Transcript

[00:12] SPEAKER_00: Gavin Wood. This is an impromptu talk on a legality. A legality. Who knows what I mean when I say. Hold on, who's got a beer? First of all. Right, yeah, this is a talk where you might want to have a beer.

It's a word. It's a relatively new word. It was coined in 2003. I didn't coin it. I thought I did, but it turned out I didn't. It was coined by an NGO who were talking about what the NSA are doing. But I think my meaning is better than theirs, so I'm going to stick with it.

A legality. I mean by it. Well, hold on, who can guess what I mean by it first? Actually, as it happens, you're not the first person to say that. But it turns out that the original authors, the original creators of the word used the double L. So I can't argue with it, can I? Well, alright, okay. Hands up if you prefer double L to single L. Three. Okay. And the other way around, more than three. Well, alright, fair enough. Maybe I'll change it. I don't think I can change it. Here, it's PDFs. Next time I talk about this.

So, does anyone know what I mean by it? Yep. It's probably something which is outside of the scope. Sort of. Yeah, I think that's what they originally, what the other guys meant by actually so when. Oh, we got other people. Other people. Legal no man's land. Yeah. So it's kind of like something which just doesn't care as to whether what. Cannot care, I suppose is more specific. Cannot care as to whether its actions are, you know, might be interpreted as legal or might be interpreted as illegal. It's sort of like, you know, it's this legal stuff. I don't care about that. I don't know about that. It's just not in my world. So kind of what you said. But I don't know, I'm coming at it from a different direction I suppose. Systems that cannot care.

Could I have the next slide please? Slide master. Yeah. Does everybody know what I mean when I say capital S, lowercase aa, capital S, Software as a Service. Yeah. So this is what happens when you go on, I don't know, Facebook or Twitter, you know, all of that stuff, all those services that you get on the web that are obviously software backed. That's what I mean when I say Software as a Service. And the talk is going to be centralized around decentralized Software as a Service.

And this software tends to specify rules by which we can interact. So like Facebook specifies the rule whereby if you click the like button on somebody, their like counter goes up by one. That's a simple rule, right? That's what it means, that's what you expect. If it doesn't do that, then you know something's wrong. Something's wrong with either your expectation or Facebook's back end.

Liking is one thing of course. In PayPal if you pay somebody some money and their account doesn't go up yet your account goes down, that's then something is really wrong that actually matters to you much more. Probably, I don't know, maybe you're really interested in Facebook likes, I don't know. But generally we care more about money than likes.

These are abstract rules. But abstract can become concrete when we bring into the balance, when bring in value into the equation. When you start paying people for things, that actually affects the real world. I'll pay you if you, I don't know, go to the shops for me, right? So when we say value transfer, that's the paying thing. And then we have these other things, actuators and oracles. So actuators, I mean basically these are things that allow you to specify things like going to the shops, right? So they're inputs, they're like, yes, they did indeed go to the shops for me, I was shopping here. And oracles do basically the same thing generally of more generalized fashion.

So what's new with the decentralized thing? Because we've already got eBay and I don't know, Amazon and whatnot. These are all Software as a Service, right? They give us services based on software. We already have this, so what's new about it?

Well, there's a big massive M difference. That's what I'm standing here telling you tonight. Massive difference. The entire talk is based around this massive difference, right? Software as a Service, at least in the traditional sense, has an operator, right? This is eBay Inc. or PayPal Inc. There are real people existing in a real legal environment with real lawyers who are doing this stuff. And if they do something, if they provide a service to you that is deemed wrong or illegal or bad in some way, or they piss off a lawyer or whatever, then they get fucked, right? They get a lawyer on their ass and they have to do something about it, which generally involves employing more lawyers, going to court and then maybe going to prison.

Decentralized Software as a Service has no operator. Yeah, this is why Bitcoin is a currency and not just yet another PayPal. Bitcoin has no operator. If Bitcoin pays somebody who the government of Iran doesn't like, the government of Iran can do nothing other than shutting down the Internet. This is why Bitcoin is different to PayPal. There is not a Bitcoin Inc. There is not a CEO of Bitcoin. There is not a team of lawyers who Bitcoin employs to take the heat off them.

Now, I like to call this a force of nature because actually that's kind of what it is. Bitcoin is based on very little other than nature. If you look at like the game theoretic aspects to Bitcoin, and there's actually not much more than that, it just uses nature. Everything comes down to individual parties acting in their own best interest. If it didn't, it would break down pretty quickly.

Now, each decentralized Software as a Service can also be considered a force of nature. So when you're creating new decentralized Software as a Service, you're actually kind of creating just a whole new force of nature. And that's kind of an interesting thing because it changes the rules a little bit.

So what have we got? We got Bitcoin, BitTorrent. So BitTorrent employs a relatively simple game theoretic core tit for tat. I'll give you a bit of this if you give me a bit of that that I find is useful. But nonetheless, it has been demonstrated to be a pretty severe force of nature, as the Motion Picture Association of America laments.

In general, if they're properly implemented, these things cannot be shut down, not by a court, not by a police force, not by a nation state. They are literally forces of nature. They are combinations of mathematics. And the thing that makes them different to most stuff that has come along is that they tend not to care about the human context of what they're doing. And this is what changes things. Forces of nature don't care about us. They don't care about our worries or our qualms or our insistence upon intellectual property laws, which are actually fairly arbitrary. If you go 500 years ago, then intellectual property laws were quite different, expectations were quite different.

So where are we going? We began at Bitcoin and we're slowly moving to the top, right? We've got crypto finance on the top and we've got sort of apps on the bottom. Moving along there and we've got this thing which I'm calling crypto law, right? So crypto law is like the abstract bit of crypto. It's kind of like the, if you just kind of take Bitcoin and pull out the fundamentals, you distill it all down. Crypto law is kind of what you get.

Very soon now, and I say this as someone who's in a position of some degree of authority, crypto law is going to come out. We're going to get platforms for implementing these decentralized software services, these a legal entities, and they cannot be, well, they cannot trivially be policed.

Ultimately, the police or the authority or the state, whatever you want to call it, have had a fairly easy time policing Software as a Service because actually what they do, they don't bother policing it properly. What they do is they go to the service providers and basically lean on them. And if, you know, if someone's selling, I don't know, in France some Nazi memorabilia or something, then they lean on eBay. They don't go to the user themselves, they just, eBay, take this down and give us their address.

But in the future they won't be able to do that because it will be out of the control of any individual authority. They'll have to actually do it properly and go to the users. And doing it properly costs a lot of time and resources.

Ultimately we are going to end up with either think Soviet era East Germany, every member of the household is an informer. Seems unlikely that we're going to put up with that. Or we've got to start rethinking what illegal is the set of. And I reckon we're probably going to see a fair few changes.

Blockchains necessarily bring on additional transparency and openness and efficiency. Some of the changes we'll see will be very positive. Things like existing laws for know your customer so you can track where payments are coming from relatively perfectly. Anti money laundering and all the things that people here who run Bitcoin cash points will sort of lie awake at night about. Become trivial to implement.

Taxes like governments are going to like some parts of the blockchain. They'll love it. Banks too. Things like financial transactions become really easy to check and audit, especially if the people doing them are at least a little bit collaborative. And if they're not, it becomes a lot easier to say, look, you were wrong, you are bad, you were actually explicitly, consciously doing this incorrectly. It becomes far harder to hide things under the table unnecessarily.

I think we're going to run into a more liberal, free society. Trends will necessarily become harder to police. It's going to become harder to shut down deals that both parties find mutually acceptable, even if those deals, maybe the rest of society doesn't much like them. And it's going to be harder to avoid the uncomfortable truths on a socially wide level. If these deals keep going down, these will be relatively easily seen and form a trend that we cannot get away from.

Inevitably in the end the whole the law will change. I mean it has to. Examples of this are relatively easy to come by. One of them is copyright law. Back when the VCR was released it was actually illegal to record things off TV. But so many people do it. Are you really going to bother policing them all? No, you can't. Law changes. So many people smoke weed. Apparently in Washington state now it's actually just legal. Ultimately law has to alter according to social norms.

However, the change that stage between one state and the other state, that can be very volatile if both sides do not collaborate. Lawmakers need to work with the new system and they need to make this change as less painful as possible.

In my opinion, the correct way to handle this is through education. Both of the lawmakers, of the institutions in society and of course of the individuals in society. All decisions need to be evidence based, open and transparent. If you talk to people like David Nutt, originally on the, what was it, the UK Drugs Misuse Board. Anyone correct me? It wasn't that. I'm paraphrasing badly. No. Okay, well it was almost that.

This guy is a very well respected professor, and he points out decisions quite literally are not taken using evidence and decisions should always be used to maximize health and well being, which all too often they're not. Morality, prejudice, misunderstanding and myths above all must really not be abused as they have been in the past. It will become very obvious that that's happening when we get this level of effectively transparency and openness and the ability to work in the a legal area.

And to everybody else of course, there's a renewed need for responsibility and inevitably tolerance and honesty with it. That is the talk on a legality. Are there any questions?

The software in crypto law description. Okay, yeah, sure. So, what is crypto law? Is basically the question. So it's, could you go back a few slides back to that lovely diagram that took me ages to make? I don't normally make diagrams for my slides because it takes so long because I'm old school. But this one I felt was worth doing.

So the idea is at the bottom you've got cryptocurrency, right? And cryptocurrency is this notion of being able to transfer value using the blockchain, which is effectively like a decentralized server with a very authenticated state transition system. Basically everyone can write checks sent to the cloud and eventually they'll get transacted, if they're valid checks, if you've signed them, right?

Crypto law is that. Except forget the idea of specifying the rules of the bank and now you can specify the rules of any institution that you want. And those rules can be very nuanced. The bank's rules are fairly simple, I mean, at least in terms of the Bitcoin bank, in that they, all they do is they subtract effectively the value from available to one person or from one account and add the value available to somebody else or to another account. So that's like cryptocurrency, value transfer.

And we have the notion of app coins as well. So Namecoin is the only one I could be bothered to put down there. But this is the notion of having a registration system, right? That's decentralized. So it's a specific application that's done without an application server, without a specific service provider.

And then you've got the crypto finance guys who basically deal with value transfer plus plus. So it's like value, but with very specific rules as to when and where that value should be transferred and under what conditions. Whereas Bitcoin tends to be under relatively simplified conditions that is always.

Crypto law is the idea where you can specify any rules and you don't just deal with value, but you can deal with anything you want. Anything that can be modeled in a computer, which pretty much these days means anything. You can deal with that and you can specify any rules that you want and anybody can join the party as long as they respect the rules. So that's what I mean by crypto law, basically. I mean any service that we have yet and any service that we could possibly have in the future. Any other questions?

How could we remove 100% of the human element from crypto law? Well, I suppose the question is how useful would crypto law be if there were no humans involved in it? I would say, at least in my, probably for my lifetime, we will not remove humans from this. I think it will augment it. I think to some degree we'll replace it with very clear, specific ways of interaction. But I don't see how, for criminal law we're going to remove humans. That's so much of a decision of rationale. If you read criminal law texts, it's all about effectively opinions, ultimately for most cases it's the opinion of a jury. Right.

But what we can do is disintermediate. We can remove the bits that are very mechanical in this process. And the bits, these mechanical bits are often bits that can a little too easily go one way or the other way, depending on the mood of the person involved. So for that I see a very real benefit and potential to replace. Yep.

Okay, so that's a very interesting point. So, the idea was that we go the other way from policing users and we go upstream to the ISPs and we start shutting down protocols effectively. Right.

So, I would say, that seems unlikely just from a technological point of view. Basically this would require a white list of what services were allowed. And to whitelist services, you'd effectively have to declare all content so there'd be no more HTTPS, SSL, none of that stuff. Because you wouldn't be able to tell whether what was going over it was a service that was allowed or a service that wasn't allowed.

So, for example, what we could do with Bitcoin, or indeed any other peer to peer protocol is run it over port 80. So it runs over actual HTTP, and rather than just sending the data, what we actually send is a set of words in an HTML document where each word is one of 256 different words with each byte encoded as a word. Yeah. So we have, I forget there's a great word for this and I've forgotten what it is. I learn it in my crypto. Steganography. Exactly. Vinay. Thanks.

So, yeah, there are theoretical ways of getting around this and I don't think it will come to that. I think that authorities will accept that there is a significant need for encrypted protocols, encrypted communications, and that ultimately they can't really stop encrypted communications and we'll just end up at this odd destination. But, you know, we'll see, we'll see. I might be wrong. Yep.

A contract that's hosted or whatever on the blockchain. I just can't understand how that works? Sure. So how would crypto law actually work?

So when I say it, I don't really mean it's going to be a drop in replacement for civil law on day one. What I mean by it is that it's a system that works in parallel to civil law as it stands. So an example would be, suppose I am a flour maker, flour processor. A florist. No, that's a flower seller. But anyway, it's a different flower, different flower. I mean the flour, that's white powder. Suppose I make white. Not white powder. No, that's the wrong white powder. Right. Suppose I'm a farmer. Suppose I am a farmer. I'm a farmer in Turkey and my crop is some wheat, processed wheat substance, and I want to sell it to somebody in, I don't know, Britain.

And the person in Britain isn't so keen on, I don't know, having lots of lawyers involved for this international deal. Nor do they really trust that this farmer, if he takes them to court in London, is going to see any of the maybe money that's owed.

In crypto law that wouldn't be a massive problem because we had something like an escrow system with a disinterested third party who would effectively authorize that this shipment has happened. And this third party would allow the automated payment or unlocking of payment, once the shipment is done.

So this, this doesn't need any contract whatsoever, standard legal contract between these guys for them to be pretty certain that it will work. It will be fine in the end, you know, providing that there's some base level of trust between them, that you know, one of them is, I don't know, paying off the guy who's checking whether the shipment has happened. And that can be relatively easily done by using a mutually trusted or much larger entity to authorize that shipment, that that shipment has happened.

These guys don't really need to trust each other. They can go on, they can have this contract, they can see the contract, they can employ a programmer who can check the contract and all is well. And this, this, they don't need to sign on the dotted line. They know it happens.

Now actually, I know, I know of this happening already. Not with, not with like Ethereum, not with like proper crypto law, but with Bitcoin. And this happens between China and the UK doing some relatively small volume of electronic goods. So just one more because we're running short on time.

Yep. Yeah, well, I was kind of trying to get to that in the talk. Right. It was kind of, the point was that the authorities, the lawmakers, are going to have to, I don't know, start getting real, basically, because the world is going to change around them. And if people want to act in a particular way, and it basically doesn't bother anybody else, then they're going to risk running into the age of dinosaurs and pissing up against the wind.

Okay, thank you. Thank you, Gav.