Video - Bitcoin Programmable Money - Nov 2014
A talk in Sydney Australia on the programmability of bitcoin. Want to hear more about Bitcoin and open blockchains?
ANDREAS ANTONOPOULOS: Bringing both of my hands I increase my speaking capacity by at least 60%. See, I can – I use my hands to speak. That’s what Greeks do so I don’t like holding those microphones. This is much more comfortable for me even though it looks really, really silly so bear with me on that.
I’m really excited to be here. I’ve been looking forward to speaking in Australia for a while now and I understand we have a great crowd here today a lot of people from Atlassian, our host. Who’s here from Atlassian? All right.
Well, first of all thank you so much for hosting us at this wonderful location. I have to say I’m a fan of your work. I’ve used your crowd authentication platform, I’ve used confluence and most importantly I’ve wrote the Mastering Bitcoin book using Atlassian source tree to manage the Git Repository where the book is being written. It was all written on GitHub and on (0:01:00) Riley’s Git server and I used your product to do all of my book commits. Thank you so much for building a wonderful product that was great to use. That’s your little plug but I mean it honestly, I appreciate that software.
I’d like to thank the host who have done so much work to bring this together. It’s not easy to organize this events and make them run smoothly. The College Crypto Network, of course, that organize this wonderful Bitcoin associations across university campuses, the Bitcoin Association of Australia and, of course, our host and sponsors CoinJar who brought us here. So, thank you again to all of those who made this happen.
Now, I sometimes talk about general topics and (0:01:49) philosophically about Bitcoin but I thought this audience is a bit more technical. I know there’s a lot of people who are into Bitcoin and understand Bitcoin fairly wall so I thought I would do a slightly more technical seminar and talk about a specific topic of interest to me slightly extending from the things that Pamela Morgan discussed on smart contracts.
What I want to talk about is programmable money used as escrow in order to achieve some of the consumer protections that people claim you can’t do with Bitcoin such as credit card chargebacks. So I want to talk a bit about that concept and show you some of the ways in which programmable money can be used to do many exciting things for consumers in a much better way than you can with credit cards. So, let’s start by setting the stage.
Today, most consumers are familiar with the concept that when they make a credit card transaction they are protected by the credit card company and they’re protected because if something goes wrong they can do a chargeback, right? How many of you have actually tried to do a chargeback on your credit card? Okay. And how many of you actually succeeded in doing that chargeback? Just about half, I think from the sample I saw before.
Interestingly enough when there’s fraud involved like clear evidence of fraud, like your card is stolen, if there’s no signature, if the transaction wasn’t actually made by you chargeback works. But if it’s more subtle like a breach of contract or more subtle form of fraud chargeback usually doesn’t work. What happens is the credit card company will say “Well, look here’s your signature. Sorry.” I’ll give you an example, I was traveling to Chicago and I took a taxi from the airport and I got into the taxi and I wasn’t paying attention, I was busy, it wasn’t a city I know very well I’m not familiar with it and, you know, when you travel a lot, you know, how many of you’ve been robbed by a taxi driver in more than one country? Yeah. So, by the time I got there I noticed there was something interesting I was running on the (0:04:04) level 3 which apparently is only for suburbs after midnight, it’s the middle of the day in Central Chicago and I was paying three times the going rate. By the time I got there a ride that should have charge $20.00 was charging me $65.00. Now, I didn’t know any better, I didn’t know what the cost was there and I was in a hurry and busy so I pay it. And then I found out later that I had been thoroughly defrauded, I mean this is totally illegal especially with a licensed service like taxi you think you’d have recourse. I asked Visa for a chargeback and they said “No, there’s your signature.” This was a valid transaction but I was defrauded but this was totally illegal. No, never mind. Here’s your signature. You don’t get a chargeback.
The idea that chargebacks as a solution that protects consumers is really oversold. But what does it do for merchant?
Now, merchants when chargebacks happen are responsible, liable for the money. Even if they’ve actually deliver the product or service chargebacks are really difficult for merchants because what it means is that for up to 30 days after a charge has been credited to their account it can be withdrawn again and pay it back to the consumer in the case that fraud is suspected and in many cases the merchant doesn’t see this coming, suddenly just money just jumps out of their account and this happens all the time.
Most of the time it happens because the cards that were presented to them were stolen or the identities were forged or they were fake credit cards or something like that happen. Basically identity theft is one of the main reasons why merchants faced chargebbacks. Wow, what a great, great way to shift the entire problem of fraud onto the merchants by the credit card companies because that way they don’t have to pay for any of these frauds. Now, the merchant has already delivered the product but now they don’t get paid so the entire problem is shifted to the merchant. If there’s identity theft and you get presented a fake card you pay for that mistake even though it’s not your mistake as a merchant, even though you have no way of controlling fraud under those circumstances, even though the fraud is caused by the fundamental broken design of credit cards where each time you do a transaction you give the access keys to the merchants and every intermediary in the chain so that they can be stolen en masse from companies like Target and Home Depot and every other company out there. As I said there are two types of companies those that had their credit card databases hacked because they failed to secure them and those that will have their credit card databases hacked because they will fail to secure them. It’s a losing game.
The idea of storing shared keys in massive numbers in a database and then guarding that is ridiculous, it’s broken by design. That will always fail and there’s nothing you can do to fix that system because the very nature of a transaction that is based on a pull where the credit card pulls money from your account and where the access token is the thing you share. Now, any of you who are programmers know that is a stupid security model, right, to use the technical term.
So, what credit card companies do is they have the system that is broken by design. It was designed in the 1950s starting with the Diners Club and at first people didn’t believe that this plastic traveler’s checks really should be honored, they wanted real hard cash instead which is kind of ironic now because now they say Bitcoin isn’t really money. But, credit card started in the ‘50s and really nothing much has advanced since then. Now, here in Australia you have Chip and PIN a few other little enhancements that somewhat improved on the problem but don’t solve the fundamental design issue. And the fundamental design issue is that this is a system based on extending credit and then with this exits token allowing a merchant to deduct or debit your card or your account. That doesn’t change whether you use a Chip and PIN or a traditional credit card. What it does however is it makes the transactions hilarious when Americans are involved because I show up here and I’ve got an American card and we don’t have any chips, no, sir. We have pieces of plastic with magnetic stripe on the back, you know, old style technology, 1980. So I’m rocking this piece of plastic and I take it to merchant in Australia, they take it and they confidently wave it over the card (0:08:38) that’s a great incantation, yeah. Give me magic money oh, card.
It reminds me of this scene in Star Trek where they go back in time and they end up on earth in the ‘90s and Spock goes into a computer store, I think he was Spock and he picks up a mouse and goes “Computer” and that’s how I feel when they use my card here and they wave it, nothing happens and then I have to explain to them that really what they need to do is they need to swipe the card and then they can’t really a PIN because the ACH network isn’t connect. They need to print out a piece of paper which I will sign and after that we will send a telegram so that the money can be load onto a steam ship that will travel across the oceans and – or something like that, yeah. It’s an interesting experience.
So, credit cards are broken by design and they’re technology that is ancient was never designed for the online world yet there’s this persistent myth that they offer this consumer protection chargeback which doesn’t really work for consumers because you won’t actually get your chargeback unless it’s a case of extreme identity theft in which case it wasn’t your fault, it was a broken design. And it hurts the merchants because they end up paying for all of the fraud and for that privilege of shifting the burden of all of the fraud on the merchants they get charged 3% on every transaction. You know, I mean really it is a brilliant scheme for making money but not really a brilliant scheme for paying and serving consumers.
There’s this persistent myth that because Bitcoin transactions are irreversible somehow that exposes consumers to fraud and risk that you don’t have in the credit card space. So I’ve been hearing this for a while and I’ve been trying to rebut this argument and debunk this argument and I thought tonight would be a good opportunity to go through this and talk about all of the cool things we can do with programmable money to solve that exact problem. All right.
So, how many of you here have heard of the concept the smart contracts before today? All right, great.
So, smart contracts I would say are mostly overhyped for what they can deliver today. But one of the key components as smart contracts is this idea that a transaction is not just a ledger entry that say “From Bob to Alice: Pay five Satoshis,” right? What a transaction is in Bitcoin is a recording of an ownership script or encumbrance that tells the system how that money can be redeemed.
So, in a traditional transaction someone actually tweeted this, just a few minutes ago I saw someone tweeted the (0:11:30) op dupe op hash 160 op check sig verify up on there which is a standard script of a Bitcoin transaction.
So, in a Bitcoin transaction what you do is you assign a script to the recipient of that money and you say in order for this money to be redeemed this script has to be validated, this script has to compute, compute to a value of true, in simple terms. And what that script says is present a public key and corresponding signature to prove ownership of the keys that correspond to this Bitcoin address. I, this Bitcoin address owner, can redeem te money. So the most simple script, the most simple payment from Bob to Alice is basically says “This money is encumbered for whoever can produce a private key signature and a public key that corresponds the Bitcoin address that corresponds to Alice.” It’s a very complicated system. And why do you need all that complexity? Why couldn’t you just write Alice’s public key and there and then, you know, transaction is easy to verify? And the reason is that then you can only have one type of payee and that’s the public key. But with the scripting language you can create all kinds of complicated transaction contracts which tell you what the conditions to redeem that transaction are and this a fort-like stack-based (0:12:51) complete language that can be incredibly powerful but most of the features in the language were disabled in the first saturation of Bitcoin because what we know about Satoshi is a brilliant scientist very interesting grasp of economics and game theory not a very good programmer. The first saturation of the Bitcoin client was a mass (0:13:12) jumble of code, it was a hairball (0:13:15) and pretty much the first year of development if you look at the Git repository commit you will see that it was mostly pulling stuff out and disabling big chucks of the code that had all kinds of nasty bugs in them. Part of the clean up in the first year was fixing this language to make it a lot more secure and robust because it had a few bugs in it. But in the last two years a lot of these features have started to be turned on again and so in January of 2012 we saw the experimental introduction of the check multi-sig verify system for multisignature transactions. It was actually a competition between two different competing standards for that and if I remember correctly those are Bitcoin Improvement Proposal 16, Bitcoin Improvement Proposal 17. It was put a vote by the miners and the miners decided to support, if I remember correctly, BIP 16 which is the check multi-sig verify OP code. What that allows you to do is set up an M-of-N multisignature. And an M-of-N scheme means that you have N declared signatories of which a quorum of M can authorize payment.
Now, in the code what this looks like is a script that says you require, for example, two of the following three keys to present signatures in order to verify this transaction. It is a very simplistic type of script but it is extremely powerful because one of the cool things that happens in the scripting language is that you can take all of that script and simply present it as a fingerprint called the pay to script hash address and then you can make the payment to that and hide all of the complexity and in that script you can put very interesting things. For example, you could say this can be redeemed by anyone who does a two-of-three multi-sig or one-of-five multi-sig and a one-of-two multi-sig. So you can do if, and, or structured conditional statements.
You can also combine it with many other criteria in the system. So, for example, one of the features of Bitcoin transaction is a system called nLockTime, so the nLockTime field with the transaction can specify a time condition which says not to be redeemed on or before date X of block X. If you give a number that’s less than four billion it’s a block number, if you give a number that’s more than four billion it’s a Unix epoch milliseconds since January 1, 1970 timestamp. So it’s a really cool feature because you can say here’s a transaction and it works like a posted dated check. It can be redeemed on or after either block 375,000 or it can be redeemed after the specific date. And the granularity of the timestamp won’t be met 100% because it’s a decentralized network and the time synchronization is not accurate but any case you’re doing block confirmations every 10 minutes. So more or less somewhere in that 10-minute range it’s going to land in a block. The note that includes it in the block will pass the timestamp it will accept this as a valid transaction. All right, a lot of technical gobbledygook.
How many people have I lost so far in the audience? Anybody who’s like really confused about this? Okay, a couple of people. All right. So, let me wrap it all up and explain it in simple English.
You can say that in order to redeem money from a specific address it takes more than one signature or it takes a specific time delay or both and you can combine all of these conditions together. So, now let’s look at how we can use that in the real world. How we can use that to do automated escrow, time-lock escrow and automated dispute resolution in some cases in order to reintroduce consumer protection mechanisms in a programmatic way right into this currency in a way that’s never been possible and that in fact far more powerful than credit card protections and far more flexible than credit card protections. All right, let’s do a quick compare and contrast.
First of all if you buy something on eBay and you use PayPal whose arbitration rules are you using?
MAN #1: PayPal’s.
ANDREAS ANTONOPOULOS: PayPal’s, great. And what if you use Visa card to coffee at a coffee shop?
ANDREAS ANTONOPOULOS: Visa’s. Great, fantastic. And if you buy your tickets with American Express you’re using American Express Dispute Resolution Mechanism so your chargeback will be subject to the rules of the provider of the payment network. So, the transport protocol defines the legal mechanism by which you get charged back. Does that make sense to anyone? No? That’s ridiculous why would it be so? Why can’t I buy something on eBay and say I want to use American Express Dispute Resolution? Why can’t I buy something with Visa and say I want to use PayPal’s rules? Because the dispute resolution mechanism is tied inexplicably and inexorably to the payment network. Well, here comes Bitcoin and it is a neutral transport payment protocol that is independent of the dispute resolution mechanism and here we can do something really interesting. I can choose on a transaction by transaction basis who I use for dispute resolution. I can include third-party escrow on-demand and in fact I can do this programmatically in a way that I can open up an entire market for arbitration providers, arbitration providers that use perhaps commercial arbitration rules that use the US judicial court or that use algorithmic arbitration for certain cases. Let me give you an example.
I could have a script that redeems a transaction only if DHL has provided a deliver receipt confirmation for a specific tracking number. So now, my transaction is tied to deliver of a product so I can buy a flat-screen plasma TV and then make sure that the merchant only gets paid if the third signature from DHL comes in that say that the package was actually delivered. And now I have a very simple mechanism of both giving the money to the merchant when the package is delivered but also dealing with any problems. So here I am now doing a transaction where DHL is my third-party, they provide one of the additional signatures. PayPal, Visa, American Express could offer arbitration services completely transparently on this network in such a way that their signature is the one that decides whether the merchant gets the money or whether the consumer gets the refund because something went wrong. And so, now I can essential escrow transactions and choose who is going to resolve the dispute for me. Is this making sense?
So, not only can I do escrow and dispute resolutions just like I did with credit cards but instead of that being tied specifically to the network I’m using which makes no sense the merchant and I can decide what we use. Now, most of the time what that means is the merchant will pick someone and you probably are going to have to go along with it unless you can select from a dropdown list. In a transaction that’s cuppa coffee it might not matter. if you’re doing a transaction that involves buying a flat-screen TV just like today you have a choice of three different insurance providers for example, in order to insure that against warranty defects, manufacturing defects, you know, three years plus you could select an arbitration provider. But it gets even better than that because these transactions can be created in advance with a time-lock. So here’s another example. I can do a transaction where I put the money in escrow and I sign my part of the transaction but it has nLock on it. Now that transaction is valid and the merchant can cash it in by signing their part of the transaction but only after seven days has elapse or 14 days, or 30 days whatever time-lock I want to put on it, right? So the merchant knows the money sitting there and they know that all it takes is for the time-lock to elapse and they just add their signature and cash the money. That gives me 30 days to do arbitration. If in those 30 days the product fails, it doesn’t get delivered I have a dispute I go to the third-party and I say “Mmm-mm, this deal isn’t working. I need you to issue a transaction with your signature and my signature that undoes this (0:21:45)” and they give me a refund transaction.
In Bitcoin terms what that does is it double spends the transaction inputs on the original transaction essentially releasing the escrow and sending it back to me. What this means is that you have an escrow system where if everything goes well you don’t need to do anything. Thirty days later “Boom” transaction executes, the merchant gets their money. If something goes wrong the consumer has the ability then to jump in request arbitration with the third-party provider and execute reverse refund transaction. Now, all of these (0:22:15) complicated if you think about what I just did. I just introduced (0:22:19) Bitcoin system of programmable way but I did it with a completely open market for arbitration providers in a completely decentralized way where no one has custodial control over the money until the transaction is executed. So I’m not trusting a third-party to hold onto the money for me. Am I making sense?
So, I wanted to really explore these options a bit because the point I’m trying to make is this – this isn’t just a currency. This isn’t just a payment network. This is the world’s first truly programmable trust and payment mechanism that is truly decentralized. And for every problem you have in this space there is one obvious, simple and straightforward solution and that solution is both centralized and wrong, right? Because if I wanted the simple version of chargeback I can do it very (0:23:19) I get Coinbase or BitPay to hold the money for 30 days and 30 days later they either give it to the merchant or if I dispute it they give it back to me and what that has done is it put those companies back into the centralized control authority of a Visa, or PayPal or any of those other providers and made my money hostage to them and also expose them to the threat of being hacked, defrauded losing the money etc. etc. so that introduced risk. But what I can also do is construct a programmable transaction that gives me all of the benefits of escrow, all of the benefits of recourse for consumer but without any centralized provider, without the arbitration provider having custodial control over the money that consumer doesn’t have custodial control over the money, the merchant doesn’t have custodial control over the money and in the vast majority of cases which 99% of the transactions everything goes peachy, everything works fine and the transaction just automatically goes through seven days later. You’ve got protection from merchants we’re not transferring the risk to them, we’ve got protection for consumers and we’ve done it without creating any intermediaries. This is just one example of how you can use the power of programmable money to create completely new classes of services that do not exist in today’s world, that simply cannot be done.
Here’s the other thing to think about – what I just did was I took a function that in our traditional financial system only occurs when you buy a house, right? You don’t do escrow to buy a cup of coffee, surely not for a cup of coffee. You don’t even do escrow if you buy a flat-screen TV, right? You only do escrow if you buy half a million dollar house, right? How many of you here have used escrow services? Okay. How many of you here use escrow services when you’re not buying a house? One, two, three. Okay, a tiny minority. Why is that? Because the complexity and friction and cost of doing an escrow transaction. It’s so high it would only be suitable for the largest of transactions. Guess what, 150 years ago the only person who got to write a check was either a king or a duke or a duchess perhaps and they would write a check to sell an estate, to buy an estate. Now, if someone pulls out a checkbook and you don’t even do it in Australia anymore, but if someone pulls out a checkbook and (0:25:53) in front of you 15 people in the line go “Ohhhh” because that’s going to take 10 minutes for them to write everything out and sign it and it’s usually, you know, it’s gone from this what I called the grand (0:26:06) of technology, it’s gone from grand (0:26:08) to grandparent, right? At first it’s only the grand Duchess of York who signs a check and then eventually it’s just the grandparents who sign checks. Now, what that means is you’ve taken a technology that used to be too expensive, too limited, too exclusive and you’ve taken that technology in the case of checks for example, and you’ve made it universal, ubiquitous, easily accessible and eventually obsolete. It becomes trivial to the point of being obsolete. But what I’m saying now is with programmable money we can take things that are currently in the purview only of the largest banks, that are things that only the largest banks, the largest multinationals use on a regular basis, the things you’d only use when you are buying a house and we can bring them into the domain of everyday purchases by massively simplifying the user interface and massively simplifying – now, I’m not saying we’re going to use escrow to buy a cup of coffee, the risk is not worth it. In fact in that scenario you don’t even need confirmations on the Blockchain because in many cases the effort required to defraud someone out of – okay, coffee is expensive in Australia I have to say. You know $8.00 for a long black is – “ouch” but –
MAN #2: Yeah, but only good coffee.
ANDREAS ANTONOPOULOS: Yeah, but it’s really good coffee I have to say that. But still I’m not going to use an escrow transaction even for that. And most merchants will do a zero confirmation transaction because the speed at which you pull customers through the line is more important than whether one of them then goes away and by colluding with a miner executes a double-spend on the $8.00 they spent on the long, black – I mean really if they’re that good they deserve a free cup of coffee. Yeah. So, but there are cases where it would make sense to use it. So if you’re buying a yacht, if you’re buying a car, if you’re buying a flat-screen TV, if you’re buying home entertainment systems perhaps, you can now bridge the gap, you know, from the trivial coffee to buying a house now there’s a whole range in between a transactions and you can expose escrow. Transactions where you’re not face-to-face with the buyer, transaction where you don’t want to establish trust.
I remember someone contacted me recently and they have a couple of (0:28:26) you know that Bitcoin ATM machines they want to sell these are about $5,000.00 a pop (0:28:31). They want to sell them for Bitcoin, an overseas transaction but it get shipped to another country well, that is a perfect case of using a Bitcoin time-lock, multisignature transaction to do escrow which will greatly simply that transaction, it will actually make it a lot easier for the two parties to establish trust and will give them a very quick and easy way to resolve any disputes without having to (0:28:54) two different jurisdictions overseas, right? So, what I’m saying is we have programmable money and we can use programmable money to take things that are in a very narrow domain today only for very (0:29:08) we can bring them into the everyday transactions. Now, right now if you wanted to do this there will be a lot of planning, pull out a terminal window and do a lot of command line stuff and cut and paste various addresses, reconstruct a multi-sig I manually constructed a time-lock transaction, we submit that transaction to test it etc. etc. it’d all be very, very complicated.
Well, you know what, I got on the internet in the early ‘90s, in fact in 1989 and if you wanted to send an e-mail you have to know Unix command line skills and you have to, you know – when I first set up e-mail on my home computer I had to compile to sent mail from scratch and then configure a sent mail configuration (0:29:52) how do it – not easy, right? And then you do command line in Unix and you send the e-mail three days later it arrives at the other end of the internet through a (0:30:02) procedure. Yeah, I’m over 40 years old.
So, the point being that that technology was never going to become mainstream but 20 years today after I sent my first e-mail through that very, very painful procedure my mom sent her first e-mail by swiping across the screen of an iPad and we’re going to follow the same exact technology curve with this technology. The power is already there, the programmability of money, the scriptable nature of transaction, it’s there for your imagination to use and find ways to combine these features and create services that have never been created before and these services will be decentralized without giving custodial control over your money. They will be instantly scalable and (0:30:48) in nature and they will be available to consumers and when you create a beautiful user experience and user interface and design and wrap this into an application that makes it simple and easy to understand for a consumer you could have the next billion dollar industry. Because in the Bitcoin space what today our problems some of us are looking at these entrepreneurs and say “Here’s an opportunity to make a billion dollar industry.” I’ve used this a lot of these expression, this analogy but if you look at all of the media publications in ’95 about the internet what they said non-stop was it’s great but now that we have all of the world’s information out there we can’t anything. And some people thought that was the problem which is going to doom the internet. You’d have all of these articles saying the internet will never work, it will collapse, it will never scale or never be able to find anything and two dudes decided that was going to be (0:31:40) company and now they have 360 billion dollar business.
So, take the problems you see in Bitcoin, find a way to solve them and this is programmable money so for the first time you have open standard open (0:31:52) open network, you don’t need to ask for anybody’s permission to go out there and create a massively successful financial service that has never existed before just with the power of your imagination and make that available to everyone, everywhere in the world simultaneously in a global scalable network.
If you can’t build a billion dollar business out of that, I mean what the hell. This is magical stuff. We are at a moment in history which is extremely rare when this enormous field of opportunity opens and people who have creative vision can now do things because you don’t have to ask for anyone’s permission. And with that thank you so much for listening. And thank you for bearing through a slightly more technical content piece there. I really appreciated all your feedback including the people who are like giving me this look which meant I needed to simplify and a few people in the audience who are (0:33:02) the programmers I can tell. All right. So I’ll start taking questions from the board and if you have any questions, you know, we can do the teenage thing which is where we sit in the same room but instead of talking to me directly you tweet at me so, I hear this is good, the mediated conversation.
All right, I’ll start with the top one – Although Bitcoin is still regarded as an experiment many have invested in the space heavily, is there any risk that an alternative currency could overtake the market cap of Bitcoin?
I don’t think there’s (0:33:36) network affecting Bitcoin and the enormous amount of brand recognition that is built over five years across the world then you probably build something really, really good. Something that has some real competitive differentiation over Bitcoin and I haven’t seen any alt currencies that really fit that bill. But if you can and you beat Bitcoin in this new brand – brand new, wide open, free market for currencies good for you I’m going to buy some of your currency. And if it’s better than Bitcoin great, let’s move on. I think you’re going to find that is very difficult to find areas where you can make enough competitive differentiation to overcome the network effective Bitcoin. The thing is that in technology in open market place all things being equal it’s not the best technology that wins. It’s the first technology that’s good enough and achieve scale, that’s the one that wins. That’s why IPv4 won because it doesn’t scale, sucks. And now we’ve spent 15 years trying to use IPv6 and we can’t and the reason we can’t is because IPv4 is embedded in so much hardware and in so many software devices and in so many different variations that unless you have a desperate need as you do in some parts of Asia it’s not worth the hassle. So, you know, most of North America is still on IPv4 because it just works, it’s good enough and there we go. And in fact if you know the history of this the internet community jumped through hoops and implemented some extremely kludgy, broken, weird things like (0:35:11) to get around the limitation of IPv4 because all of that was easier to do than do a forklift upgrade of every router out there. Bitcoin is increasingly being (0:35:23) embedded in systems, embedded in other applications, the protocol stack is being implemented in a variety of languages and when you implement a consensus network you have to implement it flawlessly. And by flawlessly I don’t mean you don’t have to have any bugs, quite the opposite. You have to have all of the bugs. You have to have all of the bugs exactly as Bitcoin has them all the way back to the beginning. So if you look at implementation of Bitcoin or Bitcoin Core it has to simulate every bug the Bitcoin Core does so it can read every block and validate it in exactly the same way. If you get it right but Bitcoin Core has it wrong, you fork off the main network and it turns out you were wrong. (0:36:05) we have the mom looking at the parade and says “Look, my son is the only one who’s properly in step” and if you’re doing that with Bitcoin you could have the most awesome (0:36:22) you’re off the main network, you’re the one who forked away from consensus. Bitcoin has enormous sticky effect because it is implementing for the first time a decentralized consensus network encode which means it’s actually very difficult to create a diverse spectrum of code that implement the same standard in exactly the same way at every block height and that’s what you (0:36:48) consensus network. So, yeah, if you could do it good for you. this is not going to be easy. Bitcoin achieve scale and is good enough. I think the only thing that unseats Bitcoin is Bitcoin itself. Bitcoin will have to catastrophically fail from the inside at this point and even that is an extremely unlikely possibility in my mind. We will see other alt currencies, we will see other massively successful alt currencies they will not probably threaten the market share or market capitalization of Bitcoin because what we’re seeing already evolve in Bitcoin is a Pareto distribution; it’s a power law, right? Which means that you have five or six of the top currencies are going to capture 99% (0:37:35) with one of them having the lion share, two or three having, you know, 10%, 15%, 20% of share and then we’re going to have another hundred thousand currencies that share the last one percentage. It’s kind of like the distribution of books on Amazon, the distribution of songs on iTune, we see this distributions, power law distributions appear anywhere where you have first mover advantage type economics and Bitcoin is exactly that.
Would you consider speaking to the Australian Senate?
I was invited to speak tomorrow unfortunately I’m traveling. I did consider Skyping from my local hotel then I experienced how wireless works at local hotels and I realized it would actually be easier for me to Skype from the New York region using my gigabit Ethernet connection than it would be to connect from Melbourne or Sydney to Canberra, that is a sad state of affairs.
And so, my message to the Australian senate is the fact that I can’t use internet to talk to you in your own country should be assign that perhaps you have regulated the industry, the internet industry in such a way – now maybe you should take wait and see approach with Bitcoin before you mess that one up. Sorry?
MAN #3: So, they’ll fly you back?
ANDREAS ANTONOPOULOS: They’ll fly me back. I would love to come back anytime. No, honestly the Australian senate hearings, I think, are important because at the moment the Bitcoin space here in Australia is at critical turning point. On one side you have the possibility of becoming a region hub of innovation and technology that will actually support not just Bitcoin in Australia but more importantly Bitcoin throughout Southeast Asia which is one of the most hungry markets for alternative currencies that exist between Indonesia, India, Pakistan, Sri Lanka, Bangladesh, Malaysia, Singapore all of the other countries around here you have plenty of closed currencies, plenty of strong currency controls and a very, very large unbanked population who are unbanked because of the lack of infrastructure who need to be connected to a currency that can work over text messaging. Now, you could be the hub that build all of that and becomes a regional powerhouse for financial services to the common people all around Southeast Asia. Or you could apply GST and throttle the baby Bitcoin tech industry in Australia in its crib and then end up being a backward of development because, you know, Bitcoin isn’t going to slow down, Bitcoin will get along just fine. Australian Bitcoin will suffer and a lot of the companies here will move abroad and they’ll take the talent with them and that would be a real shame because Bitcoin technology is very, very geographically versatile and GST is not very geographically versatile. So people will follow the path of least resistance and it would really be a shame to take a country where you have economic literacy, literacy numeracy, you have English speaking skills which by the way, you know, English speaking skills you’ll say “Okay, it deal it’s a big world.” Well, 99% of the documentation in Bitcoin is in English. It’s actually very difficult for many other countries to understand Bitcoin because for the time being it’s still all written in English so you have a distinct advantage here and don’t squander it. I would hope that doesn’t happen.
How do you (0:41:10) against value fluctuations while the payment is in multi-sig escrow?
Hmm, that is a great question. I don’t really have an answer for it right now. If you’re dealing with large amounts and at a time when Bitcoin is quite volatile that could be a problem. Some merchants in fact, I think most merchants if they price in the local currency all of their cost is in the local currency, all of their payroll is in the local currency and they just use Bitcoin as an easy payment mechanism will find it difficult to use this advanced features because that will expose them to currency exchange risk that they don’t currently have. Usually what they’ll do is they’ll take the Bitcoin payment and flip into Australian dollars instantly to limit their exposure to currency volatility. Yeah, what that means is that what we’re going to have to see before we can start using this technology is adoption of Bitcoin reaching a level perhaps where you can remain inside the Bitcoin economy for extended periods of time. And the more you can do that the less volatility you’re going to see in the overall market.
Bitcoin is volatile not because Bitcoin itself is unstable but because it is a tiny, tiny marketplace. In terms of liquidity a currency with five billion dollars worth of liquidity is like a kitty pool next to the Pacific Ocean, right? So every now and then you have this kitty pool of liquidity in Bitcoin and a fat dude jumps in and things smashed around a lot. You know, compare to that the US dollar is the Pacific Ocean, fat dude jumps in nothing really happens. So, that’s the difference, right? Volatility is you’ve got to think of a fluid dynamics, that’s actually a pretty good model for cash flow (0:42:59) volatility purposes. But the point is that Bitcoin is currently small so exchange rate fluctuations will continue for at least a couple more years until the markets are large enough, Bitcoin is traded broadly enough and enough at different marketplaces around the world. But, you know, look back where do we come from? Two years ago there was one exchange and it was managed by (0:43:23) based o PHP and MySQL code, good riddance to that but now we have a dozen exchanges that are prominent and have the majority of the volume and there’s probably another three, four dozen exchanges that have a small regional role to play. By this time next year we’re going to have a thousand exchanges and the liquidity will be much better. We’re already seeing institutional investors get involved in exchanges I don’t think this volatility is going to be a problem for too much longer.
So yes, that gives us two years to build awesome multi-sig escrow capability so it’s a race. Build them by the time you build them the volatility will be done.
What is the endgame of Bitcoin?
I believe the expression is to the moon but I would be happy to go for a somewhat more, you know, limited gameplan which is world domination.
No, seriously I think to me the most important endgame for Bitcoin is really simple. It’s the other six billion. There are six billion people on this world today who have either no banking facilities at all. About two and a half billion of them are classified as unbanked, they live in cash based societies with no ability to access to international liquidity credit to transfer money across borders (0:44:49) even transfer money across short distances, you know, that involves suitcases literally. And there’s another probably three and a half billion people who have single currency heavily restricted, heavily controlled bank accounts with very limited capabilities and there are essentially little islands, little pockets of financial activity disconnected from the rest of the world. We have the first transnational global currency, we can impact the lives of the other six billion in a way tha has never happened before. We can do to money what happened with cellphones for telecommunications. If you go to the remotest places in the world and I was talking to someone recently who treks through Papua New Guinea on a relatively regular basis and they told me, you know, you go to the middle of nowhere and you’re going to see a little solar panel and a little Nokia 1000 and that’s the way that village connects to the world. Well, guess what, that Nokia 1000 is now a Bloomberg terminal or Western Union terminal let alone origination center or mortgage origination center. A trade system, a market system, a commodity trading system, a stock trading system all of those things over text messages and Bitcoin can be done today. So we have the opportunity to help six billion people become part of the global economic community and they have the productive capacity. They’re just cut off from the world and Bitcoin has that promise within it. I know it’s a huge vision, I know it’s very audacious but I truly believe that the combination of open technology. We’ve seen how Android has done that in the world of phone market. We’ve seen how simple cellular telephony has done it. We’ve seen the miracle of M-Pesa in Kenya which over a period of just over a decade went from an experiment of exchanging cellphone minutes among people to representing 40% of the GDP of Kenya and most of that is new value that didn’t exist before. It’s not transmitted from the rest of the economy. And so, we can redo that on a global basis for the first time in history. That’s the endgame for Bitcoin – the other six billion.
All right. How will we overcome the centralization of Bitcoin mining?
I would say some kind of free market based open dynamically adjustable system that creates incentives based on game theory. So, at that point you couldn’t even burn your money for fuel because your goat was producing better fuel right next to you.
Now, if that’s the standard down there then there’s maybe another thirty or forty countries at the bottom of the stack that Bitcoin currently has a better currency. And at some point you have to think won’t either a government or just the people make the choice and say “Well, screw you 80% of us are switching to Bitcoin” and that’s it. In order for that to happen you need a confluence of circumstances. You need technology infrastructure, you need literacy, you need numeracy, you need the capability to trade and exchange Bitcoin. What that means is we have to take Bitcoin technology down and bring it to a level where it can easily be used on text message phones. We’re already seeing a lot of companies working in that space. That is a critical component.
We also need to bring awareness and a lot of times the societies where these things happen will have the incentive because of the extreme crisis to look for alternative options.
Now, what are the alternative options? In most cases you the alternative options are either to (0:48:28) or to start using a different currency. If you look at failed states like Sudan or Zimbabwe for example, a small select portion of the population trade in hard currencies, dollars, euros and in parts of Africa of course, you know, South African rands and various southern regional currencies that extremely strong. What that means is that people become adapt being multicurrency traders, right? So, if you go to certain countries the local traders can trade in five different currencies. They know the exchange rates, they can trade in all of the currencies of the surrounding nations, they’re very adapt to doing that. Most of us don’t have that experience and that makes them ready and able like how hard is it to say well, instead of five let’s do six especially if the six ones are on your phone.
So, I think you will see a tipping point reached in some countries where you will see a much bigger percentage of the population using Bitcoin as a choice of trading currency because it solves useful problems. I mean this is the key issue here. You don’t need to market Bitcoin, you don’t need to browbeat people into adopting it. Bitcoin solves problems. Bitcoin solves real problems that in some places in the world are life and death problems. And if you can solve these problems for people you don’t have to tell them to adopt Bitcoin. All you have to tell them is how and they will come stampeding towards Bitcoin when the crisis presents itself. I’ve seen that experience already in places like Argentina and it’s really palpable how people get it.
Is Apple Pay Bitcoins friend or foe?
Apple Pay is a fantastic development and it is Bitcoin’s friend. What it does is it creates a massive marketing engine behind the concept of touchless pay through NFC and Bitcoin can do touchless pay through NFC. Apple can do touchless pay through NFC by incorporating the very worst of the old credit card system, Visa and chargebacks and broken dispute resolution systems and giving your identity to everyone now including Apple to track everything you do and making you vulnerable to identity theft and fraud every time you make a transaction because nothing changes, no more secure than the old system. You’re still going to give 60 million credit cards to Amazon and Apple and Home Depot now you’re just going to give them to Apple as well and, you know, of course we know they never get hacked, right? Oh wait, well, Jennifer Lawrence might disagree.
You know, the thing is the problem there is it’s broken by design. It’s broken by design because you’re exposing credentials that should be secret to dozens of intermediaries at every moment you have a transaction. And so Apply Pay will tell people how to use touchless pay but doesn’t take away any of the problems in the past. Let people get used to using a mobile wallet on their phone and guess what, the revolution in money is not going to happen in the second and third world with iPhones. It’s going to happen on Android devices and it’s going to happen with Bitcoin on open protocols that can support very inexpensive ways of doing commerce. So, let Apple lead the way, familiarize people with this technology and then we can also use NFC but we can use it without a broken by design system. We can do push transactions instead of pull transactions. We can do a system that doesn’t need your identity with every transaction and that’s a better proposition and I think in the end we can use that new path that’s opened by Apple Pay and tell people about Bitcoin.
Also the other thing I wanted to mention on that just briefly is that Apple Pay is actually going to make mobile phone security much better. It’s also going to make people much more aware of the need for mobile phone security and we need that for Bitcoin.
What are your thoughts on national cryptocurrencies? Do you see governments potentially adopting Blockchain protocol in the future and what would be repercussions and – I think we’ve already covered that.
2015 the year of a coordinated (0:52:46) attack on Bitcoin sending it underground before it re-emerges after financial attack possible or probable?
I think to use a phrase I used recently when talking about GST, that would a monumentally stupid move. I don’t think governments are stupid enough to try and do a full frontal attack on Bitcoin. And the reason they’re not is because that’s going to have exactly the opposite effect of what they might be trying.
First of all most Western governments are not out to get Bitcoin. They’re not out to get Bitcoin because they understand that this has the possibility to generate jobs, innovation, growth, opportunity, jobs innovation, growth, opportunity and votes and campaign finance contribution. These magic words were used around politicians and they get it plus some of them want to be hip (0:53:39) with the young people. Usually the people who say that are the least hip (0:53:44) people around. I least know I’m not hip but they really think they can do that. So, you can tell them, you know, you can promote jobs, opportunity, growth, innovation, votes and campaign finance and be hip with the young people and Bitcoin is the way to do it.
Governments are not out to get Bitcoin. The governments that are out to get Bitcoin are the same governments that are opposed to internet freedom, they’re the same governments that are opposed to freedom of expression, freedom of association, freedom of determination, freedom of association, the principles of the enlightenment. If your government is opposed to principles of the enlightenment the problem isn’t Bitcoin, right? You need to take a very careful look. If a government is opposed to people taking control of their own money that is a hell of a lot more about that government than it does about Bitcoin. So, I would say that most governments today in the world are looking at Bitcoin with curiosity, they don’t see it as any major threat, they see it as some passing minor fed on the internet, they dismissed it for the most part. Some of them see the opportunity for jobs and I do not expect a coordinated states attack on Bitcoin especially since that would be massively counterproductive.
First of all in many countries around the world, especially many of the countries that haven’t yet adopted Bitcoin when the government comes out and says this is bad people go “Oh, really?” Explain to me why it is bad to have my own money. And then they start stuffing suitcases full of that particular type of money.
The Soviet Union banned the use and control of hard currencies in 1983, I believe, and the first people who started stuffing their suitcases with US dollars were the members of the politburo followed by the generals followed by the police chiefs. And once they did that and the best bribe one of those people was in hard currency so the people started hoarding it too. And so, this is what happens in countries where the rule of law is weak enough that you can go and say “I ban Bitcoin,” right? And so everybody goes “Oh, really? Why?” and start buying Bitcoin. So that doesn’t really work. And in the countries where the rule of law is strong the ‘I ban Bitcoin’ thing runs into all kinds of little niggly problems like constitutions and common law and things like that because barter is older than most forms of democratic government and it’s recognized among common law rights and also very convenient in the United States. In, what was it, 2010 the Supreme Court decides in a 5-4 decision under citizens united that money is a freedom of speech issue and money is a form of speech. So, if you actually took that to court in the US and try to ban Bitcoin someone let’s throw in a name at random, Marc Andreessen, you know, billionaire from his Netscape fortunes and in control of a massive (0:56:42) capital fund that’s invested in dozens of Bitcoin companies and has enough money to hire 10,000 lawyers a day will take this all the way to the Supreme Court and the government cannot afford to lose that because if they do that sets precedent – a precedent that is extremely dangerous because that then provides universally legal protection for Bitcoin. No one is doing a frontal attack on Bitcoin. They would much rather make it gray, make it sound ominous, talk about its connection to drugs, make insinuations about the people who use Bitcoin. You know, obviously it’s a den of thieves, pornographies and terrorists were all in it just to defraud everyone else and that message they’re going to push. The problem is that every now and then you go out and you talk about Bitcoin and your friends look at you and they say “Wow, that person I’ve known him for 15 years and they really don’t look like a terrorist, pornographer or thief to me.” And so, maybe these messages aren’t really true because Bob my friend isn’t any of those things and he really likes Bitcoin and yesterday I heard it from Annie and she really likes Bitcoin too. What’s going on here? And before you know it you debunk it simply through your actions. The number one use of Bitcoin by the way based on surveys conducted last year and this year is charity, donating and giving, tipping, right? You just keep doing that and that debunks the message very, very loudly.
So, no one is about to attack Bitcoin because to do so would be to give it a massive publicity boost and they can’t afford to do that. That really compromise their position.
When do you think the big banks in Australia will adopt Bitcoin?
Well, here’s an interesting thing that happens in competitive environments. In any competitive environment especially things like banking there is again this familiar Pareto distribution of power law. So I bet you if you look at market capitalization or revenue or stock rise or any of these other matrix and you rank Australian banks you’re going to see two or three that control 60%, 70% of the market and then you’re going to have another two dozen banks that control the other 20%, right? Is that true? Yeah, you’ve a few big ones and then you have a lot of little small ones, regional one, community banks lots of these ones. Guess what, the small ones can’t compete. They can’t outspend the big ones. They can’t out-market the big ones and they most certainly can’t buy politicians as fast as the big ones and out-lobby them. So, they have to find some other way to do that. And here comes a little secret weapon, Bitcoin that they can take and they can stick it up the nose of the big banks and disrupt their business so dramatically that when the shake up settles down suddenly a new ranking has emerge and some of these smaller banks have ridden a wave of innovation, have achieved access and expanded their region populations that never had banking and they’re doing more interesting things. That’s exactly what’s going to happen.
If you look at the early internet at first the telecom companies fought tooth and nail to stop the internet from happening. Now, a lot people don’t remember this they’re like “Oh yeah, of course” and then the AT&T became the biggest ISP we’re all happy. That’s not how it happened. I live in Greece the phone company actually put blockers on the phone systems to shut down modem calls on long distance lines. They would shut down modem calls because they didn’t want using modems to bypass the long distance monopoly. So they fought it actively and this happened all around the world and then a couple of very, very small carriers decided maybe there’s some business in this ISP thing and they started building community ISP networks and before you know it they started competing with the big guys and showing them the way and eventually AT&T picked up the message and became the largest carrier because they discovered that and this is the message I delivered to banks all the time. There’s two ways this scenario plays up. The cryptocurrency bandwagon has started rolling whether on the Bitcoin or some other name, it has happened there is no way to put that cat back in the bag. Cryptocurrency bandwagon has started rolling there are two positions you can take vis-à-vis the bandwagon – you can be on it or you can be under it because it’s going to run you over. And really there are two ways to look at this – you can embrace, you can rejoice, you can adopt, you can enhance, you can join the Blockchain systems and you can use them to extend your reach to new customers and then Bitcoin will fundamentally disrupt your business and turn it upside down in 10 years. Or you can fight, you can delay, you can disrupt, you can try to challenge Bitcoin and then Bitcoin will fundamentally disrupt your business and turn it upside down in 10 years. Either way you can play with or you can play against. This has happened before many times with technology. Do you think Kodak really believed that photography was going to disappear, right? Do you think they thought that suddenly people would stop using film? Of course not. They thought film was a form of (1:01:48) expression and nobody would use these silly low-resolution third world cameras that were coming out. And then something weird happened. In 1998 the largest manufacturer of cameras in the world was not Fuji, was not Kodak, was Nokia – a company that had never made cameras before. A company that that in fact didn’t made cameras they just slap them on phones and suddenly they were selling more cameras than anyone in the world and their industry disappeared overnight. This stuff happens again and again and again. Disruptive technologies have a tendency to do that. You can be with the whale oil or you can be with the oil, you can be with the horse or you can be with the automobile, you can be with fiat or you can be with Blockchain currencies it’s happening either way.
So, the first banks that are going to adopt Bitcoin are going to be the small banks and they’re going to use it as a competitive wedge to beat their larger competitors and what you’re going to see is this concerted march where all of the banks are saying “We won’t do Bitcoin, we won’t do bit” – “Hey, where are you going?” “We won’t do – no, this way. We won’t do Bitcoin.” One of the banks is going to say “Well, I’m going to do Bitcoin” and march off and cut off the herd. And what happens then is a stampede reaction because they’re all just scattered in all directions like “Oh shit, maybe we should do Bitcoin really, really fast.”
Every bank I talked to by the way has a project in the works, they’re doing research on Bitcoin, they’re trying to understand Blockchain technology and what’s the interesting thing is they get that they can use this in their own industry. If you don’t realize it behind the scenes bank have these massive clearing house intermediaries like Swift and DTCC that does equities in the US and all of these exchanges that take one, two, three, four, five percent cut in order to do clearing of transactions and we just introduced a completely algorithmic decentralized system that is free. You think they don’t know that? They’re like “Huh, so Swift explained to me again why I’m not going to replace your entire industry with a hundred lines of Python?” That’s the question they’re asking right, which is the question we’re asking about the banks. They’re asking about their own clearing houses so eventually they’re going to figure that out.
I am a developer with tons of free times should I work on (1:04:09) network level software, Bitcoin specific software or user experience?
I think actually probably the biggest area where we can make improvements in Bitcoin if you really know what you’re doing in terms of design and user experience that’s the area where we need desperate help. Bitcoin was designed by engineers and it shows “Oh God, does it show.” Every single word, concept and analogy used is horrible and wrong. It’s a wallet but it’s not really a wallet because it doesn’t hold coins, it holds keys and the coins are actually on a ledger that isn’t really a ledger it’s a scriptable transaction timestamp database and we called these coins but they are divisible to a hundred million units and the smallest one is called the Satoshi and you have to do a lot of math just to figure it out. We have private keys and public keys and addresses but addresses are not the same as public keys and they’re all stored on your phone which by the way did I mention has no coins on it? It’s like everything about that is wrong. All of the metaphors, all of the words, all of the concepts are wrong. Now, there is a momentum, a level of change that you can introduce to society.
When you introduced the web when you start telling people that in order to access it they have to prefix everything with http colon slash slash www dot dot dot at, people got really confused. In fact one of my favorites sites at the time was slash dot dot org which kind of played off that theme just, you know, because then you have to spell it out so people could even access it. The point being that society eventually adapted to that. Part of it is we dropped all of the prefix and most people don’t know what a www is or an http they just type something into Google and they find their way. Part of it was that society learned. You know when you look at the discussions on TV in ’94, ’95 they’re talking about is e-mail the at sign or is ww e-mail, and is ww the web and is the at sign the web or the dot or the slash and they’re trying to work this brilliant clip of the Today show in the States where they’re trying to figure this out On Air Live and they’re clueless and it’s really hilarious and they’re having that discussion with Bitcoin right now. You can expect society to move a bit but we have to move a hell of a lot to make this stuff mainstream and that’s going to take people who understand how to convey the correct metaphors and analogies and a user interface that give people tools to draw conclusions that are logical rather than misleading them. If you tell people this is a wallet they expect it to do certain things based on what wallets have done all their lives, right? And then you tell them that you can photocopy a wallet and put on another phone and wallets have never done that so it doesn’t really sound like a wallet, right? Right? You can’t make backups of your keys from your wallet that doesn’t make any sense. So if you use a word like that it has to mean something and it has to mean something to someone who has not used Bitcoin before and then it has to behave consistently so it has to do all of the things that that thing did in the past. Yes, please we need more user experience. We also need security mechanism so hardware wallets, two factor authentication, multisignature technology and making those things easier. But the real truth is that people don’t work like that. You don’t say let me choose a project and then I will become passionate and enthusiastic about it and work on that. If you try to do that you fail. In fact you will keep jumping from project to project trying to find what’s important to you, so start with that.
Start with what is important to you. You want to do development in Bitcoin, what is the thing that drives you? What is that aspect of your personality? What is the thing you love to do? What is the thing that keeps you up at night? What is the thing that you can’t possibly put down that is constantly nagging at you, okay? For me there are certain things in Bitcoin that nagged at me that I cannot put down, that whisper in my ear, that obsessed me that I keep thinking about because those are the things that are deep in my personality. They are things that mean something to me that have emotional impact and power in my own personality. Find that and do that plus Bitcoin because then you will be passionate and enthusiastic and creative and then it won’t be work. Then it will be your passion just plus Bitcoin, right? Find the place – if you’re a musician do music in Bitcoin. You know if you love social interaction do social media in Bitcoin. If you love eating out with your friends find a way to help restaurants take Bitcoin, I don’t know but find the thing you love and then add Bitcoin. Don’t go searching around the Bitcoin space and say “What could I do next?” Because the truth is that this is the most incredibly broad opportunity you’ve ever seen. You can re-imagine any aspect of money and any financial service that exist so trying to find which one to grab is the wrong way.
Where do you see Bitcoin three years from now?
I only make predictions on reasonable timescales so I think three weeks from now Bitcoin will be hovering around the 400-dollar range and a few more startups will have started. That’s the limit of my horizon. I have no clue where this thing is going and that’s part of the magic. That’s part of the excitement, right? Like could you imagine – how many people here were on the internet before 1998? Okay. Could you honestly, now could you imagine Facebook or Twitter, right? No chance. Could you imagine Uber? No chance in hell, maybe one of those. But you really couldn’t imagine where this was going. You couldn’t imagine it would move this fast, this many people would embrace it it would become global so fast you couldn’t possibly imagine all the applications. That’s where we are today in Bitcoin. You are sitting at the forefront of history. The people in this room are 10 years ahead of the mainstream watching history happen and some of you are making history happen and that’s the most exciting place in the world and that means I have no clue what’s going to happen in three years and that’s why I love Bitcoin.
How do each of you secure your Bitcoin?
I’ve talked about this a few times before. This is not a recommendation for people who are new to the space but I’ll just tell you the honest – how I secure my Bitcoins, 99.9% of all – I don’t own a lot of Bitcoins first of all. I’ve never mined. I’ve earned a bit of Bitcoins last year or so by doing this job but before that I worked for free so I don’t own much but what I do own 99.9% I store in paper wallets. I used BIP 38 encryption on those paper wallets to use a passphrase to make them a bit more robust. I have a waterproof scratch-off, stick-on on top. I sold the product that did this in fact in the past. And these paper wallets I printed out multiple copies and then I distributed them on – at the moment two continents and three countries. I’ve given them to people I trust they put them in a safe. At some point I had my paper wallets stored in a safe deposit box in a bank which is the height of irony because I’m storing my Bitcoin in a bank because the one thing they do know how to do is 20-foot walls and guards with guns, they’re really good at that. So I put my Bitcoins in the bank fault and that worked great. But I always have backups so I have multiple paper wallets they’re distributed in multiple undisclosed locations and they’re protected by encrypted keys. What that means is that effectively my Bitcoin is offline. I don’t trust – my laptop is compromised I assume that every time I turn it on. I assume that every word I utter, everything I do in front of that camera unless it has the (1:12:17) sticker covering the lens, everything I do on that computer is track monitored and available to the entire world. That’s the assumption you make as a security professional if you want to maintain security, right? You assume compromise because the chances are that if my machine was compromised I wouldn’t know it and people can do it in such a way that I wouldn’t know it. I also use some of the new technologies that are coming out for example, I’m not endorsing by I have a Trezor wallet which I find quite convenient and I enjoyed that a lot. I was one of the people who contributed to the original sort of kick-starter to get one of the first edition of Trezor. If you don’t know what a Trezor is it’s a little hardware device that does all of the key generation and signing of transactions with a hierarchical deterministic wallet, it’s a pretty cool device. It basically makes Bitcoin security easy for people who don’t know what they’re doing which is exactly what we need to be doing and there’s dozens of other hardware wallets coming out.
If you want to secure you Bitcoin the best way to do it is to store it offline, to put it on systems that are not online. And now over the next year or so I think you’re going to be able to a lot more with multisignature technology. So, here’s another consideration. If you secure your Bitcoin too well then you end up losing it, right? There’s a couple of ways that you can lose your Bitcoin by securing it too well. One is by using too many layers of encryption and going fully paranoid tin foil hat so you know you have (1:13:44) encrypted partition onto which you have Ethereum password protected wallets that only contains one of the three signatures the other two are in a backup that you gave to your uncle that’s double encypted with, you know, AS256 and a passphrase that you’ve memorized that is completely random that you generated with Diceware and within a month you will forget one of those keys and you’re screwed. Guaranteed, this has happened a number of times. That’s one way to lose you money.
The other way to lose you money which a lot of people don’t think about is that if you secure your Bitcoin really well and something happens to you your family is out of luck. How many of you today have family members who can access your Bitcoin holdings if you get run over by a bus? All right. And I’m not going for the terrible scenario where you die, I mean that’s horrible and – and actually you should plan for that, you know, even if you’re young you should – if you have people who are dependent on you, if you have children, if you have a spouse, if you have people who depend on you you should plan for the eventuality of passing some of that money onto your descendants or family if something happens to you. But let me take the really simple scenario. You’re out and you’re trying to do the double ollie on the steps of parliament right, and you execute the perfect double ollie and fortunately it’s captured on the GoPro and then you slip and you smash your head onto the concrete and you pass out and you’re unconscious for 48 hours and someone needs to pay the bill and the only money you have is in Bitcoin and none of your friends know how to get your Bitcoin, your girlfriend, your wife, your husband doesn’t know how to get your Bitcoin and now you can’t pay your hospital bill. I mean you don’t need to die, you just need to be incapacitated for 48 hours and your Bitcoin is locked up and no one can get to it. Consider using multi-sig technology to actually solve that problem. There’s a couple of ways you can do it. You can create multi-sig address where for example, you have one of two or one of three and you give a key to your spouse and you printout a key and you put it in the safe and you explain to that person how to break it out if something happens. The other way to do is to contact a lawyer. You know Pamela Morgan for empowered law for example offers the service for survivability and digital estate planning. So if you got a lot of Bitcoin like really a lot of Bitcoin then you should probably have a plan for what happens. So that’s my security little talk. We’re way out of time but I can keep going, okay. All right. I will not be offended if you have other things to do, if you need to go, if you need to bathroom just get up, go, no worries. I’ll just keep going for as long as I can stay upright and they keep us in this building. Yes? I got one here, thank you.
MAN #4: What do you think are the challenges in scaling Bitcoin?
ANDREAS ANTONOPOULOS: Okay, very good question. What do I think are the challenges in scaling Bitcoin and we’ve had a couple of questions on here. So, I think Donald Knuth once said that premature optimization is the bane of programmers or something like that. I don’t remember the exact quote I’m paraphrasing.
MAN #5: The root of all evil.
ANDREAS ANTONOPOULOS: The root of all evil, okay. So, those of us who are programmers understand that optimization, especially optimization for scalability which is really the typical problem that we come into involves making choices and when you’re presented with a trade-off picking one fork versus the other and the problem with that is that when you make that choice you exclude all of the paths that descend from the other fork. If we go into Bitcoin and we start making scalability decisions now especially scalability decisions that preclude other forks in the path we’re doing so with very little information, we don’t understand yet how people are going to use this. We don’t understand what Bitcoin will need to scale to. Like is Bitcoin going to become a micropayment network for doing under one dollar transactions globally at the rate of Visa? In that case we have to scale Bitcoin to be able to do between let’s say, 20,000 and 100,000 transactions per second, right? Now, that’s a very different computer science problem than saying Bitcoin will primarily be the reserve currency and through sidechains will empower other chains that do those things. But really it will be the money that governments use to buy aircraft carriers and pay for oil and very, very rich individuals use to move money around and eventually it will become more like kind of a high quality long-term gold holding and that’s a possible path, we don’t know yet. There might be better alt currencies that do the micropayment thing sidechains may allow us to use Bitcoin as a reserve currency as the gold for everything else. If you make choices through premature optimization now you will have to pick between one or two of those alternatives and that’s a bad idea. I think the best approach and we see this now is this – on the one hand we see experimental modeling of the scaling and capacity issues to understand where are the bottlenecks in Bitcoin. The bottlenecks in terms of disk capacity and Blockchain size, the Blockchain database size which some people argue become worse when you have data introduction like OP RETURN and hashes and fingerprints and counterparty and things like that. Other people say it doesn’t really matter. Moore’s law for disk space says that, you know, I have 23 gig now I have 23 terabytes in five years and then I’ll 23 petabytes in five years after that on my iPod in which case do I really need to worry about a 23-gig Blockchain that is growing that fast. These are all valid considerations. There are concerns about what the capacity transactions are. There’s a company that did some simulating and they were able to easily reach a hundred thousand transaction per second by simulating certain conditions and exceptions, simple optimizations that can happen on the Blockchain. The bottom line is we don’t have a scaling problem right now.
Have you ever put in a transaction that wasn’t included into the first block if you paid sufficient fee? I have it. When you have that problem – very smart people have already been working on the solution for years and they will find a way to fix it. When the Blockchain gets so big that many of the full nodes dropped out of service then we’ll start seeing pruning nodes in fact Gavin Andresen is working on exactly that right now. When bandwidth becomes a problem people start working on it. Gavin Andresen is also working on a system using invertible bloom filters in order to do a state synchronization between nodes where you can transmit a block simply as the difference in state and you can now get a (1:20:44) of magnitude reduction in the size of blocks and make them linear rather than increasing exponentially in size. Those are some interesting solutions. Optimization will happen. I have no worries that Bitcoin can scale and the simple reason for that is because I know that IPv4 (1:21:06) and yet I use it everyday. I mean it really can’t scale, it couldn’t. And if you look the discussions and I’ve been in this industry long enough, every year you get that article that says “Is this the end of the road for Ethernets? Will it ever be able to exceed one megabit per second?” And then two years later is this the end of the road for Ethernet? Will it ever be able to exceed 10 megabits per second? Here are 10 reasons why fundamental physics precludes any further advancements. I kid you not you’ll see this articles every year, the internet will grind to a halt, it will run out of addresses and we’ll all be doomed. The thing is when that becomes a problem people are (1:21:48) to find solutions and that’s on the internet where we couldn’t monetize it. This is money. It’s premonetized, right? So, if you want to find a solution you have a five billion dollar bounty to fix the shit before it blows up. All right. So, scalability I really don’t worry at all. What I do think is if you are interested in scalability this is a space with some very exciting big data analytics, statistical science, data structure synchronization, database science to be done some really interesting innovation where you can create software that will used by millions or perhaps billions of people if you get it right and you can play a small part in history. So go forth and optimize and put in a pull request when the problem comes up look at your solution. Let’s see what else.
How do you think Bitcoin will change from a core development perspective? Are there any improvements, changes or features being worked on right now that you are excited about?
Well, here’s the thing people think of Bitcoin as a static thing and it’s anything but static. Bitcoin today is very, very different from Bitcoin of 2009. A lot of things have been enhanced and developed in the protocol. New script operands like opcheck multi-sig verify which is only two years old for multi-sig OP_RETURN which is less than a year old and allows the introduction of data structures in the Blockchain which made it possible for counterparty in Mastercoin and proof-of-existence and other metaprotocols to introduce very exciting features elegantly into the Blockchain instead of through (1:23:33). The development of () prunable Blockchain nodes. So, at the moment you have the option to run either a full node which carries every unspent transaction output, every transaction and every block ever happened. That’s about 23 gigs of disk space and counting and it’s a pretty big memory footprint too. Or you can run a lightweight node where you depend on third-party verification through the simplified verification protocol, simplified payment verification (SPV) on other servers to provide you with essentially branches of the Blockchain in the form of (1:24:10) miracle paths that you can independently verify. Again, I’m getting a bit technical for those who are interested in these issues. There’s no middle ground. The proposal that’s very interesting right now is to create prunable nodes. Now a full node not only carries the entire transaction set it also carries all of the transactions that have already been spent and there’s no reason other than historical analysis to carry those. They’re not required to create new transactions. You need to carry the entire UTXO set if you want to be able to do independent verification of transactions and you need to have (1:24:44) miracle path and miracle trees for all of the transactions that you prune but you don’t need the actual transactions so you could draw up the Blockchain by an order of magnitude and create an intermediary node that is still fully authoritative and self-verifying but without carrying all the baggage then the full node essentially gets renamed. It’s an archival node, right? It’s like the way back machine.
The other developments that I’m really interested in the invertible bloom filters that we talked about, we did a show on Let’s Talk Bitcoin we have Gavin Andresen come on the show and explain this to us which was rather interesting. I never in my wildest dreams thought when we started Let’s Talk Bitcoin that I could go on a show and say “Well, I just read a post by Gavin about invertible bloom filters and if anybody knows anything about that come talk to us” and two days later I get an e-mail from Gavin saying “Can I come on your show” like Wow, amazing. So this actually allows for sharing blocks in such a way across nodes and propagating blocks that they don’t get bigger the more transactions you put in because you’re only sharing the difference in state. The bottom line with this is that at the moment there’s an incentive from miners not to make blocks too large because large blocks propagate slowly. And if they propagate slowly there is a possibility that someone else will find a block and propagate it before you and you end up on the wrong fork, right? So, all of the work you did improving the block gets wasted, you get zero reward because you got tromped by someone with a shorter block that propagated faster. So this is dynamic trade-off happening and as a result blocks don’t have that many transactions in them, the block size isn’t just increasing.
Well, the interesting thing is that if you just say let’s increase the block size the problem that happens then is that that encourages centralization of mining and the solution that Gavin is proposing reduces that problem because if it takes the same to propagate two meg block the same amount of data more or less as it does to propagate a half meg block then you don’t have that incentive for centralization. I find that very interesting. I think sidechains are very interesting.
Bitcoin core developments are very exciting space but it suffers from one fundamental limitation and that limitation is that what you are doing here is equivalent to trying to do in-flight maintenance of the left engine of a Boeing 747, and you’d better get it right. Because this is a five billion dollar economy in-flight and so you screwed up and bad things happen. In fact we’ve seen that happen in the great fork of April 2013 we had a 26 block diversions caused by a bug in Berkeley DB. That is exactly the kind of thing that happens when code development is working too fast. So at the moment the pace of development is very conservative, the testing is very extensive and most miners are running a release that’s four releases behind (1:27:52) in order to make sure it’s well tested. I think that’s a really good thing. But what that does is it means that Bitcoin can’t easily absorb all of the innovative passion and creativity and new technologies that are coming along and that’s why altchains are so interesting and that’s why sidechains are so interesting because they can try out new features and market conditions and act as a filter and testing ground and then Bitcoin can arrogantly pick and choose just the best ones, right? Plug the best features from the most successful altcoins and incorporate them after a more thorough level of testing into the Blockchain and in fact sidechains will allow that to happen even better. But I do think one of the things that’s going to happen is that the core protocol of Bitcoin is gradually going to get ossified. And that’s because it’s going to get embedded into devices, systems and software so deeply that the cost of upgrading becomes a problem and at that point the pace of chains will start slowing down. So you’ll end up with an environment where we released Bitcoin, you know, 14.5 but 90% of the network is still running 09 from two years ago because they haven’t got on around to upgrade it because it’s really running on firmware in a USB miner or whatever, that’s going to happen in Bitcoin, I think.
MAN #1: Andreas, (1:29:22) last question.
ANDREAS ANTONOPOULOS: Last question, okay.
MAN #6: Thanks so much. I’m (1:29:27) use Bitcoin but (1:29:29)
ANDREAS ANTONOPOULOS: Welcome.
MAN #6: (1:29:33)
Andreas Antonopoulos: All right, that’s a great question. So, the question was about barriers (1:30:19) entry into Bitcoin. The fact that Bitcoin technology is in many ways opaque to the mainstream audience, it’s obscure and it’s difficult to understand, it’s difficult to explain, it’s also very difficult to get Bitcoin to buy some or earn some or find some out there and the person asking the question was concerned that that would somehow stall the adoption of Bitcoin and slow things down.
I think if we look objectively at technologies that have come before the truth is that the adoption cycle and the maturity cycle of a technology usually it takes decades. But what we’ve seen over the last 50 years is an accelerating cycle where technologies get simpler and become mainstream much, much faster. So, you know, the automobile took 40 years before it was mainstream technology, electricity took, you know, 30 or 40 years, the telephone took 20, 25, 30 years before it was mainstream. The internet took 15, Bitcoin is going to take eight. You see this compressing cycle. Now, that’s not to me this Bitcoin in eight years will be mainstream. The Bitcoin that we will have in eight years will be almost unrecognizable from the Bitcoin of today just like the internet that I used when I first got on which required Unix command line skill. I tell people they said don’t believe it but when I was on the internet in the beginning I had a piece of paper in my wallet which a list of IP addresses of the 10 or 15 sites that had interesting stuff on Gopher, Archie and ftp that I could go download. This is pre-DNS, this is pre-web. There are no search engines. You need to know the IP address of the ftp server of Stanford University. Now is that the internet we were going to go mainstream with? Hell, no. Nobody is going to do. Well, nobody apart from utterly geeky people like me. Some people will look at a technology, see the long-term vision and they will overcome any obstacle to make that technology work. The first copy of Linux I downloaded was Version 095B. I downloaded it from Linus Torvalds’ personal ftp directory at the University of (1:32:44) I believe in Finland. I downloaded it onto 200 floppy disks to install it but because I couldn’t afford 200 floppy disks because I was a poor student I can only buy a hundred. So I downloaded the first hundred started the installation, stop halfway through, overwrote the 100 disk, this took four days and I had a sign on my screen that says “If you turn this off I will kill you” because it was halfway and then I would have to overwrite the 100 floppies all over again with the first set and then manually hand configure X11 Window system to find my video card and then, you know, glory of glories I got a 1200 (1:33:33) connection over TCP/IP to some obscure machine that I could only use over Telnet. Now, normal people don’t do that. Normal people say “Hey, the latest Star Wars is out. You want to catch a coke and go see a movie?” whereas me and my friends we were like “Oh, the latest version of Linux is out. Do want to spend four days installing it on 200 floppies?” Well, let me tell you something thank god for the freaks and the geeks and the weirdos who will go through all of that because I know there are a lot of you in this room. And nobody ever made history by going to watch the latest Star Wars movie but the people who did some of those things did make history and we’re going to make the really difficult choices now and use technology that is almost unusable because we can see what it can become. QR codes are not the future of Bitcoin any more than IP addresses without DNS were the future of the internet. That stuff is all going to disappear, we’re never going to see that again. The idea that five years from now you will ever see a Bitcoin address is preposterous. You will only see a name, if that you will see an automated endpoint. You won’t look at addresses any more than the average internet user knows about the Mac address of their Ethernet card, understands the frame settings, knows what their IP address is, knows what their, you know, slash 24 network subnet is. I do and that’s because I’m weird and geeky and – but the average user doesn’t have to know and that’s where Bitcoin will need to go. You’re absolutely right Bitcoin as we have it today is not mainstream ready. But that’s okay because there’s enough weirdos and geeks and eccentrics in the world who see the vision, who work hard and the simple answer here is this – all of these problems are billion dollar industry opportunities. You have open programmable money. You think something is difficult to use? Make it easy to use and you will be very successful in this space. And that was the last question I was taking today. Thank you so much. I’ll stick around.
(END OF AUDIO)