Ian Grigg – Review of the conference

The inventor of Ricardian contracts and the Chief Scientist of Mattereum gave a good summary of the conference and commented on most of the discussed topics.


As you know, my name is Ian Grigg. Normally, my main claim to fame is that I’ve been in this space since 1995; tonight apparently it’s because I’ve created this thing called a Ricardian contract. When Vinay asked me to wrap up this created quite a difficulty, because I’ve got both something to say and you had something to say, so what I’ve done is injected some of the comments that I’ve heard into this slide deck, so therefore it will be a bit of a mishmash. I understand that I am standing between you and the beer, and I did not choose that ordering [laughter], but here we are.

[Ian skips over the first three slides]

[Slide 04]

It’s all about building a blockchain, which I believe should be done from the basis of first principles, I believe we should be starting from the user and asking these questions: who is the user, what do they need and can we do that?

[Slide 05]

Today we heard that Imogen Heap is a user, she’s an artist, and she wants to get out there and talk to her audience. There might be different interpretations about this, but this is I think what she was trying to say. Can we provide it? Well, apparently Mycelia is trying to provide that communication method.


The lawyers also have talked about wanting to create these standard transaction contracts, and OpenLaw was trying to come up with a concept for that. Also, residential buyers and sellers we heard about which were trying to do this. Conveyancing is what they want to do, and Veyo is going to attempt to do that. There’s also the additional mention from Vinay, he’s saying he wants to be able to drive around London with his phone, spot a house, slap a deposit down, and 20 minutes later receive a mortgage, going ding on his phone. I think that’s a long way from where we are, but hey.


Let’s look at the options that we’ve got with blockchains. It turns out that Swanson way back in 2015 said there’s these unpermissioned blockchains and then there’s permissioned blockchains. So yeah, the ones we know about are essentially unpermissioned, and the interesting thing about that is that they do simple automated things on the chain, but complex things are left to the users. We’d really like to do smart contracts, lots of external logic; things like this are kind of left out, with oracles being needed. We want this multisig zkSNARKS thing, but we haven’t quite got there – there’s a lot of stuff that is left out. So we ended up with this layer of basic stuff where simple stuff is done, and then on top of that there’s complex stuff, but we’ve also got anarchy so it’s a little bit confusing what the hell happened here.


The answer to this was the permissioned blockchain, and in essence what we’re saying here is that trade is more complex. We want to do financial interesting stuff, like financial engineering swaps, straight finance and so forth, but this leads us into an area where we are exposed to difficulties, so our risk is going up and we need protection. How are we going to solve this problem? The answer is to put a wall around the entire thing, what we call a walled garden in the trade, so we can do more complex things in there but we’re protected from attacks, hacks, and if something goes wrong we can just ask for it back. That’s the permissioned blockchain.


Now, the problem with the walled garden is that there’s a wall, and if there’s a wall there has to be a gate, if there’s a gate there’s have to be a gatekeeper, and that gatekeeper then charges fees. I loved Michael Mainelli’s comment that monopolies, the person who is running the wall, they’re subject to indolence and corruption, and this is true – I didn’t actually dare to say that in my original slides, but that’s worth thinking about.


Somewhere between these two contrasts we’ve got anarchy and the Leviathan, this is actually the title of a forthcoming paper on this theme. We do agree, in one sense, that we ought to make everything that we can below that line, but what do we do above the line? Especially if you’re a small business, what do we do above the line? This is where for example Trent is saying we’ve got this very complicated thing, we’re driving around in a self-owning car. How do we handle that above the line? It’s now complex and we’re now generating liabilities as we’re running people over.


There’s a couple of strategies. If you’re a megacorp, you don’t change, the best thing to do is not change. All this business about large corporations wanting to get in blockchain… No, they’re actually making more money than they can possibly make by not changing. If you’re a high net worth individual, you just hold the stuff and let other people do the work. But for everybody else, we have to trade. If we want to have that fine living life, then that would create demand for products and trade and so forth. We need to generate a feedback loop where we could start to actually generate profits so we can live relatively well – we’re kind of stuck, but we have to generate trade.


Now, the thing about trade is that it’s kind of complex. If it were simple, we would automate it away, so we’re left with all small and medium employees having to stick with complex trade, which means that we’re actually living with errors all the time. What is complexity? Well, it is stuff that should be ordinary filled with errors, and we heard a lot about the errors from the various speakers. We’re in a yin and yang society where what we want to do is actually unpredictable and hence we generate that need for care, and therefore our customers come back to us because we sort out their problems, we sort out the complexity.


We can analyse this from the point of view of formal methods and software verification, that was this morning’s session from Christopher and Stephen. OpenLaw, we’re talking about we can use logic to track mistakes and I think that’s actually a good example, and I felt a little bit shall we say aggrieved by the comment that somebody could replace GBP by LBP. I mean, how could that happen in a modern-day system? That is an error, where a person would have to work extremely hard to make a mistake like that; the software should solve that problem for them. Then there’s model checking, CTL stuff – we could talk about a lot of technology there.


But what this does is it creates a substrate of working software. Around that software there’s still huge potential for error, and this is unpredictability, what’s left is always unpredictability and there’s that list of things over on the right here. Norton Rose Fulbright said all this unpredictability gives us both upside and downside and we need to analyse those risks. Scott said that if we didn’t have dispute resolution, it’s kind of open ended what would happen at the end, so that generates risk. It could also generate downside risk in the sense of value, if somebody can’t come after you. What happens there? Well, small business analyses it subconsciously and learns to live with the risk. I want to drift over to IT at this point, because IT is what builds the blockchain, and therefore they have an outsized view on how we’re going to deal with this future.


If we think about a model of risk, this is a standard risk matrix. Basically, what we’re saying is the severity goes up on the left, and as we look at the probability along the right, that right-hand top corner is dangerous and the left-hand bottom corner… Well, we’re not that fussed about it. IT people look at this, and the first thing they say is, “Well, the risk can be carried by the users.” I don’t think we can really deal with that today, but moving along… [slide A] What they do say is, “We deal with the stuff we know about,” and that’s the stuff that is highly probable over on the right-hand side, so we can deal with that sort of stuff. [slide B] If we can’t see it, we don’t believe it’s going to happen so we’re not going to deal with it, and that’s the stuff over on the left-hand side; it’s not happening, therefore it’s not going to happen. Unfortunately, some of it is very severe. What do we call that square? We call that square the black swan.


It turns out that in Bitcoin or blockchain and so forth we had a series of black swans, so actually it’s a really dangerous place, we’re losing a lot of money. The record is still held by Mt. Gox which lost something like $600 million, but there’s a lot of other shocking things going on.


Coming back to our small business, our small business would like to get in and invest, and small businesses invest serious stuff: there’s programmers to buy, there’s business and legal support, accountancy… You need a fair amount of cash, a standard startup is going to have a lot of cash, and it takes time – three months, two years, whatever it is – we call that opportunity cost. So your small business is actually not that keen to invest in blockchain because it’s too dangerous; the investor is going to hold back until we solve that problem. I put it to you that what we really want to do is build the blockchain for small business, and that means the one that solves the black swan, that means the one that permits an entrepreneur to sleep at night and not have to worry too much about a meltdown in the morning.


It’s a really stark choice, we’ve got lots of things that go on and people talked about that. But how do we solve that? The short answer is we get rid of anarchy. We use trust, in real life we use trust, which is where we’re circling back to the walled garden. If we have that trust, in essence what happens is no matter what you think about trust, the business with trust will outcompete the one without trust. But the problem is that exclusion of the walled garden.


Let’s go down a little bit deeper into what trust is about. Trust is expensive. It involves a continual analysis of what’s going on, which means that you can use that analysis to inform your decision and that creates risk, but it also creates reward, and this is information which you then take into the next round. Consequently, you need to do both multiple rounds and investment, and you need to extract some sort of profit from each round, otherwise it’s not worth doing. The theory of this sort of thing is that we need punishments outside the game. Consequently, what we need is an environment where we’re actually doing a lot of trades over and over again with the same group of people.


There’s two aspects we can look at here. Negotiation theory: the entrepreneur wants to do win-win, but what they get with blockchain is win-lose. This is one way to think about the way blockchain can be perceived by business. Another way is to go to the theory of game theory, and it’s the same thing: we want a thing called net-positive game, but what we get is zero-sum game. In theory this is often called a stag hunt, it’s similar to a prisoner’s dilemma, but basically what we’re saying is we need to move the conversation from a blockchain that’s focusing on win-lose across to win-win.


In essence, coming back to that previous diagram, below the line what we’re saying is it’s a zero-sum game, it’s a win-lose game. When you’re talking about the blockchain, all of the extant, successful applications, you walk in with a set amount of money, both parties, and you walk out with exactly the same amount of money, minus your fees. Whereas what we want to do is net-positive business: we want the entrepreneur to walk in with a certain amount of money and the customer to walk in with a certain amount of value and to walk out with more value, so we need to win above the line and somehow create that business.


Why have we got a problem with this? It turns out that the zero-sum game, which is a valuable game for some contexts, has actually earnt an over-the-top reputation. The blockchain culture, the business of blockchain has decided that this is how they’re going to do things, they have an oversized belief that the zero-sum game is the future. They also said that when it comes to ordinary business, you should be thinking about applying everything in this fashion, and that’s a problem. If you look at the way zero-sums work, you’re actually causing a problem – there’s a list of things that go wrong there. In a sense, what we’re saying is we’re attracting people who are used to a zero-sum game, which is a certain crowd that is willing to take risks and also to, shall we say, make your loss into their win. I guess my point there is that we’re actually making a toxic environment.


The alternative is to go for win-win, I mentioned before that equations of win-win come from business and so forth. We’re looking for a scenario where we get repeated trades. We need identity in there, so we need to remember who we’re dealing with, we need a set of rules about how to do this, and we need that way to trade – that’s fine, we’ve got that with the blockchain – but we also need what we call skin in the game, a way to hold an aggressor to account.


I thought we had this with the walled garden, but it turns out slightly different. We have that set of rules, the method of applying the rules and the consequences, skin in the game, but we have that problem with the wall.


If you compare that to what I’m looking for, we want many rounds, no end, and we want to know who we are dealing with.


What I’m saying here is that we don’t need the wall, we can get rid of the wall. All we need to know is a way of knowing who we’re dealing with. Now, identity is probably beyond the scope of today, but that’s a big thing that groups like Mattereum are going to look at.


[[Just to rush in quickly through here, because I know that the beer is more interesting.]] How do we build this thing, this governed blockchain? We’re doing this in one project, EOS – which is what I work with, but there’s also platforms like Mattereum – and you start off with a notion of constitution. You take that constitution, you turn it into a Ricardian contract, you put it into the substrate of the blockchain so that every transaction that goes in signs off on the constitution; consequently, everybody has agreed to the constitution. This is useful because we’ve now preserved free entry, but as you come in you have agreed to that basic set of norms, basic set of rules; you’re not blocked, but you’re one of us when you come in. That allows us to do something that’s a little bit different to the current excellent blockchains: we can now define our community. Our community is the people who have agreed to the constitution, therefore we have a defined membership set, and we could also know that when we’re dealing with somebody on the other side of that blockchain, they have agreed to the constitution.


Once we’ve got that constitution in place we set in baseline rules, and the specific things here that are of interest is arbitration, we can use arbitration to establish a forum to resolve those disputes. We can also use it for various other things which make the blockchain work better, including for example changing the constitution.


Which means that we now have a situation where we’re doing something different to the unpermissioned versus permissioned dichotomy.


If you like, we now have a third choice, which is what I call the governed blockchain, the difference being that we preserve that free entry, but we’ve established rules of behaviour which has been attractive to the permissioned blockchain group.


That’s just more detail [skips a slide], and that’s about it.


Download transcript