Hashpower is the Ends, Bitcoin is the Means
TL, DR: Hold bitcoins so miners can inflation tax you in order to fund a censorship resistant, immutable, timestamped publishing system. Creating an irreversible system is very difficult without starting from centralization. Creating a currency made it possible.
Before Satoshi’s paper, the computer world was struggling with having a good timestamped publishing system. This is an exploration of how Bitcoin might have been invented to make one.
The internet did not have a good public timestamped publishing system. There was no good way of showing that some piece of data was sufficiently aged, while being difficult for some central party to censor or fake. This seems like kind of a banal subject, but coupled with the right fraud, can be very lucrative.
The leaked files show the firm regularly offered to backdate documents to help its clients gain advantage in their financial affairs. It was so common that in 2007 an email exchange shows firm employees talking about establishing a price structure
In the old days, someone would take out a classified ad in the local paper. It would serve as notice for some event, satisfying some legal requirement. There was competition, so if one paper censored, another would publish the notice. The papers were timestamped and it was difficult to recall all the papers once they were distributed. Changing history after the fact wasn’t going to work. A reader could parse the entire contents of all qualified newspapers to prove the negative, that some notice hadn’t happened. Two things accomplished are the timestamping (not being able to backdate), and the publishing (being able to read it all if you want).
This brings us to the main point of what blockchains are for: proving the negative. By agreeing on the latest block, a community innately agrees on the complete history taken to get there. By examining all the blocks, a computer can determine if any conflicting items existed. This is analogous to courts agreeing that public notices can be found in sufficiently large newspapers.
Bit-gold was an early attempt at internet money. It was really handwavy about timestamping, deferring to centralized services. For publishing, bit-gold relied on Szabo’s property clubs which used emergent authenticity, and were handwavy in and of themselves.
An interesting pre-Bitcoin system applied features of the newspaper with a centralized timestamp. Instead of relying on their promise not to backdate timestamps, they assemble customer data into Merkle trees, make a hash chain from them, and checkpoint the hash chain in the newspaper. Note, this only proves some data is aged, but doesn’t reveal unknown conflicting data. On top of that, it is awkward to see if they reversed history, since the verifier would need to find the paper-based gazette. It couldn’t be done online, since the newspaper’s website could be altered on a whim.
Hash chains were not novel, nor was packaging multiple things into them. There was also knowledge about individuals brute forcing password hashes. Hashcash was a technology commonly known as a way to filter email spam. This email defense system never caught on, though. Combining hash chains and hashcash would make it expensive to append data. This is desirable, since it would also make it as expensive to rewrite history. There would be a large electric cost to rewrite history. Without a large electric cost, there is only a social cost to rewriting history, and only if others had already seen it. Assuming that multitudes of people were doing hashcash on the same hash chain, it would be a great way to timestamp, since backdating would be so expensive. It would also be easy for a computer to check the age, since hashchash difficulty is easy to verify. Done properly it wouldn’t have to rely on a central party. It would also be an extremely effective way to communicate through computer networks while under attack by someone with moderate resources.
It might pay for some cracker to brute force passwords, but building and running this equipment is expensive. Although people were running SETI@home a decade earlier (myself included), looking for aliens is way more interesting than maintaining a timestamping system for the public. This public service needed to be turned into a private good.
Without the glamor of aliens, there needed to be a way to entice people to brute force hash chains. A good way of doing that would be to give them money. Who would pay for all these people to burn electricity? Also, if one person paid the bills, they would be able to set policy (indirectly censor). At that point we would have an improvement on the trusted timestamping system, but the publishing would still effectively be under centralized control.
One way to pay people doing the hashing is to give them money from nothing, i.e. seigniorage of some token. Inflation is one way to get money from nothing, but to do that it must expropriate from an existing value pool. A value pool’s size is defined by how many people want to store value in the pool; a market capitalization. Inflating a large market cap extracts more value than inflating a small market cap.
This meant that Bitcoin needed to be a central bank managing a token with a large market cap. To get that large market cap, it needed to be seen as an excellent store of value, hence the economic modeling imitating historical successes.
If this line of reasoning were true, then bitcoin the currency could be considered one of so many denigrated app-coins. It is a coin to extract value from savers in order to fund a massive timestamped publishing system. It would also give one reason why Satoshi never sold out. He is allowing the miners to tax wealth from his hoard.
This central bank has managed to marshal a staggering amount of electricity. One study from 2 years ago estimated the consumption on par with the Republic of Ireland
I know that Bitcoin is so much more than just a timestamped publishing system. I am but one blind man feeling the Bitcoin elephant.
Now that you have read this article, re-read the Bitcoin whitepaper abstract. This may be completely revisionist history, and I would love to hear the thought of some pre-Bitcoin cypherpunks to see if these ideas hold water.