Rethink Trust 2018: Jonathan Levi, What are the Actual Problems we solve with Blockchain w/o tokens?
my name is Jonathan levy thank you for the nice introduction Alexei and I'm going to talk about I don't know how technical people are so I'll try to kind of have a brief overview of everything we can dive in but the main topic of the day is gonna be security and interoperability and how we scale blockchains I'm gonna present it from the angle of somebody who is working with a lot of enterprises so I'm just saying kind of we basically came into the blockchain from so before used to be an A flat goo photographer so I started photography I was originally army we did a lot of stuff that is to do with kind of channels and basically multi-party computations in the sense of like covert channels basically you want to have some communications where you don't know people to know that you're even community communicating anything so let me let me just get there and then I'll get into like what we're doing today so this is how I started I started with key photography I did a lot of stuff in finance I spent almost 10 years in finance I was in Standard & Poor's these credit ratings were trying to protect kind of transitioned credit information that is on the fly I was in Barclays Capital during the credit crisis with a lot of back-end systems actually on the credit derivatives trading nest then comma sucks we wrote the governor whisk system so it was interesting how I transitioned from kind of being and apply cryptographer our technology was acquired by SMP then I moved to banking but I still kept the my deal of a coup photographer in terms of date a curious like basically having like highly sensitive data is being transmitted and you need to have a lot of computations it have to be reliable then finish the masters went to Stanford wanted to do another PhD I dropped out of the first one then I got out of the second one but basically after two years or three years at Stanford I got back to crypto and then I started walking on I don't know it was a cryptocurrency class in Asia the first cryptocurrency class in Stanford with Dan Bonet and then it's not one I just left and what I try to do like my approach to blockchain was always I want to look at kind of sharing information sharing data building systems but I can be like the crazy crew photographer of like you know that the decipher bank basically not trying to kind of decentralized everything not to have a single point of trust but at the same time I always look at things steal from a kyc regulatory point of view AML and all the rules that you really have to deal with when you're working with real data in regulated environments so that's just quickly now the Platinum that we work to build is basically it's a Sarah the name is kind of it came from hash coming to reveal the idea was basically to make a commitment on a blockchain and then later on reveal it but if you look at exactly what we are doing we ended up creating a layer that allows you access control that is blockchain agnostic meaning if I want to be able to see who can see what and specify a policy of who can do what on a blockchain in a much more controlled way we didn't want to do it baked into a chain we didn't want to fork an existing chain so we just ended up building the whole thing on top and it is blocked in agnostic the talk after me is going to talk about like deploying stuff in a multi kind of proximity cloud way I'm just gonna talk about like why we built it and I'm gonna focus on the actual problems that we're trying to solve so I don't know you probably think like that right it's like I don't know ten years now that we are talking about broke chains and it's kind of nice like we have a solution everybody's looking for a problem sometimes people don't even care about the problem which is cool it's nice for research but let's talk about what do you need to do when you want to go to like Enterprise Solutions okay so we started with kind of caring and sharing or sharing is caring right let's all kind of you know sing the kumbaya hold hands we're all sharing everything now it's nice in and kind of in this world where all I wanted to do back in the day with Bitcoin is we were trying to kind of prevent the double spending issue right so I had a coin let's say I have a token number 17 and people are worried that if I'm gonna be out of sync basically I can pay to two different people right so we said let's have a table and every time I look at coin seventeen I know the less owners this is how the whole thing started and then when I pay to like party B if I try to pay again coin number 17 is not it's not mine anymore right this is how the whole world started but now if you look at the blockchain today we basically started with like sharing everything and we assumed it we can take the same technology that works very well for Bitcoin to prevent double spending and censorship resistance and we believe that we can just come and harness it and then build my own permission chain okay so I don't know if you basically we saw a lot of kind of institutions that it was basically mostly 2016 let's realize that they've been the prototype or a POC an oil pilot and then they ended up being kind of alone right it's like their own little blockchain it is not connected to everything so let's look at like why is it the problem okay the problem is that many block chains actually been in regulated markets are being developed and deployed in silos and the reason is mostly regulations so it was supposed to be animation but it didn't work very well nology it does so we know that there will be different types of block chains we know that we deployed block chains for different needs somebody's gonna be for financial asset transfer sometimes it's gonna be for healthcare you know the case is going to be sewing I don't know we see like issues with reward systems a lot of 2-sided markets when you have like buyers and sellers and we are thinking of let's disintermediate the exchange right so if i just want to buy from somebody let's have a peer-to-peer network when I can just like transact with somebody directly without needing to go to I don't know it's eme or any other change but what happens the collars basically represent a different blockchain type right so I can have like any theorem a Bitcoin stock I have religious fabric Soto's or whatever change you are looking at the problem is that you also have the separation of both different frameworks and different deployments okay so that's a big issue so them actually did the biggest problem that we have is that it's very difficult for me not just to move participants but to move data around right to actually and let them event propagate to different block chains so I'm really siloed okay and then I need to maintain it and then I'm exposed to different risks right so it's very difficult for people to manage these deployments so why is it happening okay let's let's I don't know how much time I have but I have no ability to control time tell me it's good so let's see what why it happens right so people are not really malicious at nature right basically people try to take the permissionless blockchain which was the Bitcoin ami theorem and all the kind of change that are permissionless they are built to basically tell you to prevent kind of censorship resistance right so I would like to be able to prevent censorship nobody can come in and block people from accessing the network so you are blocking a mining pool in China Sony will do it in Thailand and you see something in Korea and people are gonna move around and as long as they have a synchronized node that basically downloads the latest ledger you can go and validate the last block I can always look at the signature of the last block every transaction and make sure that they are all valid that was kind of the promise at the beginning and it works in it's actually great technology for this now the regular network started with okay we can't put everything in the open this is all right the whole thing started right banks financial institutions regulate like regulated entities that actually care about your privacy cannot put PII like personally identifiable information on a blockchain in declare so putting the identity on a blockchain let's forget about it putting a hash of the identity on a blockchain possible but it still leaks some information right so people started to think about okay we need to permission it we need to kind of at least at the very basic I would like to know and I'm joining a network who are the validators who are the miners okay so you know like we work with permissionless and permission block chains so this kind of this argument is kind of funny for me sometimes because you can't compare them like these are solutions there are two different families of solutions that are trying to tackle different problems here we're talking about access control in terms of luck I don't want anyone to be able to kind of prevent access and in regulated markets I'm looking at actually I can share information only with people that I'm allowed to share information with so if I'm gonna give somebody access to my data and it has some sensitive trade information or identity about the client the audit or sometimes I'm not even allowed to share it with you even if I want to so in the permissions space you have IBM was kind of one of the early starters they started a consortium they were part of a consortium called hyper ledger that's 2015 I actually joined I don't know two weeks after I saw the first check in to the public world Oracle is adopting like a blocking the service of course Microsoft so we see a lot of cloud providers a lot of vendors that are joining kind of the permissions space some of them give you like a platform agnostic kind of blockchain as a service but if I'm looking at the regulated space and the problems with it right we're trying to solve in the enterprise is mostly about kind of IOT financial asset transfer like trade finance in different ways right lending borrowing paying trading any kind of registered asset and of course if I'm looking at kind of the big kind of promise that was kind of how we started to kind of sing the kumbaya it was a supply chain right everybody kind of realized thank you that we can like have a state machine and the transitions like the state changes are gonna be on the blockchain okay so the problem that we have with all this promise that nobody's talking like and that's a real problem like for different reasons right that different like offerings people offering you different cloud services we will offer you consulting services so we will just have a specific hardware some people are trying to basically you think about it right whenever you walk in an ecosystem everybody has their own agenda but effectively today still nobody is talking to each other like these blotches are not talking to each other another thing that I think is important to note all these stuff is gonna stay like I don't think that bitcoin is not gonna disappear like even if it goes below five K okay I don't need that IBM is gonna go away any time Oracle is really pushing a lot into with a lot of kind of product offerings you know Intel is that players that you know not going anywhere that's us forget it but let's talk about enterprise application requirements very quickly so the biggest things that we saw over the last three years the people really needed right they wanted petition executions they want to be able to come in and have different validators running different things so we thought about how aetherium works today and in the theorem every validator needs to validate every transaction which is a problem especially in enterprise like if I have a sub sector or let's say I have the London trading there's validators I don't need the New York trading this validators to validate a trade maybe they need to be informed but they don't need to execute the exact same contract and to validate it so this gives you a lot of performance and this kind of partitioning really helps you in terms of regulations I'm going to go very quickly because I don't have a lot of time transaction history that's clear people want like an audit trail if a look at permission the membership this is exactly the subject I talked about in terms of kyc and all the other regulators all the regulations and requirements another thing that is really key that people kind of forget is that post trade or post deployment and once it's live we really need to move data around so we're not like I think people was so focused on like going to production and we work with a lot of kind of big players on getting you know all the check boxes from the compliance departments and the comparators and the legal departments and it is live and it stuff is traded and our platform is in production but I think that the next step of what we do is going to be like to interoperate and move stuff around between these sorry data silos so this is why we built what we've built and we kind of extended it want to focus more on the problem steal so it's not exactly that access control allows people to censor you like the way we do access control all the rules that we have are basically shared so you know what the rules are and people actually executing the same policy so whenever I want to block somebody everybody needs to agree on oh I can have a policy let's say I need 51% oh I would like to have a majority but somebody can veto it so the idea is basically not to think about oh these guys are permissioned and now you know somebody can block it yes of course the network owner or the whole consortium can't do it but I don't want you to think about it Oh access control it means like the eos kind of deployment right it whenever they think that somebody was breached they just have a phone call a video conference and they block some you know addresses so this is not the case right they're still open we're still sharing this data and we're still able to move stuff around even with perth permissionless let's see how much time I have this is an example I don't know how technical people are so if I see we was squinting too much I'll slow down but this is like a nice diploma that combines a lot of things so we actually it used to be confidential up until like ten minutes ago I just didn't prepare a lot of slides so I just took something from a spec but effectively if you look at what we have here you look at the participant that requires an identity they promise you that if you put the identity on the blockchain you know it's to be honest like now we know that it's not even legal so it's not an option okay so we cannot put an identity on the blockchain at the same time you don't want to have a central kind of authority that will come in and hold all the identities so we use in this example use the D ID so a decentralized identity so I gonna go and take my identifier somewhere else that I own and then I come in with the D ID to the use case then I'm gonna come in to the blockchain this is the second step and I want to create a request at least like an H CR the ID kind of resolver then I'm gonna come in and create my local identity for that blockchain and then I'm going to transact with it but at least here it's different because I own my identity I'm not gonna come in and it's not that the bank gave me you know a user account number or Facebook allocated to me something this is my decentralized identifier that I can move around different systems and this works kind of cross block chains so that was very quick I can talk and I don't have time for questions sorry because I ran out of time but if you would like I'm still around we can talk about more use cases and specific stuff but we do a lot of interesting stuff even in in enterprise and there is a lot of room for innovation so yeah thank you [Applause]