HN Theater @HNTheaterMonth

>Satoshi was NOT an excellent C++ programmer.

I'm not a programmer but that is what Gavin Andresen said[1]; he said something like this "Satoshi was an excellent C++ programmer but he wasn't a cryptographer" and "Satoshi wasn't familiar with Cryptography 101" in a sense that Satoshi was sometimes mixing up basic cryptographic concepts.

And yea I know Bitcoin had plenty of bugs I heard of notorious Value overflow incident[2].

>IMO it was initially written by someone deep in government (but I have no proof, just my opinion from looking at the code).

My assumption is someone from the academia e.g. university professor. Maybe someone who was teaching freshmen basics of Computer Science and basic C++ programming then your point might be valid that Bitcoin's codebase was poorly written looking from the practical and from the professional point of view.

[1] https://www.youtube.com/watch?v=rQ3e1Pzu7iI

[2] https://en.bitcoin.it/wiki/Value_overflow_incident

Like Gavin Andresen said Satoshi wasn't a cryptographer.

First reference in Bitcoin Whitepaper is "W. Dai, b-money". But Satoshi never heard of it, Adam Back suggested to Satoshi to read it and he decided to reference it in Bitcoin Whitepaper.

"I suggested Satoshi should look into B-Money, which he didn’t seem to know about at that time, and this is how I think B-Money was added to the paper."[1]

Satoshi said to Wei Dai "I was very interested to read your b-money page. I’m getting ready to release a paper that expands on your ideas into a complete working system. Adam Back (hashcash.org) noticed the similarities and pointed me to your site."[2]

Satoshi seemed to be very good C++ programmer who had interests in electronic cash and cryptography but not a cryptographer.

Look and hear what Gavin said in his presentation:

https://youtu.be/rQ3e1Pzu7iI?t=766

[1]https://cointelegraph.com/news/adam-back-on-satoshi-emails-p...

[2]https://nakamotostudies.org/emails/satoshi-and-wei-dai-corre...

⬐ heimatau

When I said 'no proof exists', I didn't mean 'no proof of authority'. I mean no document exists. Assuming Satoshi 'covered ones tracks' is a straw-man because then you open up the possibility to anyone being Satoshi, not an exclusive person.

Please understand that many people talked with Satoshi through email as well. Those emails are mostly private. When someone points to an authority, the entire argument is a straw-man, even with someone very relevant to the discussion.

> Satoshi seemed to be very good C++ programmer who had interests in electronic cash and cryptography but not a cryptographer.

What physical circumstantial evidence exists that is shows Kleiman was good at anything you've described or was even interested in these during 2008 or earlier (heck, I can't find anything prior to 2013)? What about any interest into any digital cash/gold attempts?

The argument you're giving is fairly persistent in the crypto community. I'm not entirely sure why, except intellectual laziness. The irony here is 'proof of authority' argument...is an absolute contradiction to what bitcoin's primary function is (trustless 3rd party system). Why are people lifting up a 3rd party and not the code/evidence? An opinion (no matter how relevant) isn't what we're looking for, we are looking for evidence. Did scientific reasoning leave everyone?

(edited for grammar)

⬐ dutchbrit

Is there a tldr (or in this case tldw)?

⬐ xsb

Satoshi Nakamoto is/was the anonymous creator of Bitcoin. He voluntarely disapeared a few years ago. Gavin Andersen is the lead developer of the open source Bitcoin project since then. In this video he talks about the past and the future of the cryptocurrency, and touches some parts of Bitcoin history and his relationship with Satoshi.

⬐ jsprogrammer

But what didn't Satoshi know?

⬐ marcell

From the video:

- That you can compress ECDSA public keys from 64 to 33 bytes

- Compression techniques that can be applied to the bitcoin protocol, such as inverted bloom lookup tables

- SNARKS

- Fully homomorphic encryption

- Various bugs in the initial bitcoin implementation, including one that created ~2 billion BTC

- Whether bitcoin was legal

- Whether it would take off

⬐ kolinko

Fully homomorphic encryption is a really big one. It allows to perform operations on an encrypted data - in theory you could run a computer program that would have it's memory contents encrypted at all times.

In case of cryptocurrencies it means that it's possible to create a cryptocurrency that has all the advantages of bitcoin, but also guarantees total anonymity - i.e. you're able to prove that you have the coins and not show where the coins come from. Mixers would not be necessary, and it would not be possible to track the coins.

The practical implementation of this is called DarkCoin. When we were doing Orisi we considered launching a sidechain to Bitcoin that would do similar stuff. But abandoned the project - privacy is important, but we all know who would be the first clients would be ;)

Also, it is be possible to create computer programs that run on your computer and have their own money, but that money is impossible to be stolen from them (the private key is never decoded and never reaches the computer memory).

Imagine a frustration of a computer hacker that gets a program which will pay him 1 BTC if, and only if he solves a specific equation. The program is open-source, and yet there provably is no way to hack into it.

Or a program that sends you money when it sees phrase "XXX" on Hacker News (you cannot cheat by providing altered HN website, because the program verifies HN SSL certificate on a homomorphically encrypted virtual machine).

⬐ jsprogrammer

Is there a runnable, fully homomorphic encryption program that allows running turing complete languages?

⬐ kolinko

IIRC it's a very fresh research. From what I heard It's definitely possible, but I don't think there are any working implementation of a turing complete machine on this.

⬐ mcs

Never roll your own crypto, unless you're a cryptographer.

At least he realized that.

⬐ dutchbrit

But he did design his own protocol

⬐ firasd

I converted this talk to mp3 audio to listen to it more conveniently: https://www.dropbox.com/s/cr388hk0uarg90j/DevCore%20Boston%2...