← Hidden Forces
Hidden Forces
The Future of Distributed Applications and Ethereum's Blockchain | Andrew Keys
00:55:30
In this Market Forces segment of Hidden Forces, host Demetri Kofinas covers all things blockchain related with Andrew Keys of Ethereum's Consensus Systems. Andrew's role at consensus is to help cultivate an ecosystem of distributed application developm...
Loading summary
Transcript185 lines
- [00:00]
A
What's up everybody? Welcome to this Market Forces segment of the Hidden Forces podcast. I have someone really awesome in studio today, Andrew Keys from Consensus Systems. Andrew, how are you doing?
- [00:11]
B
Great. Thanks for having me.
- [00:13]
A
It's great having you here, man. Really looking forward to this conversation. Before we get started, why don't you tell our audience a little bit about you so they know you and consensys and then we'll get into the meat of the conversation.
- [00:24]
B
So I guess professionally I grew up Wall street for all intents and purposes. I worked at UBS in investment banking and I worked at a hedge fund that dealt specifically with life insurance. And I co founded a revenue cycle management system where essentially I learned about how bad all the payment processes are in insurance and in banking. And at the same time basically bitcoin happened. Bitcoin was really cool. It solved the double spend. Essentially if I sent you a bitcoin, I couldn't send that to somebody else. And there were advancements in mathematics and computer science, but essentially we couldn't add business logic. We couldn't have. If this consideration happens, then we're able to make a payment, else no payment. And I researched the space and I thought it was relatively mature until I found what I would consider the next generation of this technology.
- [01:22]
A
You said when was this?
- [01:25]
B
So I began researching Bitcoin probably 2011, 2012 and it really started 2009.
- [01:32]
A
Right.
- [01:32]
B
So pretty early in the space. But I didn't really jump in to the space until I met a gentleman by the name of Joseph Lubin who is actually one of the co founders of the Ethereum protocol. And I met him at the first ever meetup of Ethereum in New York. And once he explained the implications of this, essentially that we could have a peer to peer next generation implementation of the World Wide Web without centralized intermediaries, I kind of went down a rabbit hole.
- [02:05]
A
Are you talking about the interplanetary file system?
- [02:08]
B
No, Ethereum are you talking about? And we can get into the interplanetary file system as well. But I believe that there is essentially a next generation worldwide web that we are on verge of. And I think that we are in kind of the equivalent of 1993 in the 1996 Internet explosion.
- [02:26]
A
Like not Odyssey, what was it? Not America Online. What was before America Online? Blanket.
- [02:30]
B
Now I think it was AOL was online. Aol. Prodigy.
- [02:34]
A
Prodigy. Prodigy.
- [02:35]
B
Prodigy.
- [02:35]
A
All right, so a few things you mentioned, the double spend problem. That's essentially the way that we deal with that in society today is we essentially have these clearinghouses we have a third party, a trusted third party that verifies a transaction. That's how we deal with that sort of double spend problem that you can't counterfeit. You mentioned peer to peer networks. Do you want to describe a little bit for our audience the difference between a peer to peer system and a server client model and how that all.
- [03:00]
B
Fits in here in its broadest sense, we as humans of planet Earth pay a lot of money to trust each other. We have trusted intermediaries like banks, like escrow agents, like Internet service providers, so things like Facebook or Google and those intermediaries extract tremendous amounts of value, providing kind of a meat space. And a peer to peer system is essentially where two counterparties can interact similarly, but without that intermediary. So without the Facebook, without the bank, without the real estate agent, without the auditor, without the custodian.
- [03:45]
A
And that's exactly what the Internet was supposed to be. And it was very early on, but immediately, once it started to scale beyond a university level, you can't host your site on your own computer.
- [03:56]
B
Right.
- [03:56]
A
So that protocol doesn't work for that type of a system you're describing. Right.
- [04:00]
B
So basically the database and the Internet grew up separately and we accessed databases through Internet service providers, we go on to Facebook, we go on to the bank account, we go on to Google, and basically we're accessing database technology through those intermediaries instead of having essentially a database technology that can go on top of an Internet protocol where two counterparties could interact peer to peer. And that was kind of the opening act with Bitcoin where we were able to do that, where we were able to send a token that represented a monetary figure and without the bank. And kind of the future of this is that we're now able to. So basically Bitcoin at its core was this gigantic peer to peer abacus where I could fire this token to you and you could fire it to a colleague. And now there's the next generation of this where it's a peer to peer world computer. So basically at every node of the environment of the Ethereum protocol is actually a Turing complete scripting language. So basically anything that you can program, you can build in a peer to peer fashion.
- [05:10]
A
All right, so I want to continue on that. Before I do, I want to take a quote that you, that I got from a lecture talk, because you do a lot of this, you actually go around the world educating people because that's actually relevant and useful for your company. This is what you said. And I want to say this to the audience, because I think it's important to find a way to make sure we're on the same page here. Before we continue, you said that the Internet simplified and expanded the way ways we humans communicated and that the blockchain simplifies and expands the way that we agree. Do me a favor, just drill that home what that distinction means.
- [05:42]
B
Sure. Basically, this next generation blockchain technology is going to commoditize trust the way that we've had the Internet commoditize communication. Right now, our agreements are typically constructed by attorneys on Microsoft Word documents with signatures and notaries. And in the future, our agreements are actually going to turn into software where two counterparties, as simple as sending an email to each other, are able to send assets to each other or make agreements that disperse assets and the contract. The notion of being able to agree is something that we as humans, I think, take for granted. And once we're now able to kind of arbitrarily codify these agreements and granularly authenticate them, it will essentially create liquidity in previous illiquid markets.
- [06:53]
A
Okay, you're talking about specifically their smart contracts, but more broadly, the agreement. And if I understand it correctly, and I just want to make sure we're clear on this, it goes further. It goes really to the point that the auditing of the ledger is distributed out to the network. X number of people on the network running computers, and we'll get into how this works, are economically incentivized to participate in the audit process. That auditing happens in a distributed manner so that you don't have to go through a central clearinghouse. And so it means that the network can spontaneously resolve its issues.
- [07:28]
B
So Planet Earth, for example, just one of a few examples runs on what we call gap accounting or double entry accounting, where essentially, if you and I are counterparties and you give me $1, there would be a credit in my books for the dollar and a debit in your books for the dollar. The two entries. Now you can imagine that, that there'll be essentially a third entry. There'll be the credit on my side, the debit on your side, and then this irreversible, this immutable watermarking, this notarization that you indeed gave me the dollar, that we agreed on it, and now we can never change that. And what we just did was automated the notion, the concept, the process of audit. So PwC, Ernst and Young, KPMG, Deloitte, get paid billions, if not trillions of dollars to facilitate. When did you send me the invoice? When did I pay it back to you? The basic double entry accounting. Now we're essentially creating in just this one example, a complete audit trail that's historically verifiable and immutable.
- [08:36]
A
There are many benefits of this. We'll get into them. The least of which in my opinion is the fact that you're going to save a huge sum of money. In fact, some of the estimates I've read just in general, in terms middlemen. Middlemen, not just in accounting, but in general is around 20% low hanging fruit. Is that about accurate?
- [08:52]
B
I would say at least, at least.
- [08:54]
A
We could completely revolutionize and make much more efficient this economy that we work in. I should also say on a high level. I want to mention this to the audience because we talk about this often on the show. This fits into the larger conversation about centralized versus decentralized systems. I want to get into this also with bitcoin cash, Andrew, for example, large societies have been able to scale because they centralize, they were able to feed large masses, get large armies, etc. And now we're trying to create a society potentially with blockchain, which is more amenable to the way that we in the west, at least our technology works and our value set, which is decentralized, which is, you know, and the rights would not be guaranteed under the, by a central authority. They'd be sort of endemic to the process, to the protocol of the system. Alright, like I said, we will get to the blockchain, to the bitcoin cash thing. I'll ask Andrew about that in a bit because that is kind of the big deal in the news. But let's draw the distinction now. So we talked sort of in general what blockchain is, which is what Bitcoin, the big capital B is essentially when it came out, that was the iteration of blockchain. So what is the difference between Ethereum and bitcoin?
- [10:07]
B
Okay, great question. So at a macro level, I think it's important to note kind of what a blockchain is before we get into that. And I'll just quickly touch on that. So basically it's really four cooperating technologies. So first and foremost it's a peer to peer network, so it's not client server. Secondly, there is a virtual machine and basically the bitcoin virtual machine is by design created. So the only function that we can do is send a token peer to peer. And so how does that work?
- [10:41]
A
When you say send a token peer to peer, can you explain that for the audience?
- [10:43]
B
Basically that means that the Only function that can be accomplished on Bitcoin is the sending of a token from me to you that represents a Bitcoin or 100 Bitcoin or 1,000 Bitcoin. And those Bitcoin are essentially transferable for fiat US dollars or British pounds or euros or Chinese one.
- [11:05]
A
And the authorization process is the consensus.
- [11:07]
B
Of the system and the consensus algorithm is, is the other one. And so basically the difference between Bitcoin and Ethereum lies in the virtual machine. And without going into the deep computer science, I think that what you should note is that at every node of the Ethereum virtual machine is a scripting language. And the Ethereum virtual machine can compile what are called smart contracts, these agreements that we've now codified, whereas the Bitcoin virtual machine is concentrating on solely transferring value peer to peer.
- [11:43]
A
Okay. And also from what I understand, I should say also, another thing I wrote down before the interview when I was thinking about this is that we're moving, as we move and hopefully will move towards a site that is able to take advantage of this type of software. We're going from one that is approval based to one that is consent based, which is important because the approval is granted by a central authority that has some level of power and the sort of classic who will guard the guardians. Right, well said. So from what I understand though, also one of the things about Ethereum, and maybe this was implicit in your answer, is that developers can develop on it much more easily. Was that what you were saying?
- [12:16]
B
Yes, exactly. So a term, complete scripting language, which basically means that any developer that knows JavaScript is able to build these applications.
- [12:24]
A
All right. And that's what allows Ethereum to evolve.
- [12:26]
B
Right.
- [12:26]
A
So it's basically that's the sort of mutation element in Ethereum that makes it robust or hopefully robust, so that people that invest early in it, whether they're buying Ethereum's currency, which I actually want to ask you about, I'm a little confused about that because when Ethereum started, there was what is now known as Ethereum Classic or Ether, what was it called?
- [12:44]
B
So when Ethereum began at the genesis block, the native fuel to run smart contracts was Ether. It is still Ether. There was an episode where early in the days of Ethereum, basically six months after the Ethereum protocol was essentially born, there was an application called the dao, which stood for Decentralized Autonomous Organization. And that Decentralized Autonomous Organization had a bug was exploited and a hacker was able to begin to move upwards of $50 million in this currency of Ether and the community essentially voted to do a fork of the protocol to essentially excise this bug and to essentially reverse what had occurred. And there was a livid debate on code being law, because there were people within the community that believed that this was how the code operated and therefore the code was law and it wasn't technically a hack.
- [14:01]
A
What was it? The hack? Are you able to explain that? Do you understand?
- [14:04]
B
It was a recursive function and to be honest, it is above my pay grade.
- [14:09]
A
Is that the loops that I hear about?
- [14:10]
B
Yeah. So basically they were able to excise $50 million worth of this crypto fuel and in order to reverse the transaction, there was a fork made and there were.
- [14:24]
A
When you say fork, you mean there was a fork literally is a divergent path. And now you have a whole new sort of chain in the blockchain. Yeah.
- [14:30]
B
And basically from that there were people that believed that that fork should not have happened. And basically they kept driving the existing chain, or the initial chain, which was deemed Ethereum Classic. And basically that has about 10% of the value. So basically, if you were to take the total market cap, Ethereum Classic, Ether, it's about a 9010 split because all.
- [14:57]
A
The computing power has gone through the new Ethereum. That brings up two points and we're going to tackle them separately and we'll see which one. One that brings up the question of security and how secure is a blockchain currency. I want to talk about that because I think that's really important for the audience and particularly for skeptics. Everyone should be skeptical about everything, certainly for this show. Then the other has to do with what you just talked about, which is the fork. I was going to get into the bitcoin cash thing through the issue of scaling. We can roll that into this as well. But that's what's happened now. That's what happened yesterday, actually, we're recording this on Wednesday, August 2nd. Tuesday, August 1st, Bitcoin forked. And now we have something called bitcoin cash. Can you explain to our audience, to the best of your knowledge, what that is and what's happened there?
- [15:41]
B
In full disclosure, Bitcoin is not my subject matter expertise, but from a macro perspective, there was a debate in the bitcoin community on how to scale the.
- [15:55]
A
Bitcoin blockchain between the engineers and then the miners. And the miners are the people who basically own the hardware that's able to do the proof of work, basically to verify the transactions on the network and that to release new blocks, Correct?
- [16:10]
B
Well, very well said. And basically a faction of the bitcoin community activated a hard fork and is now mining a subsequent fork of the network. And that has been named Bitcoin cash. And interestingly, last time I checked, it was valued at about $400 per BCH, Bitcoin cash. So what that has actually done is. And the price of bitcoin was somewhere between $2,700 and $2,800.
- [16:45]
A
Didn't really affect it, right?
- [16:46]
B
It didn't affect it. It's like a stock split cap of the bitcoin community now is the highest that it's ever been.
- [16:54]
A
So essentially the problem that the bitcoin community was addressing is that there is an issue with scale. And Bitcoin has. And this is where we can also talk a little bit about proof of work and sort of how that whole process happens. There is a deflationary component built into the way in which Bitcoin's blocks are mined. That's the other thing. This is also distributed. I guess maybe a better way to say it is, and please I want you to correct me here, is this is a distributed ledger, and it is one giant ledger. And as the transactions grow on the ledger, so does the amount of work that needs to be done in order to verify the transactions and the difficulty. And the difficulty. So why don't you explain all of that? And this actually costs real world energy, massive amounts of it. So I guess in your answer, Andrew, I'd love to know. I don't want to forget this. Why do we need to do these? Because I haven't understood this exactly. Why is it that beyond just the verification of the actual transactions, these computers are required to do ever harder problems that are completely arbitrary?
- [17:50]
B
2.
- [17:50]
A
Why not? If you answer that, and there's a good reason for the arbitrary aspect of the problems, why can't those arbitrary problems be solving other pressing issues like being used for computing power somewhere else? And I think that's it. There was a third one, but I forgot it.
- [18:04]
B
Okay, so there's a lot to unpack here. First and foremost on scalability, I think it's important to note that WeChat, for example, does 100,000 transactions per second.
- [18:16]
A
What's WeChat?
- [18:17]
B
It's the Chinese messaging. So for Americans, American Express, there's about 25,000 transactions per second, as much as about 40 to 50,000 in peak times in Black Friday visas, comparable. So let's say if we just take American Express, Visa and MasterCard on an average day, ballpark is about 100,000 transactions per second, right now, the Bitcoin blockchain processes somewhere between, I think, 4 to 10 per second. Ethereum does slightly more than that, I would say.
- [18:50]
A
What percentage is that? Below 1.00000.
- [18:56]
B
And Ethereum does slightly more than Bitcoin today. And there are trade offs. So there's a trade off by having this kind of trustless, decentralized ledger versus having kind of centralized clearinghouses that can improve kind of your transactions per second. So with that being said, the Bitcoin scaling issue was based on the block size, where basically one faction wanted to keep it at 1 gigabyte. Yeah. And then the other one wanted 2. This BCH is supposed to be 8. And there was just kind of a rapid debate on that. Ethereum has a slightly different model where there's actually three separate scaling solutions and two of them should be implemented within the next, I would say, calendar six months. So the first one is the notion of state channels. And Bitcoin has a version of this as well, called the Lightning network. The most famous state channel in the Ethereum ecosystem is called the Raiden network. And basically what that means is that if, if you and I went to a bar and I put down my credit card, I would essentially open a payment channel on chain. And then if I bought us a round of drinks, that would be kind of a micro payment, that would happen off chain or a micro transaction. And if I bought us two hamburgers, that would be another one. And then if I bought us dessert, that would be another one. But you could do millions of these before closing out. And then when we close, we're on chain. So basically opening the channel is on chain and closing the channel is on chain. Then you can have these kind of millions of microtransing transactions.
- [20:34]
A
So you can go offline, do a bunch of work, then reconnect, and then all of that gets synchronized, gets updated with the ledger.
- [20:41]
B
Correct.
- [20:42]
A
So that's a way of increasing the system's efficiency and basically giving the user some power over finding ways to be efficient.
- [20:48]
B
Correct.
- [20:49]
A
Okay, very interesting.
- [20:50]
B
So another scaling implementation is transitioning from what I would call proof of work to what's called proof of stake. And the notion behind this is that, to your point, Bitcoin and Ethereum today expel tremendous amounts of energy to form consensus, to validate and to propagate these blocks. There have been estimates somewhere between the country of Ireland to the country of Denmark's electricity spend on a day is equivalent to Bitcoin and Ethereum.
- [21:22]
A
That's crazy. I did not Hear that before.
- [21:24]
B
That's crazy.
- [21:25]
A
That's wild. And, like, how much of that is in China? Like a gazillion.
- [21:28]
B
A lot of the bitcoin is in.
- [21:30]
A
China because all the inefficiencies in the world are in China.
- [21:33]
B
Well, and what I would also say it's interesting because there are different places that have different costs of kilowatt hours. So a fun little fact. For the first month of the Ethereum blockchain of a friend of mine and myself, we mined ether and I had 1% of the hash rate of the total Ethereum network out of a little crappy apartment in Manhattan. And it was about 100 degrees in the room because there were all these, these actual computers that were doing work. And I learned very quickly how expensive a kilowatt hour is. That essentially is the metering mechanism. So In Manhattan, it's 25 cents a kilowatt hour. Just over the bridge in New Jersey, it's about 15 cents. If you go to Washington state, it's 7 cents. If you go to Venezuela, it's 4 cents. If you go to a college dorm where electricity is included, it's free. If you go to somewhere in China where it's not, of course, that was.
- [22:34]
A
A problem because a lot of college students were mining crazy amounts of Bitcoin.
- [22:38]
B
So there was kind of an electricity consideration, just the cost of electricity. But to also talk about your scalability considerations, the Ethereum network, in addition to the state channels, is transitioning from what's called proof of work, which is that this wasteful mining that uses GPUs, so like the Nvidia cards, the AMD cards, to what's called proof of stake, where people are able to make a deposit and to essentially help propagate each block and form consensus. And similar to how the miners right now earn a micropayment for securitizing the network or securing the network, the stakers or the depositors will be able to earn interest for depositing. And as long as they're not nefarious, there's nothing that would happen incorrectly with that block. And that will also increase the transactions per second and the throughput. And the last one is what's called sharding. So right now, everyone in the blockchain has to form consensus about the state of the blockchain. And in the future, shards, it's very redundant.
- [23:48]
A
You have a very redundant network.
- [23:50]
B
Right. And in the future, shards of the network will only form consensus pieces. Yeah. So for one transaction, you don't need the whole blockchain but you only need a percentage of it that will further.
- [24:02]
A
How does that work? What I mean is, what I understood about sharding is for example, if I have a company, if I run JP Morgan and we do a bunch of internal transactions, sharding to me, I guess maybe that's what state channels are. Maybe that's what state channels are. It's basically the idea of doing. You're basically internalizing. In other words, the way I interpreted quadratic sharding is that it's a form of compression. It's like a compression algorithm that you're not even an algorithm, actually not an algorithm. You're allowing the network to sort of compress itself at its own pace.
- [24:33]
B
Well, I would say it's more of a scalability solution, but it's basically evolving the network where only parts of the network have to form consensus or come to agreement rather than the entire network. In the JP Morgan case, I think it's interesting to kind of explore that and tease that out a bit because I use a poor man's analogy about us being in 93 of 96.
- [24:55]
A
Sounds more like 89, Andrew, as you're talking to me.
- [24:57]
B
All right, fair enough, fair enough. For the first time ever in the history of the bank of JP Morgan, they open sourced software, 200 plus year old bank and what they essentially did was they created an enterprise private implementation of the Ethereum network.
- [25:15]
A
That's what I had been reading. Tell me about that.
- [25:17]
B
Okay, so that's called quorum and basically the notion here is that depending upon how these blockchain technologies are used, you've got kind of the next generation of the database and or the next generation of the Internet. And being in 89 or in 93, everyone worked on intranets first where we had these kind of private networks. And today there are still many networks or many processes that never touched the public Internet.
- [25:44]
A
Yeah, you're talking about they're using the same protocol but they're not connected to the wider Internet. So they basically use that protocol internally. That's like your, your house or the.
- [25:51]
B
Local area network or Swift that moves the majority of planet Earth banking. Yeah, so what we're seeing here is kind of the implementation of these technologies in permissioned settings. And I think that the reason why we're seeing them first in these permissioned settings is kind of like technology history rhymes. It doesn't necessarily have to be the exact same, but we're seeing these kind of intranets being built while the public Ethereum mainnet implements its scalability solutions and Its privacy solutions, which is something we haven't touched on yet.
- [26:28]
A
Privacy, yeah.
- [26:29]
B
So right now, the public Ethereum mainnet, if you were to go on and look at the source code and look at the transactions, you'd basically be able to see everything.
- [26:39]
A
When you say you'd be able to see everything, explain that, please.
- [26:42]
B
You'd be able to go into the bytecode and see the transactions and see who was performing them as a wallet.
- [26:50]
A
Based on their wallets. But you don't actually necessarily connect that to the person.
- [26:53]
B
Correct. And there has been a field of cryptography called Zero Knowledge Proofs, or ZK Snarks, that was made famous by another cryptocurrency called zcash Famous or kind of brilliant cryptographer named Zuko Wilcox, who created the ability to essentially create a proof where two counterparties would be able to transact if they both knew a password in all simplicity. And in doing so, it creates a privacy layer. So while that's still yet to be implemented into this public next generation, what I believe to be the next generation of the World Wide Web, we're seeing lots of activity on these private intranets, and that's what quorum is.
- [27:43]
A
Well, the privacy thing is huge, Andrew. Also, what I would say to the audience listening is that what's really, really cool about blockchain protocol technology is that it's got a little bit of everything for everybody. I think everyone has different concerns. Certain things are more important to others. We talked about the huge savings you can potentially generate from cutting out the middleman. For me, the privacy aspect, basically finding a solution to this clearinghouse problem, which is currently you need to have a third party involved. To me, I love anything that's decentralized. I think that is the most interesting thing. Before we move on, though, when you were bringing up Wall street, you were bringing up JP Morgan. More generally, I think people, maybe the majority of the public is unaware of this. Certainly people within the industry are. When Bitcoin started, it was purely Wild West. But now that blockchain technology that it's sort of migrated, you have a lot of these altcoins, alternative cryptocurrencies, and you have Ethereum. You guys, I should say. Let me clarify this as well. Consensus systems is essentially like a mediated developer, or you guys help companies and people develop software that runs on Ethereum's world computer or what would it be considered?
- [28:57]
B
Yeah, so I think that's a pretty good description. We also use complementary technologies like the interplanetary file system.
- [29:03]
A
I want to talk about that for sure.
- [29:04]
B
Basically, we build applications, we build at the protocol layer. So if you could imagine kind of like TCP ip, HTTP as a protocol, we maintained three implementations of the equivalent.
- [29:19]
A
HTTP is what the Internet has run.
- [29:21]
B
On for the longest time.
- [29:22]
A
Forever.
- [29:23]
B
Yeah. So we drive three of the eight implementations of the Ethereum protocol. So we maintain the Java client, the Haskell client, the. NET client. Then we build open source tools. So kind of like Git, the developer framework. So we built one called Truffle, which is the most downloaded developer tool for building these types of applications. Then we build applications and then we educate governments and large enterprises on what this technology is. And then we also believe that this is kind of the next generation, or kind of a new generation of an asset class, like gold.
- [30:04]
A
What is the asset here?
- [30:05]
B
So digital, digitized, tokenized assets. So Ether, you could consider it kind of like a crypto fuel to run the next generation of the World Wide Web.
- [30:18]
A
Looks like a taxi medallion. That's not a very good analogy. So help me, because I'm.
- [30:24]
B
So now, instead of you having a website that you have to pay Amazon or Microsoft Azure to run storage on.
- [30:32]
A
Oh, got it.
- [30:33]
B
Now you've got this resource that has storage capabilities. It has kind of this trustless nature to it. And for any transaction or any computing, it takes a little micropayment of a few pennies that I believe will be a very valuable asset.
- [30:53]
A
So you're saying instead of me going to Bluehost or to GoDaddy or to WP Engineering, I can go to some kind of front company that scales this out across the network and everyone who's involved in that hosting, distributed hosting in this particular sort of example, gets a certain payment. That's interesting.
- [31:08]
B
So that's the Ethereum protocol. Similarly, the sister, I would say protocol is something like filecoin, where Ethereum, if you were to imagine a next generation decentralized world Wide Web, Ethereum is this kind of transaction layer where you actually create these smart contracts. But it's very expensive to store data on that protocol using that protocol. So basically it's this kind of trustless peer to peer transaction layer is Ethereum. And then there's this peer to peer decentralized storage layer where we can actually store interplanetary files.
- [31:47]
A
Let's talk about that. That's what I thought you were going to.
- [31:49]
B
So the interplanetary file system has a coin called filecoin. And just like we had miners that contribute to securing the network, there are miners in the filecoin system that are essentially giving computing power for Storage. So unused hard drives at its simplest form. And you know, the bet is that all of this unused processing power and hard drives can be used and rent it out like an Airbnb or an Uber. So now you've got kind of decentralized transaction layer where instead of Facebook extracting tremendous amounts of value as the intermediary, we can have decentralized social media where you and I can have our own identity and maybe we get paid a dollar a day. If we look at an advertisement, then we have this decentralized file storage. And I think the last one would be decentralized chat. So like Skype or Gmail or Facebook Messenger.
- [32:53]
A
For real time communication.
- [32:54]
B
Yeah, for real time communication.
- [32:55]
A
And the all the sort of including privacy and verification would happen on IPFS or happen on the Ethereum.
- [33:01]
B
So they all kind of work together. So you can have a smart contract that also has a file stored and basically you would have a pointer to a cryptographic hashtag and that would be where you would look at the file. So you can imagine having an actual paper contract and you would want a copy of that in addition to your computer code. And that could just be through a pointer on ipfs.
- [33:26]
A
So IPFS basically is the distributed storage and computational layer.
- [33:30]
B
I would say the distributed storage layer.
- [33:32]
A
Storage layer.
- [33:33]
B
And then you can do computation through Ethereum.
- [33:36]
A
But what I mean is, for example, let's say I want to outsource my computation as an individual, I can outsource it through the IPFS layer.
- [33:42]
B
Yeah. And there's also another one that's also trying to do that called Golem.
- [33:46]
A
But you're getting the idea this speaks to the flexibility of Ethereum. It also speaks to the beautiful architecture here. This is absolutely gorgeous. I mean, this is part of the revolution that's underway. And that now brings us to. I'm jumping all over the place. But you know what, let's just go there because we're talking very sort of. Some could say we're being naive. And why is it potentially that we're being naive? Because there is the fact of governments. What is the position of governments? Because right now what we're doing is we're disintermediating the trusted third parties in a variety of areas. At some point that's going to be a threat. It is a threat already to certain displaced corporations or individuals. We're displacing. I mean, if you were going to displace the whole middleman industry, that's a threat to those individuals. And of course there's also other sort of as the governments around the world would say, our legitimate sort of cases where they feel like they need to intervene, especially if we have taxes and we have social services and everything else. If people are making profits on Bitcoin, they need to report those profits as forex profits and they don't, let's say, and whatever else. What is the situation now with governments?
- [34:51]
B
Yeah, so I couldn't give you a broad overview of governments, but I think that it's all on kind of a case by case basis. So what we're able to do now in this next generation worldwide web is we're able to essentially tokenize every asset. So we're going to be able to tokenize oil, gas, stock, bonds, Beyonce tickets, loyalty points, and we're also going to essentially be able to raise money in that way where we can create. If you and I started a company, we could explain that idea and essentially raise money without the venture capitalists and people could send us ether and in.
- [35:34]
A
Return without the banks either.
- [35:35]
B
Yeah, without the banks, without the venture capitalists, none of that. And the SEC actually just released an investigative report two or three days ago that said that some of these tokens could be considered securities. And securities have to be registered and there has to be specific kind of processes that go around issuing securities. And there's kind of a specific test in a court of law called the Howey Test that are measures by which to determine if an asset is a security or if a person or corporation is essentially offering what would be considered a security. And so that's one example where I believe that we are seeing regulatory guidance which will actually help foster growth of the industry. So where you've got this kind of wild, wild west next generation technology that has the ability to essentially democratize fundraising in this one example, and you've got the government, in this case the SEC in the US and the Monetary Authority of Singapore released about 24 hours later a very similar approach. And basically you've got kind of clear guidelines on what is a security, what's not, and whether it's a token or whether it's, you know, something out of the backseat of someone's car. If you're doing this A, B, C and D, that should be considered a security. And if you're not doing A, B, C and D, it won't be. Another consideration here is that now you can actually bake in compliance into smart contracts. So you could have a regulator be a node on the environment. So rather than a kind of a quarterly look back in a regulated industry, you can actually have real time auditing, real time compliance, real time dashboards, real time graph theoretics in this kind of next generation database technology. So there's different.
- [37:36]
A
You basically embed the regulators within the sort of the system so that they can scale with it and it's not counterproductive. And of course that's also consistent with the. Because right now you have a basically distributed system that supposedly doesn't need a third authentication layer and therefore coming in and saying no, you need one.
- [37:54]
B
Well, not necessarily an authentication layer, but a regulatory oversight. So now you can have regulatory oversight and you have better means by which to do so.
- [38:05]
A
That's also been, I think, a big advantage of Ethereum, which is that Bitcoin came with a lot of baggage. Whether it was Mount Gox or whether it was Silk Road or whether it was just the Bitcoin community of evangelists, libertarians. Ethereum is the next iteration of that and it's much more business friendly and a lot of businesses are on board. You've got a lot of these companies on Wall street that are also on board with Ethereum. Are a lot of Wall street firms looking to really build this into their payment system so they can adopt it because they see it coming and they want to benefit from the marginal benefits of it.
- [38:38]
B
What I would say is that banks are paid to provide trust and to provide intermediation. Essentially this technology commoditizes trust and intermediation. So with that being said, there are optimizations where in some instances Bitcoin just proved that we can send value without the bank in the middle. That's, let's say, the full step. But you can also use this technology to optimize the 30, 40 year old database structures that are in the banks. You can do real time collateral rebalancing, real time margin, and in doing so there's kind of an optimization process. So in the history of open source technologies there have been kind of standards bodies. You can remember Linux, you can remember when java turned to J2EE, and that's really when it permeated, when there was kind of a standard around web APIs and the database APIs. Similarly, there has become a standards body that has become the largest open source blockchain initiative in the world, which is called the Enterprise Ethereum Alliance.
- [39:46]
A
Didn't you start that or you were a big part of it?
- [39:48]
B
I was one of the people who drove it, yes.
- [39:50]
A
Okay.
- [39:50]
B
And companies like Accenture, bbba, British Petroleum, JP Morgan, Santander, Microsoft, intel are all part of this enterprise Ethereum alliance that are using this technology in their various verticals.
- [40:06]
A
So these standards.
- [40:08]
B
Yeah, so basically we're working on standards around the protocol, the Ethereum protocol. And then we've got banking working group, supply chain working group, advertising technology working group, Healthcare working group. Merck's driving the healthcare, for example. Basically using these technologies in enterprise settings and figuring out how we create these networks, these n sided, multi sided marketplaces now.
- [40:33]
A
Yeah, this is a fully professional. This market has changed tremendously. You mentioned 2012. 2011 was when you started to. To look into the space. That's when it first entered my radar also. There was something that happened around that time for me. I caught wind of it. End of 2011. Right. Beginning of 2012. And so much has changed in the last five years. Unrecognizable. It's unrecognizable. You got guys like you in this space who are legitimate business people. I mean, it's pretty remarkable, right? Do you have a hard time, do you have a hard time convincing your parents that you have a legitimate job?
- [41:09]
B
They think it's pretty interesting.
- [41:10]
A
They get it now actually.
- [41:12]
B
I think it's pretty. But I would still say it's still the wild, wild west.
- [41:16]
A
Sure.
- [41:16]
B
It's still super early days in kind of the spectrum of where this technology can go. I believe that smart contracts can essentially re architect how every business process works and you can essentially automate that. And if you could imagine kind of economic, social, political, operating systems redesigned where now that you have this trustless transaction layer, you basically float bureaucracy away from the system.
- [41:48]
A
God, I love that. Sounds beautiful. That makes me happy. Gives me the warm fuzzies. There you go. Security. Let's talk about security.
- [41:55]
B
Sure.
- [41:55]
A
What are the risks? What are the known knowns? What are the known unknowns? The unknown unknowns.
- [41:59]
B
So for all intents and purposes, the Internet that we know is really for mostly a communication layer. Now with the simplicity of sending an email, we're going to be able to distribute assets and make these agreements that actually bind assets. And what that does is now websites actually hold money very simply. And what that presents is a whole new attack vector.
- [42:25]
A
Yeah, the attack landscape is horrific. We did an episode on this program with Josh Corman on cybersecurity and we covered all of this, starting with the scariest scenario of all, which is hospitals and medical equipment, which is the most vulnerable of all. But yeah, absolutely.
- [42:38]
B
Energy grids.
- [42:39]
A
Let me just say this also, and I want you to continue. It's an anecdote. I have a flight tomorrow. I got an email from Expedia today telling me that my data may have been compromised.
- [42:47]
B
There you go.
- [42:48]
A
Well, that's super. Thanks a lot.
- [42:50]
B
And Expedia is your middleman.
- [42:51]
A
Yeah.
- [42:52]
B
So why aren't you going peer to peer to British Airways or Delta? What is that? Intermediary now? So now this is your blockchain use case 101. You put all of your information into the intermediaries database. Because you couldn't essentially have a database layer on top of the Internet where you could go peer to peer. In the future you're going to see much more kind of peer to peer transactions.
- [43:19]
A
It's just messier. In other words, this solution is far more clean. But let's go.
- [43:23]
B
So security.
- [43:24]
A
Yes.
- [43:24]
B
So basically the first and foremost, I think security considerations that we're seeing right now is or solutions are called formal verification. So a way to essentially verify the intent of computer code. And there's a lot of work that's being done around that. It's being led by an entity called IC3, the Institute for Cryptocurrencies and Contracts, which is an academic kind of consortium between Cornell University, University of Illinois Champlain, I believe, and Umaryland. Some kind of brilliant cryptographers that are working on kind of verifying and securing these technologies. But what we're seeing is this code, instead of it just being kind of intermediation for communication, now has assets. So that's the biggest security concern.
- [44:16]
A
Right, you're talking about the ledger.
- [44:17]
B
Yeah.
- [44:17]
A
In other words, this is, and this goes to the point of like the.
- [44:20]
B
And the application layer. So I mean there's a protocol and the protocol has worked remarkably well. We haven't really seen bugs or hacks at the protocol layer. We've been seeing them on the application. So it's like saying the Internet doesn't get hacked. Website www.xyz.com got hacked. But now that website, like in the example of Mount Gox, held tremendous amounts of assets as kind of this.
- [44:47]
A
Right. The exchange. Right, right, right.
- [44:49]
B
And. Or in the dao, which was a decentralized application that held assets as well and that got hacked.
- [44:56]
A
Okay, for sure. And that can continue to happen. This doesn't provide a solution for that.
- [45:00]
B
So we're hoping that as the standards and the libraries mature, there will be kind of specific metrics to audit these contracts and there will be templates that if you wanted to do a capital raise, you use these thousand lines of code that have been used 100 times and have been audited and then that would greatly reduce security.
- [45:25]
A
Okay. What I want Though, because I don't dispute that. On the protocol level, though, let's talk about that, because I know that with Bitcoin it's a majority consensus. Right. Is it the same thing with Ethereum?
- [45:37]
B
Similar, yes.
- [45:38]
A
Can you just explain that a little bit? Because that's the level at which the majority of people don't understand the robustness. And I also would like to know more. How reliable is it and is there a way in which potentially it could be compromised, the process itself, a ledger could be corrupted.
- [45:54]
B
So neither the Ethereum protocol nor the Bitcoin protocol has ever been hacked.
- [45:59]
A
That's true. That's a big deal.
- [46:01]
B
The actual protocol has never been hacked. Essentially what you've got now is rather than this kind of honey pot where a database would be that if once you get in, you've got blocks and each block is essentially secured and irreversible. And if you were to attack, you know, one block in the Ethereum case, there would be another block 10 seconds later, and then you'd have to basically undo that one and that one. And if you wanted one four before, you'd have to attack all three.
- [46:31]
A
Because each time you move forward, it's a linear progression. Each time you consent on a block which has a maximum size, is it also 1 gigabyte with Ethereum or. No, you said it was bigger. In any case, once there's consensus reached around one block, which is basically a block of accounting, you tie it up and you put it in the closet and it's there forever. And then the next block references the most recent block and you just go one by one by one. But because each block is referencing the one before it, the entire ledger is accounted for. But then what's the sort of gold finger attack that I hear about, which is that if you have more than 50%, 51% or more miners.
- [47:05]
B
So a 51% attack is where essentially if somebody controlled 51% of the network, they'd be able to essentially undo something like that. And so you're talking about in excess of $10 billion for the Ethereum network and in excess of $20 billion for the Bitcoin network.
- [47:28]
A
And that also would be obvious, right? I mean, that's not something that can happen covertly. Right. It would be something where you'd compromise the system. That would be like blowing up the system. That's like a suicide attack. Correct. So it's not a stealth attack, it's not something where you could sort of do it stealthily. So it's very robust.
- [47:42]
B
You would need to have kind of half of Nvidia's stock of miners or half of AMD's and you would have had to buy Intel's corporation.
- [47:54]
A
It's a very robust protocol. And I think also what I'd like to, and that's what I would like to kind of say to the audience, which is we're talking about now these secondary and tertiary layers of software that are being built upon this protocol. And you don't have to exchange with your other party directly if you don't want to. You can go through a third party that's using this protocol as their base layer and therefore their transactions are more secure and they can guarantee for you. That's what a Chase bank could do or PayPal or something. PayPal could adopt this software and basically engage in the exact same function and guarantee your transaction for you or whatever else. Right, right.
- [48:32]
B
I think what this software and this technology does, it drives a price discovery mechanism for the cost of intermediation. So what is a middleman worth? The example I always use is kind of uber. Uber takes 35% of every ride revenue from the driver. And in doing so, Uber provides a beautiful user interface, user experience, UI ux. It provides insurance, it provides reputation. So those five stars, so we know that either one of us isn't a creep, the driver or the passenger. Similarly, we could create a decentralized application and we could have a peer to peer transaction where we would have a smart contract and a GPS algorithm. That said, if you took me from point A to point B, you would get the fare and you would essentially get 99.9%. But then we would have to bake in. What's the cost of creating a reputation attribute for our identities? What's the cost of insuring the smart contract for that ride? Because that's already baked in. Maybe all of those costs end up being 15% or 10%. But in broad terms, we're transferring money from the intermediary to the counterparties.
- [49:56]
A
All right, Andrew, I want to ask you one more thing before we go because I got to keep this within an hour, but we can answer this last question for as long as you want on this last question and then I'd love to have you again on it because there's so much more we can talk about it. It's a rabbit hole. It's clear. All right, so this is a question I have. Do you see any potential for this software being used, this protocol being used to address problems of the commons? Right now it's been used entirely for the private sector. Right. And when I say problems of the commons, what I mean is the things that governments now do and only governments can do, is there an emerging sort of. Is there anything happening creatively on that front, using this protocol to do that?
- [50:32]
B
Sure. I think you could do it for voting, I think you could do it for identity. I think one of the most interesting notions here is for the first time ever, you have this decentralized global database and you've got millions of people that don't have an identity where they live. And for the first time ever, they could have a digital identity and the government could provide attestations or attributes.
- [51:01]
A
I don't know that word, attestations.
- [51:04]
B
So an attestation is basically kind of a reputational consideration for how we conduct ourselves in society. You attest to who you are, it's like the five stars on Uber. And then basically you're going to be able to, to create this portable reputation. So you've got people in India that can't get credit, and now they can create a digital identity. And if somebody loans them money and they pay it back, they'd be able to build reputation. And then what you'd be able to have is that somebody in America may not know who that person is, but has seen that that person in India has paid back their loans every single time and could lend on that, for example, or you could attach an electronic medical record to like a thumb or a footprint of a baby, so you could create like an electronic medical record. And when we hear about kind of refugees, what we're seeing is that the identity providers can be the governments. And in places where governments don't have all that infrastructure, you can now create this global identity system.
- [52:14]
A
That's also where the great opportunities for innovation are. I just want to end it by saying this or hand it off to you, but this is what I want to say before we go, which is that for me, what this always comes back to in these conversations that I have with people is that this is the market of technology attempting to provide a solution, an advancement to resolve a conflict that exists within an increasingly technological society that is intermediated with layers of software, which is how do we maintain our freedom, These Western values of liberal democratic values in an increasingly digitized world. And in order to do that, we need to have distributed software that respects our privacy and our property rights, et cetera, which is inherent in the system. And I think that also speaks to why a lot of money in China is particularly going out into Bitcoin, because they don't have any property rights in China. But it's not just about property, not just about money, it's not just about Bitcoin. As you heard today with Andrew speaking, this is Ethereum. And what, what consensus systems does is not about currency. It's about the larger applications of blockchain technology, which is where it's at. I mean, increasingly that's it.
- [53:22]
B
So to your point, Dimitri, we have what I would call a very asymmetric Gini coefficient. So basically what that means is that there are 80 people on planet Earth that have the same amount of wealth as the bottom 3.5 billion.
- [53:38]
A
We talk about that a lot.
- [53:39]
B
And I would argue that a lot of that is due to the costs of centralization, whether it's of power and the advantages. Yeah. And so we've got these kind of data monopolies that are extracting tremendous amounts of value for providing database layers. And what you're seeing is this next generation peer to peer mechanism that will drive down the cost of that type of intermediation and will create these price discovery mechanisms to really ask us what is the cost of trust. And in places where we can drastically reduce that, it will happen. And I think that that will help planet Earth tremendously. And I think the other consideration is now that we can digitize assets that were formerly analog, like real estate. You are going to have new liquidity markets. And in doing so, you know, if we had 100 people that owned a building, we essentially wouldn't have real liquidity until we exited or sold that building. Now we can create basically 100 tokens and sell them on a secondary or tertiary market.
- [55:01]
A
When you say digitize, just to clarify, what you're saying is you're basically ascribe an identity, a digital identity that is robust and non corruptible to physical assets.
- [55:10]
B
Yep. And you can track the provenance of it, you can track the chain of custody and the authenticity of it. And it's immutable.
- [55:16]
A
Andrew, it was awesome having you on today of all days. It's even better than having you on yesterday because it gave me some time to prepare. I hope you come on again for us. I told you I want to do a full episode with you, man. I appreciate it so much.
- [55:28]
B
Thank you.
- [55:28]
A
All right, thank you. Cheers.