On Thu, Dec 19, 2013 at 12:43 AM, LizR <lizj...@gmail.com> wrote:
> On 19 December 2013 12:13, Telmo Menezes <te...@telmomenezes.com> wrote:
>>
>>
>> All the transactions that occurred so far are registered in a file
>> that is shared between the nodes in the network.  New transactions are
>> broadcast to many nodes.
>>
>> One of these nodes is going to be lucky enough to find a way to
>> incorporate the outstanding transactions into the file according to
>> very strict requirements. These outstanding set of transactions will
>> form a "block". A block contains the following things:
>>
>> - the hash of the previous block
>> - the set of transactions
>> - an arbitrary number (nounce)
>>
>> An hash is the output of a one-way functions. One-way functions are
>> hard to invert, so getting the original block from the hash is
>> computationally hard. The bitcoin protocol wants to make the hash hard
>> to create, in part because every time a hash is discovered, the
>> discoverer is rewarded with a predetermined number of bitcoins. The
>> way to make the hash hard to create is that the network agrees that it
>> must start with a certain number of zeros. The only way to meet this
>> requirement is through brute force, by trying random values for the
>> nounce until one works.
>>
>> Once the hash is found, a new block is created and work will begin on
>> finding the nest one, ad infinitum. This is why the ledger file is
>> called a blockchain. Each block hashes the hash of the previous block.
>>
>> This difficulty also serves as a proof-of-work (a receipt that shows
>> that a certain amount of computational effort was spent, on average).
>> This protects the network against attacks. If a node received two
>> conflicting blockchains, it will chose the longest one. This way,
>> unless the attacker controls the majority of the computing power of
>> the network, it cannot create a fake blockchain longer than the rest
>> of the network.
>>
>> So mining for bitcoins is the same process that allows for
>> transactions. There is also the possibility of transaction fees. When
>> you make a transaction, you can volunteer to pay a fee to the miners.
>> The discoverer of the next block will receive this fee. Nodes that
>> receive your transactions are not forced to accept them, so the fee is
>> an incentive for them to accept it. As mining becomes less profitable,
>> it becomes more likely that miners will expect fees. Once all coins
>> are discovered, the network will work solely on fees, and I imagine
>> fee prices will emerge naturally (miners will compete on price, users
>> will pay more according to urgency). In a market with many
>> transactions, mining can become profitable even with no new coins to
>> discover and low fees.
>>
>> What "contains" your coins are wallets. Wallets are two random
>> numbers. One is public, for incoming transactions and one private, for
>> outgoing transactions. Only you know your private address but if you
>> sign a transaction with it, the validity if the transaction can be
>> confirmed through a one-way function against the blockchain. So
>> ultimately, you keep possession of your coins by knowing the private
>> address.
>>
>> So the blockchain is a gigantic number and the wallets are numbers.
>> The actual coins are not numbers, they are a complete abstraction.
>>
> Thank you very much for that description, which I think I have more or less
> managed to understand. (I assume the 21 million limit is an outcome of this
> system demanding that the hast start with a specified number of zeroes?)

No, actually the number of zeros is variable, and is determined by the
moving average of the time between block discoveries. This way, the
rate of block discovery can be kept more or less constant (about one
every 10 minutes).

The 21 million is a convention that all nodes agree on. When you try
to generate a block, you include all the transactions your node
received plus one transaction to give you an agreed upon amount of
bitcoins. If you find a solution, the other nodes will check a number
of things before accepting it. One of those things is your reward.
There is a predefined agreement on the size of the reward, based on
the number of blocks already discovered. This is tuned so that it is
reduced by 50% about every 4 years. This just happened last November,
and the initial reward of 50 btc was halved to 25 btc. Once 21 million
btc are created, the reward becomes zero. This is all enforced by the
majority of computational power in the network.

> It sounds as though these things will eventually mimic house prices, which
> "decouple" from the cost of building after a while and go into a
> market-driven upwards spiral. (Well, except that people actually need
> houses...)

I would argue that this is different. With houses, the price went up
because the banks were increasingly lax in giving  credit for people
to buy them. They started to give credit to riskier costumers. This
increased demand, and thus the prices went up. Eventually the riskier
clients stopped being able to pay and the bubble bursted. This was
accelerated by the derivative markets over toxic credit swaps and
eventually even spread to countries that had unwittingly invested in
such swaps, like Iceland, which was bankrupt by the crash.

This happens similarly with investment bubbles, where people get over
excited about some type of business (tulips, railways, .com) and
invest too much. At some point, it becomes clear that these business
cannot possibly generate enough profit to reward all these
investments. Again, this is aggravated by leveraged investments (based
on credit).

I don't think any of this applies to bitcoin. I doubt that people buy
them on credit, and they are intangible, unlike houses or businesses.
Their value is purely a human convention. It is a direct
representation of the amount of trust people put in them. So my point
is that they can burst, of course, but it won't be for the same
reasons as houses or tulips.

I don't think it's unreasonable or even unethical to speculate on them
at this point. People who do are using their money to help bootstrap a
decentralized currency. They are taking a risk, and if the market
succeeds they will be rewarded. I don't think it can generate a
dystopia, because if wealth distribution becomes too uneven, the
majority can bail out. This is not so with fiat currency.

In fact I bet that some cryptocurrency will replace fiat money in the
long run, but not necessarily bitcoin. It is possible to imagine other
cryptocurrencies with different rules, and we will probably go through
a period of experimentation. In any case, bitcoins will be seminal,
because it will be easier for people to buy these new currencies with
bitcoins. The more money exists in decentralized currency markets, the
easier it is to experiment with alternatives.

Telmo.

> --
> You received this message because you are subscribed to the Google Groups
> "Everything List" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to everything-list+unsubscr...@googlegroups.com.
> To post to this group, send email to everything-list@googlegroups.com.
> Visit this group at http://groups.google.com/group/everything-list.
> For more options, visit https://groups.google.com/groups/opt_out.

-- 
You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to