Public vs Private Blockchains for the Enterprise

This is a very special podcast for us as it is our first debate format on Insureblocks. We were extremely privileged to have two titans/Jedi Masters from the blockchain community join us for this podcast. On one side we have Richard Brown, Chief Technology Officer from R3, builders of Corda, advocating for enterprises to use private blockchain systems. Whilst on the other side we’ve got John Wolpert, Team Lead at Web 3 Studio at Consensys, builders of Ethereum based blockchains, advocating for enterprises to use public blockchain systems. - Insureblocks - Public vs Private Blockchain for Enterprise - APRIL 7, 2019

Transcript

This is not a perfect transcription, but I took the youtube autotranscribe, and quickly broke up the text so it would be easier to navigate\reference. I also snipped some of the introduction banter and skipped ahead to the debate.

Normal text is the host, Walid Al Saqqaf.

Italic Text is Richard Gendal Brown.

Quoted text is John Wolpert

Introduction

[Music] hello hello hello hello welcome to Insureblocks, your dedicated podcast to blockchain and smart contracts in the insurance industry. I’m Walid Al Saqqaf, your host. this week we have a very special podcast today. we’re going to be debating the public vs private blockchain for enterprises and we’re very lucky to have to thought leaders or as John would put it Jutta masters and I’m very pleased to have returning to Insureblocks Richard Brown chief technology officer R3 and for the first time John walpert senior product executive at consensus.

[ …snip… ]

thank you very much so to provide some context on the purpose of this debate podcast it kind of all started when a consensus article was written on the 3rd of January 2019 entitled quote busting the myths of private blockchains and close quote on the 14th of January Richard Brown posted a counter-argument entitled busting the myths of public blockchains for business and Insureblocks produce a podcast regarding that blog post along through Richard and the 23rd of January John walpert wrote back an article called these errand the enterprise blockchains you’re looking for.

in five minutes could you please each present your main arguments on why you believe your public or private blockchain is more suitable for enterprises.

Richard

…and you have five minutes:

So we’re going to get onto the specific point raised in the articles a bit later so maybe I’ll start here with almost the take you on the journey we went on when we designed Corda you think about where we’ll be whereas in the world three or four years ago it was a pretty brave possibly people might think particularly mad idea to to build a new blockchain platform from scratch when there are so many already in existence.

The reason we did it is hopefully is interesting and maybe directive for your audience so our starting point was what would be possible what would the world be like if I could have my system or in a cloud system that’s under my control what the world be like give if I could be if I could know if I could be sure that when it tells me some things about deals I’ve done with my counterparts or if money I’ve lent or borrowed or any kind of fact I care about if I could know for sure that what it told me was the same as what my counterparts or my competitors computer told them clearly

if I had that kind of ability it would probably dramatically reduce costs complexity risk duplication all these opportunities for things to go wrong would be eliminated all this reconciliation we do which suddenly revealed the right answer rather than the wrong answer.

but it would go further than that if I knew for sure that the data I was running my business on was in sync with my counterparts and therefore was correct just think how more quickly I could make decisions how more how more quickly assets could move how more quickly deals could be done because I wouldn’t first have to go through this endless process of checking to make sure that I was operating on correct information

so we started almost from the solution doesn’t domain back to the requirements domain and said if I had this requirement for a platform that allowed me to build applications that were in sync by design with all the other people running the application how would I do that what would a platform that allowed you to do that need to look like how would it work?

and a few things sort of emerged from that the platform would need an identity model. if I was trying to model the fact that I’d lent some money to Willy door he’d lent some money to me when when I send that data to him to get him to sign it to say yes we agree this deal is correct I need to know it’s actually going to him and not somebody else. so there needs to be this this this first-class concept of identity at the heart of the platform so I know with whom I’m transacting.

I’d also need some notion of privacy is the fact that I’ve entered into a deal with Lalit well why should Jon know about that or anomaly dive entered into a deal why should I know about it so it kind of felt like privacy and an information only going to those who need it should be a foundational concept not something that we we add on at the end so we knew that not sending data to everybody was was a kind of like a non-negotiable requirement

now that easy to say but it was highly controversial at the time if I remember sitting down with the architecture working group and many of the other participants in our process at that time and it almost felt like I was new I was I was slaughtering these sacrificial lambs. I was saying things you think are innate in blockchain we have to question them to to to achieve our ends because what that leads to is probably an architecture and it’s the one we ended up with the urban Corda and in an architecture where messages and data are broadcast or gossip to the entire network but instead they’re sent just to those who need to have them

now as John will probably drill into a bit later need to know is quite a subtle concept so I don’t want a gloss over that but that the high level the idea is the data just goes to those who need to check that something’s correct and people who are not parties to the deal and don’t care about it shouldn’t get to see it and then you go on and you say well if I’m trying to enable people to run their businesses more efficiently and then from that more efficient platform then be able to do do new things and take new opportunities

I have to think about the people who were building these applications they have to be able to do it in in languages they understand in languages like Java Scala currently in any kind of modern high-level language the business people I know tens of millions of a business programmers know how to use the platform has to be scalable we have to be able to process huge numbers of transactions and perhaps perhaps most critically and this is a point I got onto in my article if we’re dealing with business transactions were real money’s at stake and we’re but were whether I’ve paid will lead back or I haven’t or he hasn’t paid me back or whether something’s happened or not is something we care about we need finality we need to know that when when we believe something has been done it can’t become undone.

so the idea is we’re with Corda, and what we are trying to achieve with it is you can build business applications that keep counterparts and participants in sync they can do so in languages they understand in a way that gives them privacy in a way that gives them certainty that when they think something is done it really is done and that’s that’s what we’ve been building it’s it’s growing massively people are really embracing it and I’ve been really pleased with just how fast it’s grown as an open source community

excellent thank you very much Richard.

John

…you’ve got five minutes:

so should go back in time to 2015 when I was trying to convince Richard to stay at IBM desperately I think I even offered you my job, didn’t I? I said Richard this is your chance to change the Internet, then Richard said that’s what you want to do John. what I wanted to do is his work is really influenced the world of finance and make a mark there so I know it was a good lesson in intent right and I think Corda is a beautiful approach to that intent writing at work, works on Finance and I see the evolution of that now it’s looking at I’ve heard the phrase the Internet of Corte

so maybe we’re converging, and the point is that back in 2015, I my team and I embarked on creating Hyperledger Fabric Richard when created Corda and with their team I do think it was a missed opportunity because an open source we probably could have been more in the same repo competing coming up with something that worked for everyone back in those days but yeah in those days we decided both of us I think Richard keep me honest here that we started with things like Ethereum and we loved it and that public Ethereum community really had some things that it needed to work on like scaling and we were focused on things like peer-to-peer channels that can can have a lot of confidence, I’ll simply say privacy, but confidentiality that is not only it can mature the transactions not be seen by parties that are not counted that are not involved in the in the transaction but the business logic should not be executed on on every node in the network even if it’s a private network

I’ve been saying that blockchain is a nudist colony and the only difference between private and public blockchain is whether or not you’re naked on a private beach or a public beach but either way people are gonna see your private parts and that’s not ok if even if you’re in a group say in a supply chain network or an insurance network that insurance in particular there are a lot of agreements between party and party B that cannot be seen by party see the logic itself behind the agreement the existence of the logic cannot be known to party see this often occurs in health care Richard would you agree?

totally I mean just making some notes as you were saying and the if it was go back to the comment you made about when we were both both at IBM and you write at the time as I said in my intro R3 was very much a finance orientated consortium and it’s kind of ironic, I call it serendipitous usually but it is also ironic that I went down that path thinking I was building a blockchain for finance and then woke up probably about probably about 18 months ago to realize that I didn’t advertently built a general purpose platform instead Brian rug who leads our insurance industry she’s based in New York which is actually in London this week and will lead knows her well I remember her coming up to me shortly after we hired her maybe two years ago now I was saying Richard you don’t realize you keep talking about banking but you were built the best platform for insurance and I said yeah yes stop flattering me that and focus on banks and but it’s a there was this probably hasn’t been revealed externally there was there was a big internal debate because it took it to it’s anything me cuz I’m often blinkered took me it took me a while to realize that what we had built something that was more general-purpose than we’d expected and we were actually doing certainly ourselves but also the community disservice by constantly talking about it in financial terms because we were inadvertently signaling to other people in other industries it’s it’s not for you if you’re not a bank when actually as we’re now seeing in in health care insurance trade finance supply chain you everywhere we’re seeing it get adopted but um but but you’re right you join a we joke but but it genuinely wasn’t the original intent

I should probably say for the record here I have been always have been my inception into blockchain was with Richard and that I am a Gendalian I’ve been I’ve been making fun of Richard by calling it the Gendalian architecture and the Gendalian architecture is basically the assertion, and I agree with it, that parties that are not counterparty to an agreement should not neither execute the code of that agreement nor even know about it nor store the state transitions of that agreement and so I think we agree on that right yeah totally totally and you clearly are public watching that’s an interesting unto itself I think my case here is not about is not about whether private or public actually I believe that the emperor has no clothes in both private and public that either one of them is insufficient to the task of evolving the Internet and that what we need is to go from this divergence that we needed to do in 2015 now toward back towards convergence and I wish I could just find all the humans and take off their flags and make them forget they ever had them and start to work in open source on saying okay well that idea in Corda was pretty good idea bad idea in Fabric was a good idea that idea in theory emits a good idea

how do I make these things work together in a in something that looks more like a seamless protocol so that we don’t have a balkanized slagheap of of a mess going forward yeah I think we can also agree that given the traction that Fabric has been having, and also Ethereum, that the world isn’t going to be an Internet of Corda unless something radically changes… so it’s gonna we’re gonna have to work together right?

[ …snip… ]

My point was if you if you get a deploy whether it’s it’s so different deployments of Fabric or different deployments of Corda or one of the enterprise Ethereum variants even if you’re deploying it for one purpose with sort of supply chain or insurance if you can agree as a group unless is that almost as a community up front to some common identity standards common governance common protocol rules - in the way the public Bitcoin network does, and in the way the public Ethereum network does.

the opportunities for interoperability within that shared network are so much greater because you’ve pre agreed your pre committed to so much and in effect and of course this is what makes it hard you’ve you’ve deliberately upfront really relinquished your right to deviate from certain things in order to maximize the chances of interrupt.

what the public in Topeka Ethereum Network has achieved multiple different applications on the same network we should strive and aspire to do the same thing with the with the enterprise block chains with whether it’s about brakes or tooth Corda and we shouldn’t deploy them as standalone states

actually that’s I had read that part of your writings and that makes a lot of sense to me in in a way that I hadn’t really apprehended before so thanks for that

but the notion hope though is yeah when people talk about security I find that they’re they’re thinking data breaches Frank what happens to companies like Facebook and others when when we find out about these breaches or these misuse of data but really the security is two things right there’s tamper resistance and there’s surveillance resistance for debut avoidance of data breaches right now you want to and of course a private blockchain certainly Fabric style and I’m starting to really rethink my point of view about Fabric that I like to Fabric channels for a while but the more I dig Richard into the corridor model and like Clement Swan is is pretty knowledgeable about it I’ve been talking a lot about this and also other people from hope we have a lot of knowledge about coordinate is that it Cordas flow system is a lot more neural plastic and

as a product guy which I used to be a CEO and when you’re not a CEO you’re a product guy so if you had I know a few things about the code but I think of myself more as a product person so what are the requirements right my requirement is I should be able to find you a registrar preferably a public permanent registrar right one that isn’t controlled by any group of companies or can’t be Jack taking control of by some oligopoly right that public registrar basically an improvement in DNS will put you and me together in a very very very private thing that we can be confident isn’t be jacked and duped by anybody else and we’re gonna conduct some business logic I’ll give you three cookies for four donuts but if on Thursday at noon you’ll give me five cookies I’ll give you I’ll give you twelve cookies if LIBOR goes over twenty percent we can make all these rules right and those rules should be known only to you and me

but the problem with that is that state machine then under traditional techniques is very mutable right weak you could chamfer with the code or I could tamper with the code or the administrator at the on the on the measure or AWS or what have you node or a host will can tamper with it or throw something in between the hypervisor and the stack and sniff out our information right so I want that bilateral or multilateral relationship to be just between you and me but I want it to be a be immutable can’t be tampered with and I sometimes in fact I’m working on a project right now in aerospace where that needs to be military-grade Trent tamper resistance and last checked there is no such thing in private that looks like military-grade tamper resistance

the closest thing we have to that the gold standard is a large globally distributed public main net now the problem with that is that the public main that has no surveillance resistance everything is done by everybody and you don’t even know who those people are so I think wouldn’t we agree that horses for courses as they say in Australia the right tool for the right job I want to do as little business logic and actual point-to-point communications on a main net a public main that is possible but I want if I can to get these these state channels these little these little state machines between you and me or you and me and a bunch of other people to be as tamper-resistant as possible while also being surveillance resistant but then then this is the last problem is I’m going to need if I’m a good programmer to pass parameters and to send into to return stuff from functions on my little state machine and send those two as parameters to functions on other people’s state machines and how do you do that without a common root core machine

[ …snip… ]

well lead we’re not we’re not buying cars here right we’re talking about open source protocol level not even house plumbing we’re talking about in the street plumbing right and just having to choose Fabric versus kurta that’s a sad conversation to have but we should be talking about is okay my business I’ll give you an example I’ve got a we just released something called bootleg yeah which is a really cool application and it allows royalty it’s basically a royalty sharing Platteville assistant and for musicians or even oil and gas and it could be insurance where anywhere where I have to share realties according to a royalty distribution formula for future cash flows that are coming in to some kind of an application or system in that in that application if I’m a music studio there’s no way I want the royalties that I am receiving from these cash flows people buying these these these videos or these music copies to for people I shouldn’t my competitor shouldn’t know what I’m taking in that royalty right and if you put it on a public blockchain you would you’d be in a nudist colony people would know they would see it in the logic that could decompile the smart contract if it’s in a private blockchain like Fabric if you’re all in the same channel same problem what court is cool great about is that it is really fundamentally point-to-point but then it has other issues with sharing state right so if I want to share a flow from kurta from a couple of parties in kurta to something that’s working on a theory a more on Bitcoin or what have you then I then I’m gonna have to deal with that right and that’s the problem there

so the what I really want is to be able to have shared state on a public network that says yeah we all have this royalty token but their business rules for who gets what in that royalty scheme need to be done on a point-to-point basis what I really want to be able to do is push some logic down into a Corda network to run that logic and then have that return to the to the main net with the shared state result right that way I have privacy and I think much more importantly confidentiality along with global military-grade tamper resistance and I’m really curious Richard to see what you think about how we could make that work so Corda and Ethereum can work seamlessly together in a stateful internet way

Richard

it’s yeah it’s it’s really good to hear you describe it that way John because if I go back a few minutes to what you’re talking about which I think you summarized at the end you it you’re talking in terms of tamper resistance and surveillance resistance and I don’t tend to always always think or speak in those terms but but I think it’s quite helpful. so if I’m transacting with you then the only people who should know about it are people who have a right to know and that was kind of like a foundational requirement against which we engineered Corda so I agree with that.

we’re trying we’re trying to tease out differences in areas of agreement in this podcast so I’m not trying to pick a fight but but there is a but I think where tamper resistance is it’s probably where we may disagree because of course and the thicknesses think something that people sometimes sometimes miss totally point to who will come into this quite quiet down position of novelty because none of these platforms are actually tamper resistant the data resides on people’s computers human beings who own computers have got data on their computers there’s nothing to stop me going into my computer and changing any piece of data on it I want just as there’s nothing stopping Willy changing data on his computer nothing stopping John changing data on his I think the requirement we actually have visit guess is almost tamper-evident so that if I try to convince you something is correct you are able to detect if the if the presumptions or the inputs if you like that I’m relying on for my argument have been changed or aren’t pre-agreed then you can spot that and therefore know that I’m trying to trick you and that for the listeners this is this is an important point because all these blockchain systems whether it’s called a Fabric Ethereum, and whatever it is that kind of like that kind of like logic engines that based on almost like inductive logic for those of you who remember your maths lessons back at school everything you want to do it is almost almost like that the grammar the language of everything that happens on it is

I want to convince you that you now owe me this much money and you will be in no position to deny it because I’m going to do the following thing I’m gonna say the last thing we agreed on is this and I’m gonna prove we agreed on it either because it was signed by lots of miners signed by a notary pool whatever it is but this is the last thing we agreed on pair the rules we’ve already agreed to know the contract we’ve already signed the code you’ve already signed up to here is some new evidence or some new data and per those rules this is the proof that I can now change that previous fact that previous deal into something else and because we pre agreed what the rules were would pre agreed what evidence needed to be provided and we had a shared understanding of what the previous state was and what the previous fact was we can each evaluate that and we’ll both reach the same answer and one we’re temporal resistance

I saw that with EDI can do that with with any often say if you could bass you can sass right I mean if you can if you could have a blockchain as a service and you can have a you could just probably you’re happy as long as you can call it blockchain so you can get funded because that’s where we’re at right now but but if if you can call it an Oracle a nice traditional Oracle database I don’t mean Oracle’s I mean Oracle as I’m Larry Ellison Oracle database or the cloud implementation but if you can call it blockchain but use a traditional system very often I think that’s probably just fine right because what people really want is to be able to collaborate we really been able to collaborate for most of my 30 years in business it’s just that we really haven’t prioritized that is what’s a new so okay so this would that’s exactly so that’s exactly the point

to get back on to how in that scenario I just described how could I do fraud well how can I convince him of something that’s not true but make him think it is true and it’s backed it’s back to that temper resistance or tamper-evident piece if it was just a database in the cloud if I could hack into that or I could persuade the operator to change something well these would have no way of evident excuse me no way of evidencing that the thing that was different with blockchain on this achieved in different ways with different trade-offs between proof of work log chains and notarized block chains or depending on how you want to call them is how do we establish what the previous state was what needs to have been signed what needs to have been minds or notarized or whatever it is so that if I try and say - we’ll eat here’s what we previously agreed and here’s the update what do we need to have agreed up front so that when I say that to him he could say hang on that isn’t what we previously agreed and here is the proof that we actually agreed something else where is the stamp that says the previous confirmed transaction was X and the point I was trying to make it you’re not worth know we can do that in many different ways you can use cloud do you yeah yeah

that was a point of my blog post which was proof of work blockchains the way they achieve that is by saying the most that the largest number of miners have been effectively voted on that previous history and because we believe and this is where the belief piece coming in because we believe the mining pool the number of miners mining the etherion blockchain of the Bitcoin blockchain is is mostly honest and highly diverse it’s unlikely that they would be feeding a false history to me and I’d come on to some of the pros and cons of that an alternative and this is the one that through through this lengthy conversation with me all the firm’s we built Corda with or as they said what there’s the probabilistic element of that the fact that history could be could be reversed slightly and things you thought were agreed become an agreed because you don’t actually know who is signing or mining these transactions you don’t know whether the probabilistic

John

yes yeah yeah but you again we both know this Richard that the laws of physics do not allow for instant or fast finality or one block finality systems to scale to massive numbers of eyeballs right you can R be up to your pbft they have very serious limitations now you get around that by just basically being almost a message bus protocol right you’re you’re you’re you’re just going point-to-point shared state is more challenging for a court of the opposite problem for Corte shared states easy for a public network shared state for court is a little harder it’s not impossible but harder point two points really hard for a public network in fact virtually impossible but having shared state is easier and so again it’s horse it’s it’s right tool for the right shoulder

I mean you’re going to have the ultimate notary why should why wouldn’t we make the ultimate notary a notary of notaries a basically a table of pinning contracts the Ethereum main-net that’s my point right so that we can have our cake and eat it too is it where we can have this beautiful private state machine where for a couple of parties it’s kind of okay that it’s theoretically yes you have evident you could you could say that it’s not just do no evil but do we you can do evil but I’m gonna find out quickly that you did evil and that might be okay for a couple of parties but we’re talking when we’re talking about the internet of transactions being possibly every transaction you cannot leave the ultimate vanishing point of control in the hands of a group of companies or governments or any other group right you need that vanishing point to be as decentralized and politically outside of the realm of politics as possible.

Richard

I do I’d agree with that but it would be it would be a more powerful argument if if let’s take Bitcoin let’s take the let’s talk about a different platform if if this supposedly highly decentralized confirmation process on one of them on the world’s largest block chain network we’re not we believe in the hands of three four five huge huge firms in China so there’s there’s theory under the ideal and then there’s the practical reality and the implications of the economics so they should do it nobody knows it’s mining

so the very specific point I’m making because it was kind of like the heart of my sa view like was this whole blockchain edifice this hot certainly in the business base this whole edifice that I can convince you that what you see is what I see without having to go through all this endless reconciliation it rests on us being sure about what it takes for the previous faction to previous second the Lord stood we previous step in the logic to have been confirmed how do I know something won’t be undone just finish the point so there’s so there’s and there’s fundamentally two ways of doing it you either go with the effectively what amounts to the proof of work and then the proof of stake approaching in Bitcoin or Ethereum which is just as an aside is no it is it was a jaw-droppingly amazing advance in computer science

when I first read it and got my head around it it was amazing and it has the potential to give you that decentralization you talk about and to avoid being beholden to any small number of firms but the practical reality of the empirical evidence of the last ten years is it hasn’t yet worked out that way in reality is controlled by a very small number of parties who nobody has any governance control over and the cost that comes with that through energy usage and the possibility of reversal through probabilistic finality makes it highly just highly difficult so the alternative approach and I’m not saying it’s appropriate for everybody is to say what this is hard the probabilistic piece is hard to deal with the fact that we don’t know who’s providing this consensus is very very trouble trouble some from many business use cases so why instead don’t we say what let’s let us surface this it’s actually make it explicit who are the five seven nine organizations who who are who are providing this consensus service how are they selected how are they governed. how are they overseen how many of them have to collude we’ll go bad in order for a problem to occur and little and then then that may then let people make business decisions about the extent to which they’re willing to trust them and the extent to which they’re willing to to run business transactions that so I’m not saying either is perfect but we shouldn’t I don’t want the listeners to believe that the decentralization of networks like a Ethereum a Bitcoin also Maps down to the decentralization of consensus forming power because it doesn’t it the economics makes it highly concentrated and all oligopolistic but with zero transparency well

John

so first of all when you and I were both hanging out together a few years ago yeah we both loved where where the public network was going and as I said the mining you can I look I sty I was one of the cofounders of Fabric and Hyperledger I definitely was in that camp I used to say things like what you just said and now I’ve learned I was pretty happy back at IBM why did I come over to to consensus well because again I’m not more interested in internet style application building rather than the hell of building consortium.

I’ve spent him a good chunk of my last 30 years trying to build heavy-handed business consortium and blockchain or no blockchain, distributed ledger no distributed ledger it is a mess and hard to do but if we can say nope I can it’s the Internet’s too but we all have control that none of us has in control of the system truly right and show me a blockchain business, a blockchain as a service network that doesn’t have a hegemon calling the shots getting everybody together and having a strong hegemonic control over that network I have not seen it yet and I’ve not seen any blockchain as a service network that has actually given their administrative keys over to the participants except one CLS right and that and that’s why I was on the record saying CLS was a really good example of using private watching because those parties would not have worked together if they didn’t they weren’t able to say that each of them had their own control their own nodes or their peers in this case

so I think that’s that’s right but I’m more of a guess and on this that again I would not put the crypto kitties in the future for example for people that know crypto kitties which is a popular game is not gonna be on a public main net it’s gonna be on a side network like me maybe something more like Gordo or more like Fabric or more like an ethereal side chain you’re not gonna put those things on the main net it you’ve just it if if the whole world was doing that one ledger is not big enough to do that but one ledger is big enough to coordinate the other ledgers and this sounds like the one ring to the rule of them all kind of story but honestly if we’re not going to have a giant mess in a battle at the gates of border over or whose protocols more important and rather say ever all the protocols have different benefits for different use cases but they need to be able to pass parameters to each other without getting into deadlocks and race conditions then they need a coordinating route chain one that’s running pinning contracts and I see that at the Ethereum main that is the way to do that so maybe this thing

well I’ve done I’ve got this result and now I’m gonna pass that result of these other things and I’m gonna wait until this block and two two before I’m gonna reset myself to my original state you can’t do that unless you have a common frame of reference

Richard

so maybe maybe I don’t want to go too far too far away from from elite audience but maybe there’s a place we can’t agree them which is if because you isn’t you made to you made to argue my limit to the two pieces your argument there John, which I agree and one I didn’t one was it wouldn’t it be great if there were some some service somewhere some decentralized service to which I could throw a transaction of any form code or Ethereum whatever to which I can throw a transaction and this is some validation to make sure that the wrong the bad ones get get filtered out and then at some point later I’m told that thing is now confirmed everybody in the world who needs to know nodes but no it’s been confirmed and it can never be undone and there’s nobody who can exert undue power over that so we know it’s truly beyond that consensus forming process and the inability to reverse it is beyond the power of its probabilities

and I should I should I should hasten to add though that also that that level should be extremely constrained right that’s not something that you do for everything most business logic should be gone done quite a point I’m sure no surely I wasn’t I was I was simplifying even further or saying if if if if someone had one of those for me I’m a buyer if I had something to which I could throw transaction and it came back completely confirmed with no chance of reversal and there was nobody I didn’t need to know who’s running a pad but I knew that they couldn’t subvert it I would be a buyer of that

where I disagree to the argument was you then made a logical leap which was to say it’s it’s the Ethereum network as it currently exists and my org how does a girl kiss because as its evolving we’re all we’re all working on the future and that’s not for today okay but it’s not a good good use of our time yeah but even then so so if that if that were in prospect or if I were truly believing that we’re in prospect I’d be I’d be all in because we could delegate some of the some of the some of the notarization pieces of call it directly to that the other pieces of my article I was trying to trying to point out where actually I don’t see that in prospect anytime soon

there’s I know seven six seven eight massive stages and upgrade to the Ethereum Network has to go to go through to get to that point but there’s no guarantee that the end state one will be one will be achieved will be reached and two that it actually will deliver what he says it’s highly risky risky cutting-edge R&D.

John

so well so we’re all in that richard great Corda, Fabric we’re all and anybody that says that we’re in production in anything that looks like what we used to think of as production is selling a bag of tricks right where none of us are in production the way we used to think of production. we’re all yeah maybe the transactions are going on but they all have training wheels

if I could sorry quickly interrupt some silly conscious on time so I had a Marco strat which is a CTO of b3i and on just to clarify so b3i I actually uses Corda but he asked an interesting question which I want to put to the bow so if you need to keep each have perhaps one minute to answer it which is considering their regions of both of your block chains and in your history and where they’re heading is it possible to assume that the differences between public and private blockchains may shrink if that’s the case is there room for one for more than one protocol or perhaps one that could sit well between the both of you

yeah I think that’s the point is that there is a there is a there that all of us should compose a protocol that is if I need a Corda flow for this thing I should do that if I need to pass parameters back and forth between that flow and something that isn’t on Corda I’m gonna need a coordinating chain for that Richard and I just to close out the point about finality, yeah we should probably take this offline I’d love to talk to with you at length about the from my point of view eventual consistency has its use and it’s not at the coalface of the application because very often I need fast finality there right but I can do fast finality in your flow on a corridor network or on any other network tender mints pretty one block finality right but I can also put a slip transmission in there between that fast finality and an eventual consistency machine that gives me global state assurance of non temporariness and a few or a lack of tamper and if if your argument is that you’re worried about a bunch of miners again that there are upgrades to that I’d say anybody worried about the decentralization of Ethereum in terms of mining should look at what’s happening with yeah proof I think it’s vitalik was right you have to start with proof of work and then you move to a more sophisticated proof of stake system the that were where it’s it’s a lot more nuanced than just proof of work or proof of stake and then you want to layer in permissioned nodes and permissioned why can’t I haven’t run a mining operation that is actually identifying itself why can’t every company have a miner on the on the main net as well making sure that the hash power does not stay and fall into the hands of parties that are that we don’t know about right so we have just that sort space we have that trust lessness part we have the finality but if instant finality part but we also have eventual consistently because you cannot scale instant finality systems for that kind of

Richard

*that kind of this doesn’t quite answer motivation but I kind of kind of feel I wanted all the best of all but kind of like the worst of all worlds which is stuff starting with starting with something which we all agree doesn’t provide finality but is extremely costly and as as a questionable governance moving to something in the form of proof of stake that’s not yet proven may work may not when you will see but then you said then why not then layer on miners who are identified it kind of feels like now adding complexity to complexity rather than rather than it feels that way so I’m copying rather than the saying what if If I transacting with you want to know that the transaction can’t be reversed in the following nine twelve thirteen odd number whatever parties that’s who we will rely on as there is the independent arbiter of what happened in what happened I think given that we can’t make any firm statements about the probabilistic systems why not go straight to that rather than layer it on is like Stage four of seven elsewhere but but Tamaki’s points so do we think these things will converge that the bid I actually didn’t I didn’t quite agree with the Marcus’s question he talks about he goes back to the original question of public and private blockchains I’m not actually sure that’s the distinction because you look a Corda network you got accorded on network you could I don’t know it’s effectively a public network it is openly governed all the all the information there on how to have to take part in its governance is there anyone who can prove their identity can join there’s no nothing stopping anyone be able to join this it’s it’s a it’s a public network albeit one that meets the requirements we think the business needs density finality point-to-point and all the rest of it the key distinction for me is how does that governance work do who’s confirming transaction did who gets to set the network parameters if there’s a hard fork is there a cabal of developers forcing it or is it openly debated by the users if that’s what I think is important but do I think there’s room for more than one of course coder is not designed to solve every problem just as no other platform is a totally agree with John this is their horses because he’s choose the right solution to your business problem but we should do it on the basis of him what’s my requirement and what are the properties or what actually gets delivered by any given platform today or is in his roadmap for tomorrow because need to be really clear on those things so gentleman we are out of time

Concluding Thoughts

so if you could please perhaps give each a one or two minute concluding thoughts to this podcast that would be great I’ll be very grateful about it John if you wanna have a go ahead I give you the final where to go so first of all I believe thanks for thinking thanks for putting this together John’s right he and I used to speak so much and we’ve both been so busy but like following our own track we had a good chance to catch up Susie almost felt like they want to our old oral chat

concluding thoughts really just reiterate what I just said I outlined at the start of this this podcast what problem we think we’re trying to solve with blockchain in general with Corda in particular and if your listeners if you’ve got if you have if you’re in a situation where you are trying to do business with your counterparts competitors peers customers and you and you want to reduce the amount of error you want to reduce some of the situations where you disagree about things and importantly you’re doing it you want to do it just to unlock new opportunity to be able to move faster move quicker and then seize new opportunities then blockchain in general and I would argue Corda in particular is what you should be looking at that is the that’s why this this industry exists I won’t rehearse the reasons why we went with the design we did but I just also just pick up on something that John said as well it’s the truly open platforms that are are going to succeed so interview listening to this wondering which one to go with I’d love it if you chose Corda but whatever you do for for your own sanity and that of your your employees your employers and your customers choose an open source want whether it’s Fabrics or one of the open source etherion variants or Corda because there are some there are some close source Ethereum enterprise variant so you have to be wary of but choose an open source one because of the network effects are so strong but net-net i thought this is a really really helpful discussion but do pay a lot of attention to finality it does matter yeah

right on so i first of all this is such a great richer we really should talk first of all I just I love you and always have so we the it what I used to say this when I was over at IBM that yeah I think very highly of Joe Lubin and Jerry Cuomo and Richard really as as three great coaches in a threeway soccer banter or multi of sight a soccer match where the coaches understand good sportsmanship and the team is more or less like each other right so I we can we can we can argue and even even yeah I hope my article was it was so much fun because I wanted to show you respect Richard but I also wanted to poke some fun and she’s like today you could do because I was if I took my gloves off in my face and you always came back it was gentle it was it was funny it was great well I was channeling you because you taught me gentleness in a lot of ways because I can be more gruff sometimes and I want to be more like they’re okay you really modeled that but so look the problem I see is I’m not I’m not here to advocate for corridor or Fabric or even Ethereum I’m saying that we are about to get we’re gonna approach a hard freeze a blockchain Ice Age winter if we keep on running around in our private or public camps and by weight the

one thing I will quibble with you or Richard on is that a private blockchain or distributed ledger technology implementation that is that is accessible publicly is not but you might as well say that a SAS system is a public network so public means something different than that I thought is it useful sure SAS is useful is it the databases are useful so this is useful stuff and if calling a blockchain is how you get clients to to buy this stuff great that’s fine but I think that the stateful Internet which is what I’m advocating for and not that I’m trying to make that those two words into some kind of meme in fact I chose them because they are on and descriptive and probably don’t turn into a hashtag right but a stateful Internet is one where we need to be able to seamlessly find parties connect them privately and then be able to pass information into and out of the functions that live on them without running into double spend race conditions non determinism and other nasty problems that happen with distributed systems and that’s why we need both kurta and things like what’s going coming out of Cornell’s gun sears lab things like credi b and drivet it’s why we need public networks like Ethereum and we need those to evolve all of this stuff needs to evolve and if we don’t evolve together from here we’re gonna find ourselves in an ice age because the things you can do purely on public main net black chains are very limited and the things that you can do sensibly on purely private blockchains don’t really stand the up to the argument that they are you there delivering unique value you couldn’t have done it some other way so but together

private and public together is powerful and is the Internet of the future and I think I’ll predict that when CEOs start just like back in the day remember Richard that you and I there was a guy at IBM that was managed out of the business for saying blockchain too soon and then for most later we could all say blockchain yeah because because the world had changed right it can happen pretty quick I think that right now saying public main net or public network watching scares a lot of CEOs but I predict that as soon as we show how you can have very very private things happening on these side chains and It was an unfortunate term side-chain I would call Corvo I would even call Bitcoin a side chain but I say that in respect it that side chains are where the action it is when those things can operate in conjunction with a public main net and where the where we have that join between a fast finality system on the side and a eventual consistency of machine gun system coordinating them I think then CEOs are going to wake up and say I not only am okay with public blockchain I demand it in the system just not all of it it’s like chili powder it should be it’s not the stake it’s the it’s not the meals it’s the chili powder it’s a small piece of of how we make all this stuff work if the poison is in the dough

gentlemen I’m afraid we’re out of time but I’ve absolutely loved this podcast I mean we touch on some incredible points which as you mentioned John you guys are gonna have to take off line and and come talk and debate about it but some of the points that came up we’re we’re not nearly near not production-grade there’s conversations about building this one protocol there is still some issues regarding finality nan some differences in your opinions but I very much the idea the like of the public and the private becoming the foundation on this future of the Internet so I’d like to thank you both very much for this wonderful debate and this wraps up this in intro blocks podcast we’ve hope you’ve enjoyed this episode and I hope to have you both back in the show maybe even at a conference perhaps we can organize it so I thank you very much both of you for attending this podcast like so much great speaking with you John pleasure as always and thanks for elite it was great you