00:00 to 30:55
hi kids this is Charles Hoskinson broadcasting live from warm sunny Georgia the country not the state also known as the kingdom of Georgia and that place that everybody invaded and somehow they managed to keep it all together throughout the years a wonderful lovely place when I come to these countries and I have an opportunity to explore these places I always get a nice briefing in fact this one came from Dan Friedman and he's a big star Wars fan so it says Georgia a long time ago in a former Soviet republic far far away and as 32 pages long and pretty magical stuff but you know what literally was amazing was that what was a deeply technical document covering everything from the GDP to every index to how the banking system works I had a chance to go with many of the former ministers of Georgia on a pretty grand tour throughout the entire country and it gave me a great opportunity to get to know the culture the customs the religious practices and basically their their myths and their beliefs for example we toured one church that was 1,400 years old and it was a really amazing seat up on the mountain top and it has tunnels underneath it for people to escape during sieges and they've had to use them multiple times throughout the last millennium and a half we also had a great time going to a Cathedral where apparently they had part of the crucifix according to legend that Jesus himself was crucified on so overall this was just an amazing cultural trip but the real meat and potatoes was the work that we started doing with the Georgian government directly so many people aren't aware but Georgia in the last 10 years has gone through one of the most incredible Reformation of any government and probably in the entire world has gone from around a hundredths to one hundred and tenth ease of doing business to about six they used to have one of the most corrupt police forces they completely decimated it shut down the whole force and restarted it using us practices and training and now they have one of the least corrupt police forces they have high GDP growth salaries go up by year and actually the country of Georgia is passing a constitutional amendment that will force them to spend six percent of GDP currently 25% of the Georgian budget on educational expenditures in addition we met the Minister of Finance and he told us how the tax system in Georgia works took three minutes because they only have four taxes business tax a property tax a consumption tax and the income tax which is a flat tax I think at 10 or 20 percent I forget the exact amount so really just a well-put-together country they're very forward-looking and one of the few countries in the world that is already using blockchain actively and several of their systems in particular they're exploring blotching now for everything from health records to land titling and registration and this is principally because bitfury and other partners in this jurisdiction have done really an amazing job evangelizing the power of blockchain and the kinds of things that can be done and should be done to make governments more modern and records more portable and also to promote concepts like self sovereign identity and so forth so what are we doing here well George is one of the few places where we can immediately hit the ground running and actually do an academic credentialing project so basically the idea is that professional credentials whether that be a credential saying you're a lawyer or a doctor or a high school diploma or a university diploma your transcripts we would like to take those hash them sign them put them into a specialized format that standardized throughout the entire jurisdiction and put that onto a blockchain so that employers can easily verify when an employee comes up to them with that credential that that credential is legitimate and not counterfeit not adulterated in any form or fashion so we signed several MO use one with the Free University one with BTU went directly with the Ministry of Education and then we had a meeting this morning with the prime minister of the country who's a wonderful guy very very lively and we had a very deep and detailed 45-minute conversation with him and several other ministers you knew I was a mathematician so he brought his own mathematician who is a topologist who used to teach at University Sylvania was also friends with one of my advisors at cu-boulder so that was that was a great experience and we really enjoyed it and then we got a great photo outside and and we got to see the beautiful Georgian mountains I also had a chance to see this pilot we're gonna hope to try to combine it as something between atala and Cardno this type of a system requires innovation not just in hashing of academic records but also Identity Management as well as the payment system and so it would be really cool to see if we can explore how to build a payment system to pay for these transactions when people are doing verification events on a consortium ledger in this case it's hala and it would be great to see if we create that token to carry the instrument of value using the card on network so this is one of the first opportunities we've had right out of the gate to be able to build a hybrid ledger where there's a so it's still early days our professional services group will be coming to Georgia in July they'll be sitting down with ministers and university partners and other consortium members and collecting business and technical requirements and basically that will result in us being able to put a feasibility report together and then eventually a pilot and should that pilot be successful we should be able to actually figure out the scale out cost for the entire country as well as the value for transaction and the estimated transaction volume so this is just picture-perfect trip it's been amazing to have just so much packed into just a few days unfortunately I've had a gout attack and this is the country of wine it's actually where wine came from the very first vineyards were here in Georgia and I couldn't drink any of it so that's basically a modern-day story of Tantalus there okay so that's what Georgia is all about that's what we've been doing here next up is Israel I'll be flying out very early in the morning I believe 5:00 a.m. so I hope I get the bed at some point and then from Georgia I go to Israel to Tel Aviv and we'll be doing a hackathon and a lot of card auto related events I'll be talking with a lot of partners people I work with like Endor the cryptographic community developers and people the financial industry Israel is a huge cryptocurrency hub there's a lot of really amazing things going on there and it's a country I should visit more often given a significance and prominence to our industry so I promise you that we're gonna have a lot of fun and that's going to be interesting and I'll take a brief private tour to Jerusalem and the Dead Sea and we'll have a chance to kind of experience history as we did experience history here in Georgia so Israel's gonna be quite nice so anyway let's talk about Shelley and the progress of Cardona so hard on oh is red at a major point in its evolution we are converging to a viable test net for the Haskell code which is really remarkable because it was an engagement rewrite effort using very rigorous standards and on the rust side we are now ready for a self no test net in fact some people didn't even want to wait for our documentation and the website and the docker files and the other things we prepared and we're also preparing packages for snap craft and chocolaty and homebrew they said well I'm just gonna stall ourselves and they actually started doing that last week and we've already gotten quite a bit of feedback from it so we're doing a rolling release the people who sent in all the forms the 150 people who filled out all that information will get kind of a sneak peek of the stuff that we're planning on releasing on Friday and it would be a nice soft launch so that if there's anything we missed they can help us make sure we can put that back in and we're going to do a big dump so there's going to be a blog post and probably some video content there's pretty extensive documentation about the node and other such artifacts and so David's going to do an AMA on Friday and basically the first part will be covering what is being released and how to use what is being released and then we'll be available for your questions so I believe that Tim will probably post a link in reddit or something like that either today or tomorrow for questions to accumulate for the AMM Friday and will also announce Oh time for that AMA usually we do these around 3:00 p.m. Eastern Standard Time because David is based in Silicon Valley okay so we're rapidly updating things and what we're doing is every every week to two weeks we're cutting another release of the rust class of the rust client in fact I think we've done four releases in the last three weeks and the Haskell wallet back-end is rapidly evolving as well so the wallet back end and the rest client are colliding with each other and getting fully integrated together and that integration should be towards the end of this month so the initial self note test net will just be for the rest code and then the next major update will be probably rolling out the rest note alongside the Haskell wallet back end or something like that alongside a lot of fixes that come from the QA process and comments from the community now basically what this is doing is giving our community the opportunity to have a blockchain in the box basically this is the first time that anybody the general public would have the ability to run workhorse Genesis and you're just running it as if you're in god mode you basically decide the distribution of the token you decide how many nodes you want you decide your delegation mechanics so you know who you want to delegate to and then basically or force is going to run so this is this is a really exciting four step because it's kind of a feature test to show end to end what Shelly's going to look like and it's a really convenient parallel testing method to the next iteration of the test that we're going to be doing which is the open test network it reflects what you would normally see with a traditional test now so basically there's going to be a common State amongst all the actors so right now you have God mode and it's a great way to get used to the code the API is and generally speaking what this engineering is going to look like and then at some point in the coming weeks we'll make an announcement of exactly how we're going to launch that unified test that for everybody and that's basically going to be a network stress test and not an incentives test so basically there we're trying to figure out make sure that the system is going to work with lots of users in it then run a bunch of Hazard scenarios and so forth and we've kind of roll over the group of people that haven't playing around mastering the use of this test that the self no test net to actively participate in the system now once that's solidified a little bit the final version of the test net will be an incentive test that and the point here is to take what we've written in the formal specifications and what we think is a good incentive scheme for how much state pool rewards are going to be etc etc and to that roll it out and then solicit one final review of community feedback we're probably also going to have a third party auditor come in and take a look at the incentive scheme that we've developed and basically this will give us a a good sense of whether we're close to the mark or not so there's various waves that review now if not to be outdone the Haskell team is converging to a test that we probably will see siren ish test net for fire an arrow functionality with the system sometimes in July and we should probably see Shelley functionality late July to August so as we enter August it's entirely possible that we'll have a Haskell test net an arrest test net running there are design differences and differences of opinion between the rust cloud and the Haskell code and what's going to happen is we're going to basically converge to a final design what are the advantages of our new development processes is that we actually had competing teams build the same product in parallel giving one a mandate to follow a very rigorous formal methods systematic process and the other one the mandate to follow a more traditional Silicon Valley agile software development process but of course we're a functional programming group and we have formal methods people so we couldn't resist but you throw in quick check and all these other property based testing techniques for the rest side and you know it's this stuff happens so so both of them are quite well designed but there are differences between these two clients and so some decisions are going to have to be made about whether to go left or right and that will be the major discussion internally as we get closer to the maintenance so things are moving along quite well the teams are working in parallel finally very well there are small teams usually about three to seven people they're mostly agile at this point we're getting weekly updates and we're seeing very quick releases and rapid iteration of things and this simply just was not the case with our old engineering standards last year so that's a huge organizational win and it's a testimony to the resilience and rigor of our engineers and desire discipline of our engineers to just keep with it it's really really hard when you keep getting beaten up and some cases you're encumbered with very bad code with high technical debt to keep motivation to just show up every day and take the punishment another event this month probably towards the end of this month will be shipping 1.5.2 to card auto maintenance this is going to be the last patch we do there's a few minutes last-minute configurations we wanted to do four or worse PFT this and at that point we can turn on or bourse BFT and that gives us a out great path to either the rust node or to the haskell node and there's a very clean upgrade path in both cases so that'll be a fairly smooth transition so many things are happening there's an overwhelming desire for commercial adoption of our products and there's a lot of passion in our community so it's exciting to see that it's fun to see that and I really do enjoy all the feedback we get one it's read again when it's reasonable and you know it's nice to actually have products and market test nets and market not to be outdone we are making great progress with Plutus and Plutus Plutus is coming along well these tests these hackathons that we're doing are really great acid tests for the design of the system the ease of use of the system the quality of the underlying software and basically what's happening is that Politis is being pulled out as a service so basically Cletus is being decoupled from the main client in as a consequence it's pluggable in two different code bases so once that architecture is done the Plutus infrastructure we've constructed can either be plugged into the Haskell code or the rest code or conceivably other cryptocurrencies as well so this portable VM concept is quite useful for people wanting to host their own pluto server internally if they want to do off chain computation with it or if they want to put it into their own cryptocurrency products both permission to permissionless for example we might use this with atala so that architecture is being finalized we have a series of papers that are coming out they've been accepted at various conferences and we're real excited to share that and these hackathons really help us evolve so the second half of this year is going to be spent very aggressively getting Plutus to a ship ready state so that we can bring it into Toronto for smart contracts and that same can be said for Marlow I think the ultimate success of this paradigm is going to stem on three factors one factor is our ability to train new developers and bring them into the ecosystem we've started this product the process with the udemy courses and we are actively scaling that effort up and we're also doing some on-site training for people the second factor is obviously the model itself so this off chain on chain iteration and this concept of easing dsls to solve blockchain specific to domain specific problems within the the space so for example Marlowe for financial contracts where it is for non-financial contracts so these concepts are I think new and they ought to propagate and third I think it depends on our ability to interface with the broader functional programming community and convinced them that they ought to look to Cardno as their first choice for deploying smart contracts this is a huge community its untapped it has hundreds of thousands of people in it if you look at everybody from the Lisp people to the closure people to the Haskell people for the oh camel people and they are among the best programmers in the world and they are among the most credentialed and qualified programmers and usually the ones who tend to work more senior roles at companies and does have deciding power we've had many approached us and be very excited about flutist and Marlowe and very excited about the entire paradigm that we've generated for it and so we think that by interfacing with that community and evangelizing that you need to be very heavily with the help of the foundation and our programmers within our organization we should be able to see a lot of really cool interesting gaps produced for Cardinal rather quickly and what's really exciting is that we are going to be able to extend the Haskell side of this so that's the off chain code to webassembly and to JavaScript so so basically you can use this in the web browser you can use this as a node package so basically people are going to begin to use more standard tools so if they want to do things in a more imperative way they have that option without sacrificing necessarily the overall architecture design of the system so there's a lot to do more dsls to write up ludus itself is a product of three years of R&D we've done probably 20 redesigns at the system from start to finish because programming language theorists are perfectionists and they really want to get this done right also remember they live with the consequences of their mistake we we witnessed that with JavaScript no man in the alive would want to do over more than him Brenda Nick and we also witnessed this pretty much in every popular programming language there's the Olaf I knew what I knew now what would I have done differently conversation and Philip Wadler helped create the Haskell programming language back in the 1980s and he has the advantage of over 30 years of hindsight to inform his language design so in the development of flutist he was very systematic and trying to avoid the mistakes of the past which has been a great refreshing exercise and a real inspiration for the younger members of the flutist team who are very passionate very skilled and putting in a lot of hard hours to get us over a certain threshold so a very complicated product overall in terms of its design but the use of it and the features of it should be accessible to mainstream developers so Shelly's definitely on the road we know what to do and every week you're gonna see some progress and David's gonna get out there and start doing some amas and those weekly reports are starting to get richer and richer and guys just follow that along the new roadmap websites recently been released and we're gonna keep updating and iterating that there should be at gokhan refresh at some point on the website to reflect more about where we're at in addition to that we'll just keep drilling down and adding more specific information on each of the features and bullet points within that website additional content like blog posts and other such things so product is moving on Rails and your Roy's looking good at virgos doing great work I had it I was just in Japan before I was here in Georgia and you really is basically releasing all the time and every four weeks or something they're cutting something out Caesars looking good as well they're getting great partnerships I believe crypto comm just announced a to support for merchants my secretary actually runs crypto supreme and he's looking into crypto comm for accepting ADA payments on his shop it's been something we've been wanting to do for quite some time and we're really excited at it for the prospect that this can be done and hopefully share it with as many merchants as possible the Foundation's also negotiating relationships with other commercial vendors and exploring things like more Ferber Wallet support to debit cards and the kinds of things they should have been doing a long time ago but at least they're doing now so that's exciting so overall we're starting to see everybody getting it to flow and things works as well now I will return to Colorado at some point hopefully soon and I'm gonna sit down and work with the entire product team about basically the last major feature and technical requirement of spring for Cardinal so certain components of our system we have very clear well scoped ideas about exactly what we want to do scalability we really really really know what to do there we have a great team it's getting a larger erection it's probably going to double in size who are finishing up or bored Heidi a beautiful brilliant paper we have a great team working on side chains for interoperability and we have a very good comprehension of the interoperability design space using inter ledger and lightning in fact Rob Cohen is helping us lighting support and we talk all the time to the inner election of people is very trivial for us to support these types of protocols in the system and to see where they can go what they can do and what pitfalls they've encumbered so given that there's great industry intelligence and there's and there's also also starting to put together a product team to begin Volterra's implementation which has to start soon if we're gonna hit 2020 for that and now that all the code is rewritten we have a beautiful low technical debt codebase that's very easy to integrate with and we have great processes so we should see a lot of progress there in that respect so those are the things we know how to do very well the things we also know how to do but we haven't captured and put into a great design system yet our hat talking or ytf recommendations about embedding of metadata and compliance information into cryptocurrency transactions so the challenge with cryptocurrencies if you look at the transaction itself in a most abstract sense as an atomic element of a financial system cryptocurrency and non cryptocurrency the heart of the transaction has four branches coming off of it first you have the asset that the transaction is carrying alongside the information of the transaction you have second the entities who are involved with the transaction third you have the story the why are you doing the transaction and fourth you have some sort of answer tation of what is the regulatory environment upon which this transaction exists and how will you determine whether the transaction is in compliance or not and in the case of the cryptocurrency space there's an added fifth value which is the code so if it's a smart contract or if there's code embedded within transaction that's additional terms and conditions which are machine understandable and they live within the transaction itself the problem with Bitcoin is it's a push system science information or sophisticated transactions so with aetherium we were able to start exploring the design space of the code side and then we had some arbitrariness with that design that would conceivably allow you to invent metadata and other such things into the transaction if desired but there really was no standardized way of doing that and also it creates the higher cost per for those particular types of transactions furthermore there are transaction patter scription you subscribe to something with your credit card and then every month some charge appears so there's a lot of innovation we can do with the subscription model about basically authorizing a subscription within a window and then allowing people to pull from an account until some time period or until a chimeric ledger standard so this is something we're going to study quite a bit metadata embedding and identity abetting are quite controversial as topics and rightfully so because they breached the fungibility and they breach the privacy of transactions so a fair compromise has been historically saying perhaps we just associate the transaction with a hash and then hash as a pre-image that map's to something useful for example a contract or compliance information or other artifacts but because it's a hash you really from just looking at the transaction would not be able to ascertain what the nature of that information is so you retain your privacy so in accordance with these if atf ideas i sent to their transactions people will started banning some form of identity data into their transactions and people are going to care more about the compliance ideas of where is this transaction going to travel with jurisdictions that's going to live in and something will be done to assist that transaction in becoming more compliant so that said we have a lot of work to do to build an end-to-end model for these features and functionalities so part of the team that's delivering shelley will go handle different transactional models like this mehta notion of accounts you know what does this cluster of identities look like cluster of transactions look like relational to some known identity in the system as well as layering on other architectures like a pub/sub architecture into the system so we can then use that for contingent settlement you've probably seen continued settlement before in the context of multi-sig but there's a huge elegant spectrum of use cases for that and it's again something that the cryptocurrency space has not historically done as well as it needs to so we'll be able to bring these features into card auto rather quickly they don't actually require a lot of code to be written they're more of a question of a philosophy statement of well what do you want to enable and how do you make sure what you've enabled doesn't cause too many issues through the system but at the end of the day our transactions will end up being among the most useful flexible and potentially complex in the entire space so you simple if you want simple and it's complicated if you need it to be complicated and that's that's really the magic magic of what we can do with what we have so what I return to Karl Colorado we'll begin that process and a dedicated team will be smaller from shelly team once they've finished to work directly with those requirements and finish out building these compliance layers of the system and then frankly we have man ten system it'll support everything we want people want to keep regulators happy that said we will also begin exploring some privacy ideas and how we can also increase privacy a bit on these transactions because once you have the ability to embed metadata then it makes sense to increase the privacy of the underlying transaction and then leave it up users decision of whether they want to bring those transacting those transactions into compliance or not so there are a lot of ideas on how to do that we've even looked at it at the consensus layer with a paper we wrote called org or script Cena's so thank you guys so much for listening and I have a little bit of time so I'll just take one or two questions and I do apologize for the connection I'm in a hotel in Georgia a country and guys one of the consequences of going to the developing world is internet is not always amazing so deal with it and thank you for the haircut I said I wanted an eight I said two point five centimeters and he gave me a military buzz cut it was a little short and it's showing that I'm really starting to get bald Charles will I which can't ever work on etherion classic again we're gonna make some form of proposal for a theorem classic about what we can't do but we'll submit it to the different stakeholders in the community it'll come with a funding request and if it's approved we'll work on it if it's not approved well then will formally leave the ecosystem we made a big financial commitment well over a million dollars we spent a year and a half and we built a great client which is being used by multiple parties and it's being used internally for a lot of different projects at RHK so it was time well spent money well spent engineering effort wills event and it was a commitment made and honored so if we're gonna go beyond that there has to be an economic model that is sustainable and it's the etherion classics community's decision whether they want to go down that road or not and if they don't okay that's fine we have bigger better things to do right see one more question then I'm off to the next big meeting come on guys make it good and I'm not gonna talk about Facebook coin or any of the magic there and Saturn is of course cool all right well nothing interesting well that's the case until next time thank you for listening it's been a heck of a lot of fun we have really really enjoyed the time we spend here and I'll see everybody in Israel and hopefully we'll be able to talk real soon Cheers see you on Friday