is that you can't make an improved guess of the public key or hash from a near-guess.

This is why if you have an android or iphone, you can set a fingerprint or face scan, but you can't use it to encrypt the phone's storage. You need a "real" digital passcode like a PIN to unlock it on boot.

If you want to use biometrics, tattoo a QR code onto your ass. QR codes are digital and have built in redundancy.

>>54519982
Programmers will say this all the time, it is ridiculous.
Do security issues just spontaneously come out of nowhere? No. They come from increased complexity and the addition of new features.
If you just audit and patch the software without adding additional shit, it eventually reaches a state where it has no more bugs.

>>54167905
As the twitter user explains, jamtis will also "fix" it by increasing the ringsize, so the threat of output spamming is decreased.

But to be fair, if someone is going out of their way to use tx_extra to indicate they are fucking with outputs, that's really more of a blackball than output spamming.

>>52680880
bro it literally uses trusted execution for privacy. That is the most insane retarded thing I have ever heard of. Intel can literally just break it whenever they want.
The whole security model is based on the idea that it is impossible to inspect or modify silicon (it is impractical, but you only have to do it once to break the privacy for everyone).
If there is a single failure point (Intel Corp.) then why not just use a central bank?

DCA is for suckers. XMR at $70, BTC at $7k. Don't forget it.

>>49297887

>>49254948