Hacker News new | past | comments | ask | show | jobs | submit login
Bitcoin Non-Technical FAQ (oleganza.com)
60 points by snitko on Oct 2, 2012 | hide | past | favorite | 33 comments



"Bitcoins are not created upfront and distributed to some privileged persons."

Well, yes and no. Around half of the bitcoins there can ever be have already been created and distributed to some privileged persons (certainly far less than half the world's population). IIRC around half of those belong, or belonged at some point, to its original inventor.

I admire the mathematics of bitcoin, but I'll never buy any until they make the production curve fairer.


No production curve can ever be "fair". If you imagine that one day every living person has equal amount of bitcoins, they will be very quickly redistributed from those who doesn't care to those who do for a relatively small price. And then, when the network becomes more stable and useful, those who have sold their fair share early will whine again about being unfair distribution.

Second: even if a lot of bitcoins are initially belong to a single person, it poses no problem. Either you hold money that nobody needs (value is very low), either you hold money that people a willing to buy from you and every day you are not selling them, your opportunity cost rises.

In other words, it's entirely possible to create inflationary effect by stashing 50% of money and then suddenly selling them out. But the imaginable effect is very limited:

1. The longer you keep the money, the longer you bear the risk of uncertainty. If your bitcoins today are worth 100K USD, there is no guarantee they will be work 1M USD next month. This creates an incentive to sell earlier.

2. Each successive act of selling naturally lowers the price as supply is increased. So if you start selling your stash, the remaining part will be worth less (or will grow in value slower and slower)

3. Generally, you cannot sell in bulk. If people are trading 1 BTC for 10 USD, it does not mean that somebody will buy 1000 BTC for 10000 USD. There necessarily be a discount or simply nobody will spend 10K USD at once. This returns us to points #1 and #2.

So even if Satoshi had a lot of bitcoins initially, he has either sold significant part of them already, or will be able to "disturb" the market by selling them at once only to his own great expense.


>No production curve can ever be "fair". If you imagine that one day every living person has equal amount of bitcoins, they will be very quickly redistributed from those who doesn't care to those who do for a relatively small price. And then, when the network becomes more stable and useful, those who have sold their fair share early will whine again about being unfair distribution.

I said "fairer", not "fair". While there is no perfectly fair distribution curve, the deflationary one bitcoin uses is far worse than any other currency in existence. The very concept of one individual owning 1/4 of the entire world's supply of dollars or gold or even sea shells is absurd, because such an individual would have to be ridiculously, obscenely wealthy.

If you want a practical suggestion, simply removing the exponential halving would make a far fairer curve. People would then obtain wealth proportional to how long they'd cared about bitcoins (assuming that investments and computing power grow at approximately the same rate).

>Second: even if a lot of bitcoins are initially belong to a single person, it poses no problem. Either you hold money that nobody needs (value is very low), either you hold money that people a willing to buy from you and every day you are not selling them, your opportunity cost rises.

>1. The longer you keep the money, the longer you bear the risk of uncertainty. If your bitcoins today are worth 100K USD, there is no guarantee they will be work 1M USD next month. This creates an incentive to sell earlier.

But they're inherently deflationary, so actually you gain money the longer you hold them. If you have 4 million bitcoins now, you can guarantee that you will always have at least 5% of the entire bitcoin economy, and economies generally grow.

>2. Each successive act of selling naturally lowers the price as supply is increased. So if you start selling your stash, the remaining part will be worth less (or will grow in value slower and slower)

>3. Generally, you cannot sell in bulk. If people are trading 1 BTC for 10 USD, it does not mean that somebody will buy 1000 BTC for 10000 USD. There necessarily be a discount or simply nobody will spend 10K USD at once. This returns us to points #1 and #2.

Sure, you'd have to ration them out over time and not spend them all at once, like an arab prince who owns millions of barrels of oil. Doesn't make you any less wealthy, just means your assets are not entirely liquid.

>So even if Satoshi had a lot of bitcoins initially, he has either sold significant part of them already, or will be able to "disturb" the market by selling them at once only to his own great expense.

I'm not worried about him disturbing the market. I'm worried about concentrating an enormous amount of wealth in his hands. You can say he and other early adopters deserve some profit from creating the system, and I'll agree with you. But 1/2 of the entire future economy being concentrated in the hands of so few is ridiculous.


If Bitcoin becomes the currency of the entire future economy, the initial distribution method will have been proved to be a non-issue. More likely, Bitcoin will be one currency amongst many, and the early adopters will sell their windfall gradually according to supply and demand.


Gets a few facts wrong— E.g. the maximum block size is 1MB not 10MB, and there is no facility to increase it without replacing all the software—it's as fundamentally hard to change as the supply of coins because it's also a scarcity which is essential to the security of the system. (without it, the system would rapidly become centralized as miners produced enormous blocks taking every single with-fee transaction; and then no one but miners could afford the computational, storage, and network costs of validating the gigantic blocks; nor would there be reason to pay substantive transaction fees to fund security).

Overall it's pretty good.


Also wrong, 'Can the government shut it down?'

Should be: Q. Who can shut it down? A. Anyone who is willing to pay 51% of the monthly operating cost of all miners. Plus an initial investment to manufacture the hardware. Neither of these numbers are out of the range of Governments or Banks. And they probably won't be for a long time. This is the death of Bitcoin scenario that worries me most. And it seems no one is really worrying about it or taking it seriously.

edit: Here are the numbers someone calculated at the height of the first mining bubble [1]. We're currently at less than double the hashing power of then [2]. So a current rough estimate would be to double all those numbers.

[1] http://bitcoin.stackexchange.com/a/1094

[2] http://bitcoin.sipa.be/speed-lin-ever.png


Controlling only 51% of the current operating power of the miners would take over eventually but it would take a very long time. Since each client trusts the longest chain, you would have to recompute a new chain that's longer than the entire chain so far. If you only pay for an extra 1% over the "legitimate" chain, it would take several years to make a longer chain.


You're thinking of the wrong scenario. If you have 51% of the power you start mining a new block off the latest legit block but you include 0 or no "real" transactions in it. After you find one, you only compute new blocks off of _your_ "bad" last block. Since you have 51% you will control the entire blockchain from this point on. And you can basically shut down all transactions.

Note that we aren't trying to reverse any transactions or steal people's bitcoin. We are just trying to stop all transactions. This is what bitcoin is all about, so do this and you've destroyed bitcoin.


51% doesn't mean you always "win" a block, just that you win 51% of the time.


Ok I'll bite and explain it.

You're right, sometimes someone else will find a block before you and will add it onto the longest existing chain. IF you always appended your bad blocks to the longest chain then the blockchain would look something like bad->bad->good->bad->bad->good->bad->bad. BUT you WONT be adding your blocks to the longest chain. You will always add your blocks to YOUR OWN last bad block. So you will always be working on a blockchain that looks like ... ->bad->bad->bad->bad. And since you have 51% of the power your blockchain will eventually always be longer than the other chain(s) miners are working on. For a while the two chains (legitimate and your bad one) will go back and forth between which is longer. But at some point yours will get long enough and outpace the legitimate one.

If you don't understand this you don't understand how bitcoin works. I don't say this to be a dick, but I say it because this is a real concern that a lot of people don't realize or think about.


I hadn't thought about only working on your own blockchain like that.

A piece I'm missing is the heuristic most clients use to determine "longest". Depending on how that works, this plan could be really easy or really hard.


I believe the way longest chain is determined is the sum of the difficulty of all the blocks. This basically boils down to just the chain with the most cpu power that went into it. So as long as you control the majority of the cpu power, you can control the longest chain.


Why do you need to recompute the entire block chain. At the moment, the block chain is in some state. If you control the majority of the processing power, you can devote it all to expanding the chain from that state. Because you have more CPU, your version of the block chain would replace the old version when you choice to release it. This would not let you re-write history from before you started mining, but gives you complete control of events after you started.


"Controlling" would only give you the power to block legit transactions and not to change the ownership of any coins, correct?

If so, could people could react by making btc nodes that consider that rude behavior and begin blacklisting rude nodes wrt which block chains they consider accepted?


Correct, it would only block legit transactions.

This is one of the proposed solutions. Basically centralize bitcoin by somehow whitelisting good blocks or blacklisting bad ones. The problem is this would require a change to _all_ bitcoin software. Meaning EVERYONE has to upgrade to this new centralized approach. Plus the decentralization of bitcoin is lost.

Pretty sure that solution would spell certain doom for bitcoin.

Gavin offered a solution which _might_ work [1]

[1] http://gavintech.blogspot.com/2012/05/neutralizing-51-attack...


On a side note, I recall hearing that almost all of the TOR traffic is routed through about 12 exit nodes. That seems like the government is investing the resources needed to weaken that system. Of course, I have no idea how large the TOR network is relative to bitcoin.


Thanks, updated with the correction.

However, the block size limit has nothing to do with scarcity and competition. If there were no limit, nothing would change today as number of transactions created is not even approaching 1 Mb per 10 minutes. And if somebody wants to create a lot of spam, they would have to pay for that in form of fees to compensate for the wasted bandwidth and blockchain storage.

If I understand correctly, the block size is limited to make block distribution smoother across the peers + limit the disk space consumption until there are lightweight clients.

Please correct me if I'm wrong.


That's correct. 1MB has a lot more to do with the mechanics of moving validated data across a peer to peer network than anything else.


I would prefer a tutorial on how to safely use it to another pretty high-level overview, just in the QA format this time.


Mentioning things like "Peer to peer digital currency" and "protocol of storage and exchange" in the first paragraph seems to me like it would turn a lot of non-technical people off.

The "Why is it any good?" answer would be a much better introductory paragraph I think!


This is going to sound a bit naive on my part, and perhaps it is. If the bitcoin transaction database is signed and stored on every computer, what keeps this database from growing too large to handle?


Nothing keeps it from growing too large. In fact, this is the problem that has several theoretical solutions.

First, people may develop and use lightweight clients. They will only store wallet and sign transactions, but will not store and verify the blockchain themselves. For instance, they may refer to some public service to figure out the current balance.

Second, there is a possibility to develop a client which stores only the block headers and downloads only necessary blocks to validate particular transactions.

Third, it is also possible to compress/discard old blocks that contain transactions with empty recipient addresses.


Would using only a partial block chain result in severe security issues?


No. If someone does not want to keep a full block chain, then there are two alternatives: a personal client with partial blockchain or an escrow service. First option is definitely more secure.


It was proposed long time ago alternative approach to the storing and using chain where only certain number of latest blocks are used. And obviously - longer partial chain - more trust.


Long term, it looks like storage is getting cheaper faster than (human population)*(transactions/person) is growing. That might take a couple decades before it's practical, though.


This needs some editing by a more experienced English speaker if it's to be widely read.


This comment needs some editing by a less haughty writer it it's to be widely read.

Error-wise I didn't notice a ton of things, but:

> This [could use] some editing by a more experience English speaker if it's to be widely read [without grammar pedants coming out of the woodwork].


Can you point to some part of the text that needs editing? I don't think the English in the FAQ is bad written at all (as a non-native speaker myself). I agree the "non-technical" thing is a bit misleading though.


Error-wise I didn't notice a ton of things, but:

> Bitcoin is designed to be a faster, cheaper and a more secure currency

> Security is achieved by having every participant do the verification themselves using [an] open protocol based on well[-]known cryptography methods

> The original software source code is open and available [to] everyone for review and improvement.

> The reward is halved approximately every 4 years until [a] total of 21 million bitcoins are generated around year 2140. Currently more than 9 million of bitcoins are available already.

Also "bitcoin" is not consistently capitalized.

That stuff aside, it's also not very idiomatic. I'm not sure what the ideal audience is, but if it's, say, my 60-year-old mom who doesn't really understand browser tabs, I don't think it's going to reach them. It's good for me though, as a programmer who doesn't specialize in crypto or know much about currency theory (if that's a thing).


Thanks for the remarks. What do you think about this way to capitalize "Bitcoin"? http://blog.oleganza.com/post/32746652183/how-to-capitalize-...


Makes sense to me.


I would appreciate if you point out some mistakes. My email is oleganza@gmail.com Thanks.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: