Ask HN: What are we decrypting with bitcoin? | Hacker News Reader

Ask HN: What are we decrypting with bitcoin?

11 pointsbrianbreslin12 comments

So I looked and couldn't find a straight answer. When we are mining bitcoins, we're effectively decrypting hashes (security tokens, encrypted items), but what exactly are we decrypting? Are we cracking MD5 hashes for some nefarious purpose? Are we helping to break into some security system? Is it completely benign?

View on HN

It would be nice if the proof of work could do something like SETI at home did, or work on some other big problem like analyzing data from particle accelerators. It seems like such a waste to do work for works sake, just to generate money, but I guess there is some kind of commentary in the whole thing that reflects the true value of money. i.e we waste a lot of time chasing something that at it's core is immaterial.

To me that would be an amazing currency, one that is generated by fueling progress, the more we contribute to solving problems the more the money supply grows. I am no economist but that sounds like a win/win. I wonder how hard it would be for Bitcoin to retrofit something like that for the proof of work. Another cool idea would be to allow people to choose which project they wanted their cycles to work for. It would be pretty cool to be able to choose "analyze cancer genomes" or "search for new matter" and be paid for contributing.

It isn't just work for work's sake. The work done is actually protecting the blocks, the transaction record. It is difficult and the difficulty increases with computing power to make sure that no one could effectively forge past transactions. The mining difficulty is the mechanism that makes Bitcoin transactions secure and trustworthy.

I think part of the labor should go into something useful to society though.

Nothing is being decrypted by mining. Mining is trying to generate a valid block which hashes to a low enough value.

The only purpose of it is to be time consuming, as the idea is you can't altar historical data because it would be too time consuming to "catch up" with the main blockchain

More details here: https://en.bitcoin.it/wiki/Proof_of_work https://en.bitcoin.it/wiki/FAQ#How_does_the_proof-of-work_sy...

Cracking passwords would require finding data which hashes to an exactly specified value,

Bitcoin's proof of work requires finding data which hashes to 'any hash with the given rare property' (given number of leading zeroes, smaller than the target)

During mining, you're actually not decrypting hashes but encrypting a certain piece of data until the initial N digits of its hash are zero. By increasing the number of initial zeros required, you decrease the probability of it occurring and increase the time which it takes to figure out what data to encrypt to get that hash - therefore it gets harder to mine coins with each block.

As far as I can tell, there's no purpose behind the hashing, besides block generation. Miners maintain the shared ledger by forging blocks which contain transactions.

The reason it is hard to do, and increases in difficulty as more and more people mine is that it protects the transaction record. If the difficulty keeps up with the current computing power it makes it effectively impossible to forge transactions. Typically 6 blocks or about an hour is when transactions are considered rock solid. No one could have enough computing power to change a transaction 1 hour in the past.

Why don't you RTFM and see for yourself?

but aren't we creating a giant look up table to the MD5 encryption function? And couldn't that be used to decrypt as well?

Yes, but only for very specific data. If you stored the results, you'd have a lookup table of a tonne of 'salts' that appended to the transaction data. The only real danger is hash collisions I guess - but MD5 is a broken encryption scheme anyway.

Oh! I know this one! It turns out to be 42 in the end.

Reply on news.ycombinator.com