A Fun Introduction to Bitcoin / Crypto-Currencies / Math-based Currencies

[muh_roads]

NEW UPDATE:

Check out Elwar's thread for free Bitcoin for RPF members.
http://www.ronpaulforums.com/showthread.php?454121-Free-bitcoins-to-the-first-50-Ron-Paul-Forums-users

--------------------------------------------------------------------------

This is an invitation to come join us on the Bitcoin / Crypto-Currencies Board.
http://www.ronpaulforums.com/forumdisplay.php?91-Bitcoin-Cryptocurrencies

Not the biggest fan of neon pink pants but this guy explains things WAY better than most videos I've come across.

http://ultraculture.org/blog/2014/03/04/vibrantly-fun-introduction-bitcoin-youll-ever-watch/


https://www.youtube.com/watch?v=84lEZpFML6Y

Do you own any Bitcoin's (or Litecoin's?) If no, why not? While I've found myself guilty of this at times, it would be nice if we could debate without ad-hominem attacks and cheap one-liners.

[FSP-Rebel]

bump

[Weston White]

Investing in Bitcoins, is now really the best time for something like that? You have heard that its primary founder is somebody that the media cannot get a hold of that nobody knows anything about him, a virtual ghost (ironically sort of like how Bitcoins resemble the ghost of currency); another founder is avoiding the press entirely; and still yet another founder apparently just jumped from her apartment building, killing herself? Also that its primary exchange Mt.Gox filed bankruptcy a few days ago. And not to mention that odd surge in its value from April of last year, resulting in a three-quarter loss in overall value.

[kpitcher]

Investing in Bitcoins, is now really the best time for something like that? You have heard that its primary founder is somebody that the media cannot get a hold of that nobody knows anything about him, a virtual ghost (ironically sort of like how Bitcoins resemble the ghost of currency); another founder is avoiding the press entirely; and still yet another founder apparently just jumped from her apartment building, killing herself? Also that its primary exchange Mt.Gox filed bankruptcy a few days ago. And not to mention that odd surge in its value from April of last year, resulting in a three-quarter loss in overall value.
Hi Weston. I can't give any investment advice as that's a personal decision. I will say that many people on Wall Street, Silicon Valley, and on this forum have invested. I know my initial investment has returned amazingly well over the past few years and continues to do so. As to your questions...

- There is one primary founder, Satoshi, who created it all. He - or they as it may be a group - hid his identity from the start letting his work speak for him. Newsweek broke a story about finding the guy. It isn't conclusive at all and I think newsweek got it wrong. Also a bitcoin discussion place that the original Satoshi used years ago had an update saying Newsweek was wrong on the ID. For the past number of years Bitcoin has been developed by the community and Satoshi hasn't been involved.

- There are no other founders. Other people in the open source community acting as lead developers working on consensus of the users, yet no other founders.

- One girl who ran a bitcoin exchange - basically a commodity exchange letting you convert bitcoin to other things - died. There are many exchanges around the world. It was a smaller exchange, I'd never even heard of it.

- Mt. Gox has not been the primary exchange for over half a year. They were big years ago yet they had issues of slow deposits and withdrawls for many months. The bitcoin community was saying avoid them. Even places like Ben Swan was saying avoid them.

- What time frame are you talking about for loss in value? If you go back a year, March 2013 Bitcoin was $50. There have been surges for that did drop. The latest surge was Bitcoin hitting $1200 in December when China got involved big. It dropped when the Chinese government ordered banks to not work with bitcoin exchanges. With all the negative Mt Gox issue hitting Bitcoin is sitting in the $600 range.

I think the recent amount of bad news with Bitcoin holding its value shows that the support for Bitcoin is still strong. I don't forsee bitcoin getting much cheaper again.

[FSP-Rebel]

For me, it's personal secession from the state and their henchmen as much as it is an investment in myself and all people that want to be freer.

[Elias Graves]

I have a hard time trusting anything run by anonymous individuals. If it's up front, be up front.
But I'm still waiting for my tulip bulbs to hit $10,000 again, so what do I know?

[KCIndy]

I have a hard time trusting anything run by anonymous individuals. If it's up front, be up front.
But I'm still waiting for my tulip bulbs to hit $10,000 again, so what do I know?

With all due respect, Bitcoin isn't "run by" anonymous individuals. As kpitcher noted earlier, "Satoshi" was the anonymous guy/gal/group who created bitcoin. But Bitcoin is a peer-to-peer network and the code is open source. It's not run by any cabal or clique.

I'm not sure how much more "up front" it is possible to be.

There's no way "Satoshi" or anyone else could control Bitcoin once it got going. That's one of the many aspects of BTC that I find appealing. It's largely beyond the control of any one person, group or nation. I truly believe it would be easier for a country to corner the gold market than control Bitcoin at this point.

But in the end, if you really dislike or distrust Bitcoin, there's an incredibly simple solution: don't use it. :)

[KCIndy]

Paladin69, I just realized you posted this in General Politics. Great idea!

I'll add a "seconded!" to your invitation for more people to come check out the cryptocurrencies board. I'm guessing there are still a lot of people, even here at Ron Paul Forums, who might have heard a few vague mentions of Bitcoin and/or cryptocurrencies but have no idea what it's really all about.

So if anyone is reading this and has any questions or comments, or particular insights you would like to share, please do as Paladin69 suggests and drop by:

http://www.ronpaulforums.com/forumdisplay.php?91-Bitcoin-Cryptocurrencies

Believe me, I can sympathize with anyone who has questions. It took me longer than I really care to admit to learn about this stuff myself, including the basic question of: "How the heck do I actually buy a Bitcoin???" so I'm a firm believer in the old adage that there are NO dumb questions. I can speak only for myself, but I will absolutely guarantee I won't be laughing at any legitimate question posted by anyone.

Curious? Interested? Skeptical? Drop in and give your opinion!

[ghengis86]

I think there are some fundamental problems with Btc and other crypto currencies. That's not to say things won't evolve in the future. I don't own any and I think there's still some growing pains coming. It's speculating right now; not investing IMHO (and I'm not here to piss in anybody's Cheerios, just throwing out my two cents).

But, anything that undermines government issued currencies is excellent. I see Btc having a place in the future as a medium of exchange complimntary to a store of value like gold. Also, crypto currencies are pretty amazing and I'm sure will evolve with good results.

[ghengis86]

Paladin69, I just realized you posted this in General Politics. Great idea!

I'll add a "seconded!" to your invitation for more people to come check out the cryptocurrencies board. I'm guessing there are still a lot of people, even here at Ron Paul Forums, who might have heard a few vague mentions of Bitcoin and/or cryptocurrencies but have no idea what it's really all about.

So if anyone is reading this and has any questions or comments, or particular insights you would like to share, please do as Paladin69 suggests and drop by:

http://www.ronpaulforums.com/forumdisplay.php?91-Bitcoin-Cryptocurrencies

Believe me, I can sympathize with anyone who has questions. It took me longer than I really care to admit to learn about this stuff myself, including the basic question of: "How the heck do I actually buy a Bitcoin???" so I'm a firm believer in the old adage that there are NO dumb questions. I can speak only for myself, but I will absolutely guarantee I won't be laughing at any legitimate question posted by anyone.

Curious? Interested? Skeptical? Drop in and give your opinion!

Yes, I'll drop by and have a look see.

Thaks for the invite and OP!

[KCIndy]

I think there are some fundamental problems with Btc and other crypto currencies. That's not to say things won't evolve in the future. I don't own any and I think there's still some growing pains coming. It's speculating right now; not investing IMHO (and I'm not here to piss in anybody's Cheerios, just throwing out my two cents).



Personally, I think you're absolutely right in that "speculating" is a much better word to use for Bitcoin and the other cryptocurrencies right now. To me, "investing" means that there is, or should be, a very high likelihood that the product in question is going to be worth more in a few years than it is right now. And in these volatile times, I can't think of very many things that fit that definition.

Even gold and silver have their price swings. I can only imagine the frustration of someone who might have sunk their life savings into gold back in 2011 when the price was topping out over $1700/ounce. Three years later? Ouch. Same thing with silver, which is still down by about 50% from those heady days. (And to all the gold and silver fans out there, please understand I'm NOT knocking gold and silver. I'm a big fan of both, and own a lot of silver myself. I'm just saying that there are NO guarantees...)

Do I think Bitcoin (BTC) is going to be worth more in a few years than it is right now? Heck yeah. But I understand I may be dead wrong. If so... well, I'll take my lumps. :)


But the return on investment isn't really my main interest.

For me, the real joy in these "altcoins" (aka cryptocurrencies) is the fact that they're completely apart from government produced and controlled currency. Used the right way, they're difficult to trace, hard to seize, and damn near impossible to tax.

I'm also fascinated with the potential for how the various cryptocurrencies are going to change and adapt due to market forces. Right now there are well over 200 different "altcoins." Obviously, not all are going to survive. In fact, I'm guessing only a handful will be around in just a couple of years. But the great thing about this is the fact that the survivors are going to be selected through sheer, raw, free market forces. The bad ideas die. The good ones prosper. There won't be any government subsidies, no "too big to fail" bailouts.

To put it bluntly, Bitcoin and the the crowd of cryptocurrencies following on its heels may be one of the last and best chances for economic freedom in an ever more oppressive world.

[torchbearer]

bitcoin helps the agorist.

[ghengis86]

To put it bluntly, Bitcoin and the the crowd of cryptocurrencies following on its heels may be one of the last and best chances for economic freedom in an ever more oppressive world.

and i would add 'freedom' in general. If one understands what makes an altcoin or even 'alt-currency' (gold, silver, copper, lead, salt, bacon, etc.) superior to government currency and its inherent problems, its a very short leap to alt-organization-of-civilization being superior to the current state. Another avenue to talk freedom and the message of liberty.

i was asked by a co-worker about what BitCoin was and I was able to talk about limited supply, gold money, dollar inflation, currency debasement, government 'funding' through debt, the Fed, etc. and why things like BitCoin have come to pass. When you have an inquisitive, attentive audience, its great to see the gears turn when you tell them, "the government funds almost half of its budget by issuing debt, which is bought with dollars conjured out of thin air. Every time the Fed buys a Treasury bill, note or bond, the value of the dollars in your wallet are worth less. Altcoins, like gold, are a way to protect yourself from the hidden inflation of a runaway government"

The conversation can go many different ways from here, but you get the idea. AltCoins are dangerous to the state, because the put the fundamental premise of government instituted currency on trial and expose it for what it is.

[muh_roads]

and i would add 'freedom' in general. If one understands what makes an altcoin or even 'alt-currency' (gold, silver, copper, lead, salt, bacon, etc.) superior to government currency and its inherent problems, its a very short leap to alt-organization-of-civilization being superior to the current state. Another avenue to talk freedom and the message of liberty.

i was asked by a co-worker about what BitCoin was and I was able to talk about limited supply, gold money, dollar inflation, currency debasement, government 'funding' through debt, the Fed, etc. and why things like BitCoin have come to pass. When you have an inquisitive, attentive audience, its great to see the gears turn when you tell them, "the government funds almost half of its budget by issuing debt, which is bought with dollars conjured out of thin air. Every time the Fed buys a Treasury bill, note or bond, the value of the dollars in your wallet are worth less. Altcoins, like gold, are a way to protect yourself from the hidden inflation of a runaway government"

The conversation can go many different ways from here, but you get the idea. AltCoins are dangerous to the state, because the put the fundamental premise of government instituted currency on trial and expose it for what it is.

Supply is half of the equation.

The second half of the equation that gives these "units" their value in the first place is a completely 100% decentralized payment processor with no middle man. Governments use economic sanctions on weaker countries for geo-political gain to bully them into giving us the resources we want.

Crypto-currencies can get around the great firewall of China and can go anywhere. There are millions of nodes out there turning on every day and no Government can anticipate who you will connect to next.

From 2012 when we were putting sanctions on Iran.
http://www.businessweek.com/articles/2012-11-29/dollar-less-iranians-discover-virtual-currency

People within Iran could still get paid by people outside of the country.

Our Government is so evil. To hurt average Iranian citizens that are just trying to make a living is something that NEEDS to end. And now it can.

[Acala]

When I first read the title I thought it said "myth-based currencies". That would be the FRN.

[FSP-Rebel]

Just for the record, I had heard about BTC as far back as late 2011 as Blockchain.info and others were big advertisers on Free Talk Live. I was bombarded everyday w/ ads and live reads for BTC related websites and I was like, "Stop talking about F-ing Bitcoin, God D-m it!" As a non-tech savvy person, I had no idea what this crap was or wtf I should care about it. But, things started to click after a year and a half and I became sold on it. After plenty of research on my own and still having not grasped some of the more technical and tedious aspects of it, fast track to now and I'm an official BTC slappy.:D

[PRB]

One thing I never understood about cryptography is, why is it hard to solve, easy to verify, and irreversible? (it's that like saying i can build a lock but I can't get the key?)

Or more specifically, how do we know the person who started the program doesn't have a backdoor to solving problems and shortcutting the proof of work mechanism?

And lastly, isn't the possibility of 51% attack, in short, the vulnerability for corruption? Or is corruption only defined as stealth actions which are too late to stop when done?

[PRB]

Just for the record, I had heard about BTC as far back as late 2011 as Blockchain.info and others were big advertisers on Free Talk Live. I was bombarded everyday w/ ads and live reads for BTC related websites and I was like, "Stop talking about F-ing Bitcoin, God D-m it!" As a non-tech savvy person, I had no idea what this crap was or wtf I should care about it. But, things started to click after a year and a half and I became sold on it. After plenty of research on my own and still having not grasped some of the more technical and tedious aspects of it, fast track to now and I'm an official BTC slappy.:D

I remember that, or maybe what I remembered was an advertised site which paid bitcoin for amazon commissions, at that time it sounded so silly I couldn't wait to see it fail (as 99% of all other internet inventions do when it comes to money)

[CPUd]

One thing I never understood about cryptography is, why is it hard to solve, easy to verify, and irreversible? (it's that like saying i can build a lock but I can't get the key?)

Or more specifically, how do we know the person who started the program doesn't have a backdoor to solving problems and shortcutting the proof of work mechanism?

And lastly, isn't the possibility of 51% attack, in short, the vulnerability for corruption? Or is corruption only defined as stealth actions which are too late to stop when done?


For the first part, imagine what you would have to do to calculate by hand an exponential, like n^3. It's simple repeated multiplication, but if you had 8, and were trying to get back to n, you would have to answer the question "8 is a power of what?"

It becomes a search problem, where the naive approach would be to start with 1 and incrementally try to answer the question. Because of other things we know, we can eliminate a lot of those numbers, like 1, since the number we were given wasn't 1. And we could eliminate numbers greater than 4 (assuming we are dealing only with natural numbers). Luckily, for our problem, the next number, 2 didn't take much work to get to. And we can easily verify by taking 2^3.

The numbers in crypto are a hell of a lot larger than 8, and brute force search has gotten better to where you only need to try prime numbers, because any composite number can be broken down to prime factors.

As you see, I'm pretty much telling you how the puzzle is structured, and it is no risk for me to do so, because I know it takes a long time to search for the solution.

[PRB]

For the first part, imagine what you would have to do to calculate by hand an exponential, like n^3. It's simple repeated multiplication, but if you had 8, and were trying to get back to n, you would have to answer the question "8 is a power of what?"

Good analogy, but how do I know the person who asked the question isn't asking what he already knows? Meaning, he has the answer, and therefore can cheat me if he wanted to by answering first, while I'm wasting my time trying to find the answer.



As you see, I'm pretty much telling you how the puzzle is structured, and it is no risk for me to do so, because I know it takes a long time to search for the solution.

Good point, but how do I know you don't have a shortcut or secret to formulating your questions so that you have an unfair advantage over everybody who "thinks" that nobody has shortcuts or advantages?

[CPUd]

Good analogy, but how do I know the person who asked the question isn't asking what he already knows? Meaning, he has the answer, and therefore can cheat me if he wanted to by answering first, while I'm wasting my time trying to find the answer.



Good point, but how do I know you don't have a shortcut or secret to formulating your questions so that you have an unfair advantage over everybody who "thinks" that nobody has shortcuts or advantages?

Because the question is out in the open. Find a natural number n such that n^3=8, which is equivalent to solving log_n(8)=3 for n. All of the standard encryption algorithms are stated just as explicitly, albeit they are more complicated.

The only way to get a guaranteed solution from scratch on a machine is to break 8 down into prime factors, which we know is a NP-complete problem. A machine can solve this example in a matter of nanoseconds, but the larger the number gets, the larger the set of possible solutions gets- and the increase is exponential.

[muh_roads]

Anybody ever take masters numerical analysis courses?

Here is some of the math behind mining and pool types...

https://bitcoil.co.il/pool_analysis.pdf

http://images.wikia.com/uncyclopedia/images/archive/b/b5/20120914120348!Exploding-head.gif

[PRB]

Cpud. But couldnt the bitcoin creator have designed the questions long in advance, as well as predictied the future questions, but lead people to think questions are new to everybody?

[CPUd]

Cpud. But couldnt the bitcoin creator have designed the questions long in advance, as well as predictied the future questions, but lead people to think questions are new to everybody?

I was really only talking about how crypto works in general. Adding bitcoin into the mix involves other elements, which are probably better explained by someone else.

[muh_roads]

Cpud. But couldnt the bitcoin creator have designed the questions long in advance, as well as predictied the future questions, but lead people to think questions are new to everybody?

The creator has a rough estimate of 900k btc. Even if I knew the questions in advance I'd never do anything to sabotage my own system and bring my near $1b value to zero.

Programmers around the world have analyzed the code thoroughly and the consensus among those that can read it say the answers are random everytime. The question varies based on the block itself combined with the number of unpredictable transactions made in a 10 minute interval.

[PaulConventionWV]

Investing in Bitcoins, is now really the best time for something like that? You have heard that its primary founder is somebody that the media cannot get a hold of that nobody knows anything about him, a virtual ghost (ironically sort of like how Bitcoins resemble the ghost of currency); another founder is avoiding the press entirely; and still yet another founder apparently just jumped from her apartment building, killing herself? Also that its primary exchange Mt.Gox filed bankruptcy a few days ago. And not to mention that odd surge in its value from April of last year, resulting in a three-quarter loss in overall value.

That's actually precisely why now is a good time.

[ghengis86]

Anybody ever take masters numerical analysis courses?

Here is some of the math behind mining and pool types...

https://bitcoil.co.il/pool_analysis.pdf

http://images.wikia.com/uncyclopedia/images/archive/b/b5/20120914120348!Exploding-head.gif

Yeah, I fancy myself to be good with them number thingys, and I got a few pages in....boom, my head went like that gif.

I do understand gold however. So for now my simple mind will have to settle for simple gold.

Now, if someone develops a picture book explanation of Btc and crypto currencies, send it my way!!

[brandon]

Regarding encryption algorithms, CPUd is right that solving the discrete log problem is the basis behind RSA type (such as ECDSA used by btc) public key schemes. But I think for an outsider looking in you need to first look at a more general view of cryptography.

PRB asked specifically about the proof of work mechanism, so I will try to address that. The proof of work algorithm is not based on discrete log or any type of public key scheme. It is based on something called a cryptographic hash function.

A hash function can be defined as H: m -> x, or in english, a function that maps any message into a fixed length string of bits. By it's very nature this function is not invertible. As a quick proof, there are infinite messages you can put into the hash function, but only a set number of outputs. So given any output from the hash it's impossible to reconstruct the input. The property that makes the hash function "cryptographic" is that given a specific hash output, it is impossible for a person to construct an input that would produce that output. The property is called collision resistance. A collision resistant hash function outputs sequences of bits that are indistinguishable from a random sequence of bits. Note, this property is only assumed of hash functions and there is no proof of it.

bitcoin uses the SHA256 hash function which outputs 256 bits. By the birthday paradox, someone would have to compute 2^128 hashes before they could generate a collision. This is currently infeasible even if all the super computers in the world were working together. Even quantum computing doesn't break SHA256. However, older hash functions, such as MD5, have been compromised in less time than brute force. MD5 is actually completely broken and collisions can be produced in a matter of seconds. SHA256 is a open peer reviewed algorithm that has so far stood the test of time. It's not likely to be broken any time soon. However, if it were to be broken there are contingency plans to switch bitcoin to a new hashing algorithm.


Disclaimer: I'm only a student of crypto and I probably said something that's not exactly correct above.

[brandon]

I don't own any bitcoins because I'm not convinced they are a viable currency. I think they're in a deflationary speculative bubble. I am absolutely fascinated with the technology. I would love for it to succeed. I do think in the next hundred years a decentralized digital currency will become extremely successful. I'm just not convinced bitcoin is it. I think the block reward decay and total supply were chosen almost completely arbitrarily. In some aspects the currency was centrally designed by one guy who is probably wrong about how it should work. I also worry about the 51% attack and worry about NSA breaking ECDSA years before the break enters the public domain.

[PRB]

Regarding encryption algorithms, CPUd is right that solving the discrete log problem is the basis behind RSA type (such as ECDSA used by btc) public key schemes. But I think for an outsider looking in you need to first look at a more general view of cryptography.

PRB asked specifically about the proof of work mechanism, so I will try to address that. The proof of work algorithm is not based on discrete log or any type of public key scheme. It is based on something called a cryptographic hash function.

A hash function can be defined as H: m -> x, or in english, a function that maps any message into a fixed length string of bits. By it's very nature this function is not invertible. As a quick proof, there are infinite messages you can put into the hash function, but only a set number of outputs. So given any output from the hash it's impossible to reconstruct the input.

In creationist terms, do we call this "Loss of information"? Much like you can't reconstruct the original photograph after you've made a thumbnail resize of it?

[PRB]

I don't own any bitcoins because I'm not convinced they are a viable currency. I think they're in a deflationary speculative bubble. I am absolutely fascinated with the technology. I would love for it to succeed. I do think in the next hundred years a decentralized digital currency will become extremely successful. I'm just not convinced bitcoin is it. I think the block reward decay and total supply were chosen almost completely arbitrarily. In some aspects the currency was centrally designed by one guy who is probably wrong about how it should work. I also worry about the 51% attack and worry about NSA breaking ECDSA years before the break enters the public domain.

I agree with you, that the reward and incentive to mine and check will decrease, soon enough that miners will have no incentive to waste their computer power for close to nothing, which in turn mean the currency cannot remain as decentralize as it's imagined to be.

[PRB]

Even quantum computing doesn't break SHA256. However, older hash functions, such as MD5, have been compromised in less time than brute force. MD5 is actually completely broken and collisions can be produced in a matter of seconds. SHA256 is a open peer reviewed algorithm that has so far stood the test of time.

Sorry, back to the entry level question again.

How can an encryption mechanism be open source but still secure? Is it like CPud said, that knowing the question is only half the story?

Is the security of bitcoin reliant on the fact that bitcoin creators didn't also create their own encryption mechanism, but relies on SHA256, which is known to be reliable and even bitcoin creators can't cheat it?

In bitcoin, when a block is solved, is it akin to 'cracking' encryption? (like if a person cracked the key to SSL 128bit encryption)?

In SSL or PGP, all you need to lose your security, is for the SSL holder to lose or give you his key, how is the same not true with bitcoin, that the creator has a master key? Is it because brand new keys and locks are generated every transaction?

[CPUd]

In creationist terms, do we call this "Loss of information"? Much like you can't reconstruct the original photograph after you've made a thumbnail resize of it?

Yes. Hashing is one-way. Passwords stored in a database (like the one here for RPF) are usually not the actual passwords, but their hashes. This way, if someone gains unauthorized access to the database, they won't get everyone's passwords. When someone logs in, the plaintext password will be run through the same hash function and compared to what is in the database.

It is possible for two different inputs to result in the same hash.

[amonasro]

I agree with you, that the reward and incentive to mine and check will decrease, soon enough that miners will have no incentive to waste their computer power for close to nothing, which in turn mean the currency cannot remain as decentralize as it's imagined to be.

This is a valid concern. However, due to the limited number of bitcoins, if it becomes adopted worldwide there's a very good chance that the decreasing block reward incentive will be enough to grow the network. Miners also receive the transaction fees. Difficulty adjusts relative to the network hashrate so that it's always marginally profitable to mine, so there will most likely be enough power to secure the network even if miners are only making tx fees.


Sorry, back to the entry level question again.

How can an encryption mechanism be open source but still secure? Is it like CPud said, that knowing the question is only half the story?

Is the security of bitcoin reliant on the fact that bitcoin creators didn't also create their own encryption mechanism, but relies on SHA256, which is known to be reliable and even bitcoin creators can't cheat it?

In bitcoin, when a block is solved, is it akin to 'cracking' encryption? (like if a person cracked the key to SSL 128bit encryption)?

In SSL or PGP, all you need to lose your security, is for the SSL holder to lose or give you his key, how is the same not true with bitcoin, that the creator has a master key? Is it because brand new keys and locks are generated every transaction?

Yes, you're correct. I'd check out the wiki on how the blockchain is stored as a merkle tree of hashes, one block on top of the other. There is no master key to the blockchain. You can generate new addresses (keypairs) for every transaction if you want, or use the same address. If you lose your private key, you don't own your coins. If you lose the public key, however, it can be regenerated from the private key. So all you really need is a piece of paper with the privkey on it hidden somewhere. There are ways to generate offline addresses that ensures your privkey never touches the internet, then send funds to it.

Miners are simply hashing away, looking for a huge random number. Like a lottery. If they find it, the network accepts it and they bundle the transactions from the last 10 minutes into the block, which is added to the blockchain. It's computationally difficult to hash fast enough to have a chance to solve/find a block, therefore when you do (or a pool does) this constitutes proof of work and is accepted as such by the majority of the network. This is a simplified explanation. It actually has to do with finding a hash of a block header that is equal to or lower than the target, a random 256 bit number.

[PRB]

Miners are simply hashing away, looking for a huge random number. Like a lottery. If they find it, the network accepts it and they bundle the transactions from the last 10 minutes into the block, which is added to the blockchain. It's computationally difficult to hash fast enough to have a chance to solve/find a block, therefore when you do (or a pool does) this constitutes proof of work and is accepted as such by the majority of the network. This is a simplified explanation. It actually has to do with finding a hash of a block header that is equal to or lower than the target, a random 256 bit number.

Are you basically saying, that rather than solving, it's more like brute force guessing, and getting lucky, those with more computational power merely have a higher chance of getting to the right answer first (do all miners approach the problem at the same angle?).

"Proof of work" then, is just the assumption that it takes time to find the answer, so the first person who finds it wins, regardless of how it was found, as long as it can be verified. Is that right?

Now for another question, I read here
www.washingtonpost.com/blogs/the-switch/wp/2013/11/07/no-bitcoin-isnt-broken/

"Eventually, one of the honest nodes will find a lottery ticket of its own and present it to the rest of the network for validation. The moment that happens, an evil node will present its own winning ticket to the network."

What's the incentive for the evil node to wait and hoard, while risking losing his hard earned reward? Isn't he just wasting the honest node's time? How does that benefit himself?

[muh_roads]

For the record, it's not an exact 10 minutes on the nose. Sometimes it is 8 minutes, sometimes it is 12.0286 minutes. The difficulty tries to adjust itself every 2016 blocks to keep block solving as close to 10 minutes as possible relative to the growing hashrate of the network.

[PRB]

Yes. Hashing is one-way. Passwords stored in a database (like the one here for RPF) are usually not the actual passwords, but their hashes. This way, if someone gains unauthorized access to the database, they won't get everyone's passwords. When someone logs in, the plaintext password will be run through the same hash function and compared to what is in the database.

It is possible for two different inputs to result in the same hash.

If these are 2 actual passwords, but same hash.

100D
PRBisAnIdiot

Does that mean in a collision, they can be used interchangeably as actual passwords?

If so, doesn't that mean we don't need to reverse reconstruct the input, as long as we know what can collide to create the same output?

[2young2vote]

I currently do not own any crypto/digital currency. I have nothing against it and I understand why it can be useful, but I have no incentive to own any at this time. Until one or more becomes standard currency on the internet to the same extent that Visa and Mastercard are the standard accepted credit cards for physical in-store purchases, I do not have any reason to own any right now - especially considering how risk averse I am at this point.

[CPUd]

If these are 2 actual passwords, but same hash.

100D
PRBisAnIdiot

Does that mean in a collision, they can be used interchangeably as actual passwords?

If so, doesn't that mean we don't need to reverse reconstruct the input, as long as we know what can collide to create the same output?

If the hash function by itself was used, then input strings that cause a collision can be used interchangeably. But with something like MD5, you are probably not going to find many collisions with strings shorter than 64 bytes, which is a lot longer than the average password (even if it has a 32-byte salt).

Many web based apps will also use a salt- often a hardcoded string in the source code, with a length of around 32 bytes. When installing one of these apps onto a server, part of the procedure is to generate the salt, so known collisions can't be used globally. Unless, of course, the attacker knows the salt.

[PRB]

If the hash function by itself was used, then input strings that cause a collision can be used interchangeably. But with something like MD5, you are probably not going to find many collisions with strings shorter than 64 bytes, which is a lot longer than the average password (even if it has a 32-byte salt).

Many web based apps will also use a salt- often a hardcoded string in the source code, with a length of around 32 bytes. When installing one of these apps onto a server, part of the procedure is to generate the salt, so known collisions can't be used globally. Unless, of course, the attacker knows the salt.

Is the point of hashing mainly to passwords from being stolen as plain text?

Is salt basically an additional encryption on the hash?

[brandon]

PRB, the concepts you are asking about are complex and take a lot of study to understand. There are various free online cryptography courses you can take that will help you understand. I took this one and highly highly recommend it: https://www.coursera.org/course/crypto

Hashing is used for a lot of things. One example is it's use in message authentication codes. If I send you a message and a hash of that message, you can detect if the message was altered by a 3rd party if it doesn't match the hash. (This is actually a giant simplification of MACs)

[muh_roads]

PRB you are asking things that I am sure have already been asked before. You have this strange pursuit so you can say "Ah HA, gotcha". I do admire the pursuit of knowledge though. If there was anything devious within BTC it would've been brought forward a long time ago by top people in these fields of math, cryptography, and encryption. If there was anything devious there the value wouldn't be anywhere near it is today.

It's like metals. I don't need to fully understand how they mine for gold and silver to trust it. It is how you protect & secure your crypto is all that really matters. Just like securing your own metals.

[mczerone]

Good analogy, but how do I know the person who asked the question isn't asking what he already knows? Meaning, he has the answer, and therefore can cheat me if he wanted to by answering first, while I'm wasting my time trying to find the answer.

Good point, but how do I know you don't have a shortcut or secret to formulating your questions so that you have an unfair advantage over everybody who "thinks" that nobody has shortcuts or advantages?

There was actually just a problem uncovered with cryptographic schemes in general:

To generate keys, an encryption program needs to generate large primes randomly, so that they can't be guessed by crackers. A number of "randomizers" were giving results so that something like 5% of private keys had primes that were commonly used, and so could be broken by "guess and check" methods within seconds.

The problem was with proprietary encryptors that were not using vetted randomization subroutines. The encryption theory had no problems, but once you start using the same keys again and again, you run into problems.


But most major crypto-currencies use open-sourced programming to generate keys, and they will be continually vetted for problems by people with a large stake in the currency. They know how to properly run high-entropic randomizers, and they know how to prove that to people auditing their code.

By being open-source and free-software, BTC and other currencies are basically asking for other programmers to find their "secret backdoors." Because of this, it ensure that none can be put into the code. Because there's no individual scheming programmer or cartel of privileged programmers, anyone could find a hypothetical advantage, exploit it, and ensure that programmers change the code before much damage is done.

Basically you're worried about central planning problems, when the whole blockchain was devised to avoid being centrally planned.


As to your earlier question about the 51% problem: We've already seen 1 mining pool approach 51%. The result was that other miners joined the network to block them, and some of the members of the large pool actually voluntarily left to form a second pool to avoid being part of the "problem."

[PRB]

PRB you are asking things that I am sure have already been asked before. You have this strange pursuit so you can say "Ah HA, gotcha". I do admire the pursuit of knowledge though. If there was anything devious within BTC it would've been brought forward a long time ago by top people in these fields of math, cryptography, and encryption. If there was anything devious there the value wouldn't be anywhere near it is today.

It's like metals. I don't need to fully understand how they mine for gold and silver to trust it. It is how you protect & secure your crypto is all that really matters. Just like securing your own metals.

I am genuinely interested, and admit I am ignorant.

Here's another question.

Why are mining equations easy to solve, but hacking a person's wallet private key not? Are they different encryption mechanisms?

[ghengis86]

PRB you are asking things that I am sure have already been asked before. You have this strange pursuit so you can say "Ah HA, gotcha". I do admire the pursuit of knowledge though. If there was anything devious within BTC it would've been brought forward a long time ago by top people in these fields of math, cryptography, and encryption. If there was anything devious there the value wouldn't be anywhere near it is today.

It's like metals. I don't need to fully understand how they mine for gold and silver to trust it. It is how you protect & secure your crypto is all that really matters. Just like securing your own metals.

My only quibble is that alchemists haven't found a way to turn lead into gold, whereas some man or men can turn math into crypto currency. It's actually pretty simple to understand gold mining; Btc, not so much.

Granted, there's the open source protection and whatnot. I think it's just inherently easier to trust gold than Btc at this point in time. The future? It's bright and who knows !

[muh_roads]

Why are mining equations easy to solve, but hacking a person's wallet private key not? Are they different encryption mechanisms?

It isn't easy to solve.

http://www.forbes.com/sites/reuvencohen/2013/11/28/global-bitcoin-computing-power-now-256-times-faster-than-top-500-supercomputers-combined/

[brandon]

SHA256 is easy to calculate. You could even do it by hand if you wanted to. That article is weird. Why is he measuring mining in FLOPS? As far as I know there are no floating point operations at all in SHA256. Can ASIC miners even do any FLOPs?

[muh_roads]

SHA256 is easy to calculate. You could even do it by hand if you wanted to.

lol

ok I'd like to see you calculate 2^256 by hand. Hell I'll even make it easy for you and let you count in order starting with 1. You'd poop out.

I will agree with you that the theory of quantum computing could break ECDSA, but we aren't there yet for QC to be a threat. And for the record, QC would also be a threat to every cryptographic financial instrument on the planet if it came along, not just Bitcoin.

Luckily Bitcoin was designed to be upgradable if QC does become a real thing. I think we're safe for at least another 10 years. At least.

http://en.wikipedia.org/wiki/Post-quantum_cryptography


Can ASIC miners even do any FLOPs?

No it is an integer operation. I think 1 intop = 2 flops. So cut his values in half if one is emulated on top of the other.

[PRB]

lol

ok I'd like to see you calculate 2^256 by hand. Hell I'll even make it easy for you and let you count in order starting with 1. You'd poop out.

I will agree with you that the theory of quantum computing could break ECDSA, but we aren't there yet for QC to be a threat. And for the record, QC would also be a threat to every cryptographic financial instrument on the planet if it came along, not just Bitcoin.

Luckily Bitcoin was designed to be upgradable if QC does become a real thing. I think we're safe for at least another 10 years. At least.

http://en.wikipedia.org/wiki/Post-quantum_cryptography



No it is an integer operation. I think 1 intop = 2 flops. So cut his values in half if one is emulated on top of the other.

2^256?

2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2

give me another 20 minutes

[muh_roads]

2^256?

2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2

give me another 20 minutes

No. Write it out by hand in order...1...2...3...4...etc until you reach 1.1579208923731619542357098500869e+77

No short cuts...that was the challenge he offered. And allowing you to go in order is just me being nice. No sequence is in order.

We're just talking about hashing here.

http://cynic.me/wp-content/uploads/2013/11/bitcoin-laws-of-the-universe.jpg

[PRB]

No. Write it out by hand in order...1...2...3...4...etc until you reach 1.1579208923731619542357098500869e+77


It's taking longer than I thought. But my head hasn't quite exploded yet :)

I thought your original challenge was to calculate it out, not count one at a time. We know that 256^2 is a big number and you can easily get 64,536 in a few minutes, do I need to count 1 to 64,536 to prove to appreciate how big the number is?

2^256?


2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2
=((17,153,635,286,310,109,696)^2)^2

I'll continue this tomorrow, I haven't checked with a calculator at all (I hope when I'm done, I don't get the fate of the guy in Pi)

[PRB]

http://cynic.me/wp-content/uploads/2013/11/bitcoin-laws-of-the-universe.jpg

Sure, I couldn't count to 2^256 before the computer runs out of energy, but why should I be the standard for comparison? Computers CAN count 100s or 1000s of numbers at a time.

Until computers are built from something other than matter, impossible. Occupy something other than space? Hardware cannot, but software can.

[PRB]

duplicate post.

[PRB]

=((17,153,635,286,310,109,696)^2)^2


17,153,635,286,310,109,696 = x
2x= 34,307,270,572,620,219,392
3x= 51,460,905,858,930,329,088
4x=
5x= 85,768,176,431,550,548,480
6x=102,921,811,717,860,658,176
7x=120,075,447,004,170,767,872
8x
9x=154,382,717,576,790,987,264

[PRB]

And allowing you to go in order is just me being nice. No sequence is in order.




Is your point basically, that brute force attack against a 256 bit key would require 2^256 tries?
It's one thing to try, and another to first know what to try.
Can't computers, in advance, list out all 2^256 keys, divide them up, and when needed, attack one division at a time between multiple computers?

[ghengis86]

It's taking longer than I thought. But my head hasn't quite exploded yet :)

I thought your original challenge was to calculate it out, not count one at a time. We know that 256^2 is a big number and you can easily get 64,536 in a few minutes, do I need to count 1 to 64,536 to prove to appreciate how big the number is?

2^256?


2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2
=((17,153,635,286,310,109,696)^2)^2

I'll continue this tomorrow, I haven't checked with a calculator at all (I hope when I'm done, I don't get the fate of the guy in Pi)

Lol. That movie was crazy! My tip to you; don't look at the sun and stay away from power drills!


Sorry, back to the nerdery

[brandon]

lol

ok I'd like to see you calculate 2^256 by hand. Hell I'll even make it easy for you and let you count in order starting with 1.

I don't understand what calculating (or counting???) to 2^256 has to do with sha256. I think you might be confused about what sha256 is.

Anyway, calculating any power of 2 by hand is very simple, you just double the number repeatedly(Not that this has anything to do with sha256...). I don't know why you keep posting that terrible infographic.




No it is an integer operation. I think 1 intop = 2 flops. So cut his values in half if one is emulated on top of the other.

No there is no comparison. That's like saying 2 apples = 1 orange.

[muh_roads]

Sure, I couldn't count to 2^256 before the computer runs out of energy, but why should I be the standard for comparison? Computers CAN count 100s or 1000s of numbers at a time.

Until computers are built from something other than matter, impossible. Occupy something other than space? Hardware cannot, but software can.

You weren't actually supposed to count that. My point was that the amount of character combinations is immense for a human. I was just arguing with the guy who thinks he can break bitcoin transactions that move thru the system like droplets of water coming out of a shower head and break it by hand.


I don't know why you keep posting that terrible infographic.

It's refering to combinations, dude. Giving neg rep over something you don't understand is kinda immature, mang.


No there is no comparison. That's like saying 2 apples = 1 orange.

You should go read up on CUDA applications. It is an example of emulating integer operations.

Roll it back, I will agree that flops wasn't the best measurement in the article.

[CPUd]

Is your point basically, that brute force attack against a 256 bit key would require 2^256 tries?
It's one thing to try, and another to first know what to try.
Can't computers, in advance, list out all 2^256 keys, divide them up, and when needed, attack one division at a time between multiple computers?

In computer science, when someone talks about the complexity of a problem, it is almost always a worst case scenario. Like if the problem is a list sorting problem, assume the list is sorted in reverse order from the desired output. And brute-forcing 2^256 keys, the correct key would be the last one tried. The reason they do this is to get an upper bounds for a class of problems.

Arithmetic operations have to be wired into circuitry of the CPU. The most basic one is an incrementer. To add 2 numbers, it would literally take one operand and the second operand would be the number of 1's to increment the first. Subtraction could use the same circuit by flipping the "sign" flag to make the net result a decrement. Multiplication could also use the same circuit by constructing a double-nested loop of increments. Division is more complicated, because the result is stored in 2 locations- the quotient and the remainder. Modern CPUs have more operations wired in hardware, but arithmetic on a machine still has simple counting at its core. Floating-point operations have their own circuits to store and do arithmetic.

Staying at this low level, suppose we want to add 2 integers at locations c and d, and store the result at d. The number at c could be the counter, and the number at d could be incremented based on the counter value. The problem is, in a 32-bit system, c, d and the other registers are only designed to be able to represent 2^32 different numbers. So for integers that could be positive or negative, the range would be [-2147483647, 2147483647], and for unsigned ints [0, 4294967295]

If you wanted to take a set of keys of size 2^256 and split them across multiple machines, let's say each machine gets 2^64 (or 18446744073709551615) keys. That is a good bit of work; might take a few years on a really fast machine. Then you would need 6.28 x 10^57 machines.

[FSP-Rebel]

This Fun intro to Cryptos just took a turn for the worst and now the average, interested party is locked out.:p

[muh_roads]

This Fun intro to Cryptos just took a turn for the worst and now the average, interested party is locked out.:p

Yeah we sure geeked the fuck out of the place didn't we? lol

I'm trying to keep my responses simple because most don't give a shit about this e-penis dick measuring. If arm-chair analysts really have a problem with BTC on a technical level they should take it to bitcointalk.org where they will be schooled appropriately by those that have degrees in this stuff.

I just wanted to clear up the economic concerns with some fun debate.

[PRB]

It's taking longer than I thought. But my head hasn't quite exploded yet :)

I thought your original challenge was to calculate it out, not count one at a time. We know that 256^2 is a big number and you can easily get 64,536 in a few minutes, do I need to count 1 to 64,536 to prove to appreciate how big the number is?

2^256?


2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2
=((17,153,635,286,310,109,696)^2)^2

I'll continue this tomorrow, I haven't checked with a calculator at all (I hope when I'm done, I don't get the fate of the guy in Pi)

So here I am

=((17,153,635,286,310,109,696)^2)^2
=(294,247,203,535,743,319,043,768,834,135,553,212, 416)^2

Can somebody verify for me if (17,153,635,286,310,109,696)^2 is 294,247,203,535,743,319,043,768,834,135,553,212,41 6 ?

Or more importantly, is 2^128 = 294,247,203,535,743,319,043,768,834,135,553,212,41 6 ?

I did it entirely by hand, pencil and paper. Lots of paper, lots of mess, no calculators, and obviously no computer science background.

[PRB]

If you wanted to take a set of keys of size 2^256 and split them across multiple machines, let's say each machine gets 2^64 (or 18446744073709551615) keys. That is a good bit of work; might take a few years on a really fast machine. Then you would need 6.28 x 10^57 machines.

Let me try to digest this first.

So if a key is 100bits of information a kilobit holds 10 keys?
18,446,744,073,709,551,615 or about 20,000,000,000,000,000,000 keys would occupy 2,000,000,000,000,000,000,000 kilobits
or 2,000,000,000,000,000,000 megabits
or 2,000,000,000,000,000 gigabits
or 2,000,000,000,000 terrabits?

Meaning for just 2^64 keys, you need 2 million people each with 1 million hard drives holding a terrabit each?

[Madison320]

I'm betting Bitcoin will fail because it is not a reliable store of value.

[muh_roads]

I'm betting Bitcoin will fail because it is not a reliable store of value.

Future volatility decreases as a new market grows. They call it "sticky" economics.

http://en.wikipedia.org/wiki/Sticky_%28economics%29

Also when the merchants expenses are also transacted in BTC, not just what they sell, the exchange rate won't be looked at as much. To the point of being nearly irrelevant.

It's kinda like day-trading alts. I don't even look at the USD spot price except to anticipate a possible ratio drop between two crypto's. My primary goal is to grow the BTC tally. Merchants would look at it the same way if everything they did was in BTC.

[CPUd]

Let me try to digest this first.

So if a key is 100bits of information a kilobit holds 10 keys?
18,446,744,073,709,551,615 or about 20,000,000,000,000,000,000 keys would occupy 2,000,000,000,000,000,000,000 kilobits
or 2,000,000,000,000,000,000 megabits
or 2,000,000,000,000,000 gigabits
or 2,000,000,000,000 terrabits?

Meaning for just 2^64 keys, you need 2 million people each with 1 million hard drives holding a terrabit each?

Perhaps we should just give them each 1 key to start with, and let them try keys sequentially. ;)

[PRB]

Perhaps we should just give them each 1 key to start with, and let them try keys sequentially. ;)

lol. sorry

[nayjevin]

So ultimately bitcoin is majority rule by participating programmers? And if they integrate something fatal folks need to switch to another fork or whatever?

[muh_roads]

So ultimately bitcoin is majority rule by participating programmers? And if they integrate something fatal folks need to switch to another fork or whatever?

That is correct. If something goes wrong they back-peddle to a previous version and fork to a new branch including the blocks that were solved on the other path.

This happened February 2013 and it was fixed in a matter of minutes after the problem was discovered. The previous time before that was 2010 sometime. They gave everyone ample time (3 months) to upgrade to the new fork. All major pools were ready to go in just a few hours.

The clients can broadcast when a new version is ready immediately.

[Elwar]

So ultimately bitcoin is majority rule by participating programmers? And if they integrate something fatal folks need to switch to another fork or whatever?

This happened once last year. A flaw was found, it was discovered quickly, fixed and the blockchain was forked. Only transactions that took place over a short period of time were affected and there wasn't much interruption.

If the NSA were to break the cryptography, it would be discovered, the cyptography upgraded and the blockchain forked.

If it were not in everyone's best interest, it would not fork.

[Elwar]

bump

[PaulConventionWV]

It's taking longer than I thought. But my head hasn't quite exploded yet :)

I thought your original challenge was to calculate it out, not count one at a time. We know that 256^2 is a big number and you can easily get 64,536 in a few minutes, do I need to count 1 to 64,536 to prove to appreciate how big the number is?

2^256?


2^256 = ((2^8)^2)^16
=(256^2)^16
=64,536 ^ 16
=((((64536)^2)^2)^2)^2
=((4,141,694,736^2)^2)^2
=((17,153,635,286,310,109,696)^2)^2

I'll continue this tomorrow, I haven't checked with a calculator at all (I hope when I'm done, I don't get the fate of the guy in Pi)

I'm pretty sure it's 2^256, not 256^2. By comparison, there are approximately 10^80 atoms in the universe. I don't think you're going to finish counting.

[CPUd]

Yeah, any time you see something that refers to n bits, there will be 2^n possible values.

[Warrior_of_Freedom]

I'm betting Bitcoin will fail because it is not a reliable store of value.

dollars aren't either so pick your poison

[PRB]

dollars aren't either so pick your poison

sorry, the dollar hasn't failed for over 100 years.

[Warrior_of_Freedom]

sorry, the dollar hasn't failed for over 100 years.

it fails me every day the money i hold is worth less

[buenijo]

sorry, the dollar hasn't failed for over 100 years.

The U.S. currency system has shifted at least twice despite the currency unit consistently being called "dollar". The most recent was 1971. The system in place after that failure has been maintained partly with political and military interventions, and increasingly the latter in recent years.

The vagaries of language often facilitates deception. Using the same term to describe different things obscures the characteristics that distinguish them. Poor language both suggests and reinforces poor thinking. See Orwell.