Surprise AMA 03/31/2019

hello everyone this is Charles Hoskinson bribes casting live from warm sunny Colorado okay and there might be an issue with the stream okay there we go and people should trickle in now apologies guys with the stream I was trying out a new type of software and I couldn't get the software running in time I gave myself just about ten minutes to play around with xsplit and I could had a trouble with the key anyway I hope everybody can hear me hope everybody can see me and thank you all for waiting I have have family currently over and so I had entertained them for today but I didn't want to do an AMA when I got back to Colorado so thank you for waiting alright let's take a look here I hope you all have saved up some questions comments concerns other such things and let's see first question okay so while we wait for the questions to come trickling in and those who are on the live stream please do make sure to go ahead and submit the live stream link to the telegram channels and elsewhere so we can get as many people as possible into the live chat make it more of a dynamic nice ama things are going pretty good we're in the nuts and bolts and wheats of getting Cardenal and the final mile the Shelly work is coming along we we have a huge amount of work on the specifications a huge amount of work on the Haskell society a huge amount of work we're doing on the rough side of things we are now in a weekly date cycle for the wallet back end Matias Bank Court is doing very good work there Marcus has done some great work getting the rust node running on a rock by device I share whenever he sends me cool videos of various devices working I share that on Twitter basically you have these small devices are about this big they cost about $70 and you can connect to all kinds of things to them like external storage and backup battery we've even experiment it was getting into work with solar so when you go to the I which case summit you'll learn how to do this yourself but basically our hope is that these little rock pipe warts can end up becoming kind of the de-facto reference open source open hardware a stick pool for people they run on less than 10 watts of power they're pretty easy to configure and set up they're very reliable so I which Kate does have some ADA so we set up our stick pools we'll probably use for our company state pools these little devices as reference boards so look for that we're pretty excited about it test network working SuperDuper hard we really wanted to get it out today but we're still still doing some last-minute configurations but the shellye test net should come out fairly soon not like six months from now but like more more in the week to several week time horizon and basically that will be a great opportunity for us to roll over all the people we've accumulated the state pool task force channel into the shell a test net and begin the hard stress testing and hopefully be able to get people running state pools as quickly as possible we hope on open hardware but basically you can run them at any configuration whether it be pushing something to an Amazon server or running it on your own computer at home we were going to see how easy the setup happens to be so so good progress there and that last mile and software is always so difficult because a lot of things are always coming together and there's tons of little things that people never thought about that they always become last-minute issues punching software through that last percent is always a challenge but we're almost there we ran in the same issue with Leger we had most of it done in February and there were these tiny little things here and there that came up and we just kept pushing and pushing and pushing to get it through but we now have let your support thousands and thousands of happy people we got a lot of great messages most people can use them of course for software there's always some configuration somewhere that for some damn reason we can't get it through this whip USB standard or whatever it is some people can't use it we don't know why but for 99.9% of people most people can use their Ledger's or have a good time with it so we're pretty happy with that anyway overall things are looking great on the software side progress is pretty rapid and as we approach the summit I'm pretty proud of how hard the team has worked I and I'm glad that were within a good range for how we ship things and we're getting kind of weekly feedback and weekly releases for the wallet back-end for the rust client and soon-to-be the half Haskell's software as well so we're all moving in the right direction okay let's go ahead and answer your questions let's see what you guys have are all right well eight ASP or Colts taking the answer is yes we've talked to vacuum labs about Colts taking support for the alleged devices we're also in discussions with the vacuum labs guys and our own engineers about it making sure that we have multi six support so not only can you cold stake and so what that basically means is that you create a key that you then import so that's a sticking key into Daedalus or your ROI and then you can use that key to stake with that or delegate with that but then the spending keys remain on the cold wallet device so you never lose control over the spending of the of the key but you've now made a hotkey specifically for sneaking rights so that capability will exist and we're in discussions about how to add that support the ledger it probably will come out not soon but in the ledger nano X timeframe so that's the late May June timeframe just because there's a lot to do there but multi stick is also part of that conversation as well so the next major update will be saying how do we get a good multi-sig cold wallet experience so basically you have kind of a copay like multi sick experience where you can build and read that and kind of construct that transaction a copay pub/sub system where someone initiate the transaction amongst an Emma and keychain but instead of having the actual signature take place on the cellphone device the signature will take place on a ledger device and then the signature will be transmitted to the cellphone and then sent to a pubsub server which will then relay it to the other people on the keychain so we'd like to figure that out and we have some ideas of how we want to construct that quickly a lot of the infrastructure we need for that is already built it's just a matter of putting it together and then kind of the next step is saying let's also make sure we can do this with delegation as well it's very important for us it's important for mergo is important for the CF and it's important for you guys the community okay can't talk about coinbase listing that's something that's off limits crazy here on the side of your face yes I am getting older and I need a haircut I just got back from a 23 day world trip and boy it was fun I Charles with Steph's partnership has n cash / horizon PS they really shouldn't have changed her name actually I like to send cash a lot but horizon okay but everybody's always gonna misspell it I wasn't not involved in the name change I'm just there to answer questions we do consulting work and we're right now negotiating the next phase of consulting I've always liked Rob and the rest of the team I think that they're good actors small cryptocurrency but they're damn good actors if I have a Raspberry Pi node my internet sales will my stake be slash this is an example of where people take concepts from other proof of stakes and they import them automatically into our proof of stake there is no in the incentives design that we have for a Boris at the moment concept of slashing where we destroy your your stake you do not have to bond data for our protocol to work this is in our view an advantage of our protocol over last generation protocols bonding was kind of a clever mechanism to introduce some sort of cost to lying but if you use the epic design then you don't actually need to do this because you're delaying payment and some sort of closure several days later so so anyway there is a isn't the concept that somehow if you don't perform optimally your money gets destroyed or you lose a bond or something like that now your payments will go down if you don't do your job well at all you get no payment so obviously people won't want to state to you but we don't have the concept of slashing in our system [Music] okay and by the way we have answered a very big collection of questions from the stick pool task force and we're now in the process of taking that collection of answers and putting them into a more accessible public format so I just talked to Jeremy Roni today he's I which case director of operations and chair is right now exploring with a few different people in our marketing department about the best way of taking that question-answer list and converting that question-answer list into basically accessible content about how delegation and staking and incentives and rewards what all these things are going to look so look for that we have the answers it's just getting them into a website the N which case summit has really sucked a lot of the air out of our marketing capacity most of our people are focused on that and getting through the summit so that's almost done but it's slowed down some of our ability to release other marketing assets for example the stateful task force website [Music] let's see what else we got here well we have a debt platform at the end of 2019 I don't see any major delays that would prevent us from having smart contracts support by the end of 2019 and certainly before the end of 2019 Goggan is a parallel project to Shelley so they have separate teams and they've been working now for about two and a half years on designing the languages Plutus and Marlowe and basically building up infrastructure for them now there are lots of things we can do to improve flutists we've already learned a lot of useful usability lessons for the Plutus playground and for meadow that were a direct consequence of the course we taught in Ethiopia because the girls but the last month basically being Politis developers so we'll keep doing that and we've already incorporated some of those improvements for the hackathon we're going to do in April for the average-case summit so that'll come now there is a broader discussion of what is an app platform and what does an app platform mean for a cryptocurrency now my goal and this has been something that I've been wanting to do since 2014 is this concept of adapter so basically the idea is you have something that's kind of the Chrome Web Store or the Google Play Store where you can search for an experience and that can be a game that can be a gambling application that can be a wallet comes up there's ratings there's an author there's metadata and there's a button to install you click it something happens you install it that it goes into your App Catalog and basically that state can be updated by the author of the system now the hope is that the backend for that is that some of the logic will be running in decentralized infrastructure and then also that the centralized infrastructure will handle tasks like author management registration reputation management and triggers for and you found a tional technology like graph QL and for about roughly a year our companies explored this with Darko nation as the product manager and there's Ries and a few other people and we'll probably do some sort of workshop at the summit explaining what we've learned and what we're considering some people are moving away from electron for example you may have read recently that the mists project has been retired and that was built on top of electron that I believe the brief browser was in some way built on top of electron and now they've moved to a pure chromium fork and then there is a lot of discussion about potentially using Waze and plus huazi as a as an alternative we also explore docker and we explored some other containerization technology the problem with containerization is it's simply too heavy for a lot of these types of experiences so we were still studying that but I think it's that's the consumer moment getting to that easy search easy curation easy installation which seamlessly blends the client the server and the blockchain infrastructure together and has a good author management a good distribution story and in some way does not require a full curation that's what we call the the def browser and no one in the industry has really figured out an optimal way to do that but there are a lot of foundational technology that potentially can let us go get in that direction there are some challenges like for example across application communication so ideally a situation where every piece of cryptocurrency infrastructure or blocked in infrastructure lips within the sandbox so you would no longer have a separate wallet and then that somehow communicates with this DAF browser you'd like them to be one in the same so you download in some part of the future basically an empty socket an empty framework and then you search and you decide what types of apps you want so you would enter something like card on a wallet or Bitcoin wallet and there'd be vendors who have deployed there with various trusted and then you click on it yes I want the aisle educate her on wallet you download that and then once you have it you can store ADA and then you have other apps with dependencies so for example you download the Charles house Constance wild poker app or something like that well that needs a permission to access the cardinal wallet so when you make eight of bets you can drag money out and put it into whatever escrow mechanism that particular software happens to have well the minute that you introduce two applications that live in the same system where the apps can communicate with each other through some sort of bridge and there's permissioning ie the app can take money from one or do something with the other then you have a whole litany of attacks escalation of privileges and other things that can occur and so we from an infrared information theoretic standpoint info security standpoint have been thinking about well can we build some sort of permissioning system that makes sense and it's fairly tunable and also it's not gamma Buhl's so that users don't accidentally hand away the keys of the kingdom and this is the intrinsic challenge behind these types of the systems so the problem is that the more you do to enhance the consumer experience meaning the more seamless you make things the less menus people have the less configuration people have the less prompting you give the user the more likely it will be that people will basically inadvertently say yes to something that could steal their money or download an untrusted application that can cause quite some problems and so this is an openly usability issue that everybody has to contend with to give you an idea of how bad these problems can be just in our own ecosystem when ledger support was announced there were people who are shipping fake software to Ledger's to actually steal people's keys and there every time we do an update for Cardinal hackers wait for the announcement and as soon as they do they should fake binaries for devilís that again will steal people's money so they're sure a lot of great hackers out there who are terrible human being but they invest enormous amounts of money into trying to scam innocent people with to stealing their money so when you create an app platform the smart contract side of it can be trust less and and with the best of intentions but then when you put all these things together and you have the ability for cross application communication you you could emergently create security issues for people that are in some cases difficult attack no amount of formal methods or computer science magic will kind of protect you because at the end of the day you may be as the user authorizing the system to do something against your best interests it's a great conversation to have you build things in layers decentralisation smart contracts are kind of the foundational layers of the system and then getting good developer experience as the next and then having a good platform to deploy these things on it's the next after that we have kind of a collection of teams that work on all these different things and we're trying to work as much as we can parallel and we do have some good ideas about what an app platform would look and currently our best guess is to do this with electron [Music] now let's see what else we got can we expect legend with deathless integration the answer is yes we are planning on doing that it was less of an up priority because at the end of the day regardless if you access your alleged device with a cell phone application a chrome application or a heavy desktop application basically the key management is always done on the legend device itself so at the end of day that's just a pretty front end it's a window into your device at a command center that allows you to work with your device but it does not control the keys themselves so there really isn't a security difference accessing your ledger from Daedalus then there is accessing your ledger from your ROI because the way you ROI was constructed and it was on a much better codebase due to being inherited from Icarus on the rust codebase it was considerably simpler for vacuum labs and mergo to build that support into euro we're in the process of completely changing all the code behind Daedalus and that's almost done there are two teams working a Rusty Moore the Haskell team and we're just paying that price of rewriting once that's done it'll be very straightforward for us to add things like ledger support multis sake it just simply makes no sense to build that support on top of the old code base as frustrating is that it can be the point of deploying Icarus when your ROI was to basically give people a thread that we can add features rapidly to so you could get ledger and multi-sig in these types of things and it didn't slow us down while we're in the process of rewriting everything for Shelley and for Cogan so once that rewrites done let your will happen and actually other initiatives will happen as well for example we're going to start implementing the proof of stake side chains work and move Daedalus into a light client by default and it'll upgrade into a full node in the background so basically what will happen is we have a basically a cryptographic structure which will ship with the wallet which basically means that when you boot up the wallet for the first time the only thing it's going to do to start is grab the existing UT Excel will put something in the protocol to we ship this to the client it'll be able to check that very rapidly and then use that cryptographic structure the validate that you TXO is correct at that point you'll be able to very easily and straightforward use your wallet spend money from it restore a wallet into it with no problem and then it'll gradually download the blocks before so you won't have to have this waiting for the entire block chain to sync or other things like that so that's it that's a big priority of mine for a usability viewpoint I would like for Daedalus to be significantly lighter there's also some discussions about how can we make electron lighter for the for the user because it's a very heavy framework and for some people it just seemed to be a bit much we've made huge wins in 2018 for example we went from memory utilization of about 2 gigabytes to about 500 megabytes and then also we considerably reduced the amount of storage requirements for downloading the Cardinal blotching by a factor of about 5 or 6 but we can do better and improvements the wallet backend improvements the core code proven it's the data based technology that we use and also just improvements to electron itself can probably push us in in the right direction and then having a light client mode by default I think will dramatically help usability for the court notes of the system [Music] can you speed up research with more recruit people in this project Charles this is a question that comes often like how why can't you go faster can you hire more people there's a Michigan the conception of software engineering and research that somehow you get linear scaling so if you have 10 engineers and you go to 20 engineers you're twice as fast the reality is that software kind of has a critical mass for any one part you're working on in the architecture the code base you have and the collection of things that you want to do that if you go beyond that critical mass of engineers for the methodology or following and for the code base you're worried of the project you actually have people tripping up on each other IBM was one of the first companies notice this and I think they have wrote a paper about it 1970s or 80s and I forget the term of it so we're at maximum capacity for engineers occurred on we have a lot of great people that are working very hard and right now the only efficiency gains we can get our gains of when we actually have proper things like better processes so Bruno Palio's our director of engineering and he wakes up every day and he's trying to systematically install better processes throughout our organization that was one area we were weak in 2017 and throughout 2018 we just simply had the wrong philosophy for how to build software it was in a too much of a waterfall esque way for my comfort so the good news is that we've already transformed several teams the RUS team and the haskell wallet back-end team are both operating I'd say a near peak efficiency or as close to as good as we can get I and I the models that they're using and we're seeing week by week a transformation of the legacy Haskell groups where they're starting to gain better acceleration and they're starting to work pretty well the formal methods team also needs a little bit of acceleration due to the nature of the work that they're doing it's a little difficult to accelerate things and also on the research side researchers researchers this is why you start foundational research as far back as you can and it's important remember that we've been doing research since 2015 so on all the core technologies that we intend on shipping at 2019 the research has either concluded or it's reaching the refinement and proof cleaning stage meaning that the core protocols have not changed so those have been available for a while people can play with them looking on to the road map to the future we have a little bit more to do on the single threaded or aside so there's an upcoming paper called Wars Kronos and that paper solves some issues with the clock in a little of the major advancement a proof of stake it's currently under submission I think the crypto and look for that paper soon and then the very next paper will be something involving spikes of dishonest majority at that point outside of quantum resistance checkpointing quantum computer check pointing there's really not much more to do that will add substantial value to the or horse research stream and we'll be in a position where we could basically say this is it it's complete and we'd be 30 to 40 times faster than Bitcoin and in terms of throughput and then order of magnitude better in terms of latency if not and we probably improve that more and we would be a hundred times more decentralize so that's a pretty good win but we didn't want to do sharding and we have ideas on how to do that and that's really taking the parallel chains research stream and taking the side chains research stream and combining them with what we've done with Ouroboros and those three things together become war pours Hydra so once we finish off with Kronos and we finish off with all the refreshes of the orb bores paper family that we have with make considerable improvements to the proofs we've simplified some of them extended other ones resolve some theoretical issues here and there that came up during the peer review process or from our own analysis things are just more readable and there's a lot more clarity that we have then then the research team will immediately begin working the final steps of the sharding park but then there's a lot of open questions about things the incentives side of the protocol the delegation side of the protocol the stateful design of the protocol and there are a lot of interesting research questions that we can pursue there and will pursue there we're pretty happy with what we've done out of Oxford and we're pretty happy with what we've done with the specifications on how delegation incentives work but there's always room for improvements so there's going to be some scientists to play in that area and work in that particular area but all things considered we've gone from aspirational from scalability and interoperability sustainability is the three pillars of thing that we'd like to play with to a massive tsunami wave of papers and designs and the end result the output of these processes are are basically us converging to a really good concept of what the permanent system needs to look like that will leave great stubs for future researchers and future engineers to improve it so all things considered with the resources we had the engineering time the science time we put in we're pretty happy with the outcome of that adding more researchers and scientists would be frankly counterproductive with a few exceptions we probably could use some more people on the pl side just to build up the ecosystem around it for example we are scaling the team that's working on compiling haskell to javascript and compiling haskell to web assembly and that team will be finished in terms of its scale by end of april mid-may we're just onboarding people right now and they'll reach peak capacity in terms of their output sometime during the summer and so then this will make it considerably easier to deploy fleetest contracts and a node setting or in the browser or these types of things so there's some stuff to do there there's also some stuff to do on libraries and SDK is immuno things like that but those things come organically as a back and forth between developers and the maintainer z' of the but core people we certainly have enough and we certainly learned a lot let's see what we got here why is your live cam quality support looks you're being filmed underwater I'm on a 4G connection I live in a town this is this is the great tantalizing irony of my life I think it is a great analogy for it I live in Longmont Colorado just outside of Boulder in Longmont actually has a municipal ISP that has one gigabit up and down fiber optic internet connectivity for $60 a month I live just a few miles outside of the five burgers own and not only did they not service my area Comcast and Centrelink do not for high-speed Internet so my only option is to go with satellite or 4G connection both are about the same speed of about 30 to 40 mb/s down rate with about 10 up and one is an order of magnitude higher latency than the other so I'm on a 4G connection so and when my co-founder Jeremy what he lives just in town and he's on a one gigabit per second connection and so he often reminds me how wonderful his 4k web cam and his video conferencing happens to be and I have to make do with with my connection so I hope when 5g rolls out I get better options but at the moment this is as good as it gets for where I'm at but I have a much better view can you elaborate on the work that karna is doing in Ethiopia so Ethiopia is is basically one of our flagship and beautiful places that we we work in so you have 106 million people seventy percent are under the age or at the age of thirty and it's a country that's going through enormous economic growth and at the same time it really does need to change the way it's doing things eighty-five percent of the country is agricultural most of the food is grown or so we connected to about fifteen million smallholder farmers they have a big problem with diaspora on the perimeters of the country they're landlocked and they consist consistently have border issues with some of their neighbors the countries had historical issues with establishing free and fair markets there are a lot of monopolies who have to be complete and solid broken up and there's a lot of people that do like taking advantage of some of the poorest people in the world and they need to go away or be disintermediated so the Prime Minister and ministries there are quite forward-thinking and there are some that are better than others in particular we really the Ministry of innovation technology we've worked very closely with them we also the cultural transformation agency led by mr. Khalid bomba I used to work for Bill Gates it's actually one of the first ministries I think of that is funded by private interests but yet it's still a government ministry and they've done some great work taking about 3.2 million of the 15 million smallholder farmers and putting them into collective bargaining units called ACC's agricultural commercialization clusters so basically they can unify the farmers on standards from what fertilisers they use to what types of commodities they grow there's about ten commodity areas that the ACCC focuses on and then they can use these standards and they can use these better practices and better data collection to increase output get better prices for the food and ensure that farmers by collectively bargaining they they have a bit more economic voice of power than a normal smallholder farmer would because when you're only on an acre of land and you don't have a lot of output you kind of just have to be at the mercy of whatever the market happens to give you so no talking to the ATA they have basic problems they'd love to solve for example they say hey we got our fertilizer in Jim booty and they come in as 50-pound bags and by the time they get to Addis Ababa still a lighter they're like 45 pounds or 44 pounds so somewhere along the way some fertilizer is being taken or they come back to the same weight but when you open it up it's adulterated so maybe it's 35 pounds of fertilizer and it's 15 pounds of dirt or sand or something like that somebody basically swapped it or the bags just magically grow legs and they they kind of walk away so when you have situations like that the conversations usually still around can we build a blockchain solution for that supply chain so when the bags get checked in at Djiboutian and they work their way to Addis or wherever they end up being shipped to all along the way you can kind of discover who's touched it who's done things with it where houses are storing it and basically where it's getting adulterated or stolen or slippage is occurring so projects like that and we're working real hard to try to create pilots along those lines because those pilots not only deploy blockchain solutions on an enterprise side to the country of Ethiopia but they also work at creating digital identities for people digital wallets for people and this is fundamentally going to be compatible with Cardinal so once you have these digital identities and you have really cheap cards that have trusted Hardware modules in them like at an gem style card you can then start talking around that infrastructure and saying well we also can now go ahead and give people loans and by the way the ACC has extensive records that are cross-referenced with different data sets on those 3.2 million farmers and at some point they can consolidate them and use it as an identity Colonel that's reliable and in the existence of things the ACC allow you to do voucher programs and lending programs in fact they did an experiment 2014 with a voucher program or they gave out loans as vouchers and they had a hundred percent repayment rate for the thirty five thousand farmers that they gave vouchers to one hundred percent every single farmer repaid at the end of the day because the way they structured things so there's there's really a tremendous amount of innovation there and if you get in early you can get blotching solutions to solve very simple and easy to understand problems that do require a lot of social dynamics to get solve and in the process you've now created a huge space of reliable people for lending insurance stable coin proliferation and so forth so throughout all of Africa we are exploring stable coins we are exploring alternative market places for security tokens we are exploring kind of out-of-the-box ways to market and explain crypto currencies the masses including potentially tokenizing sports teams will make some sort of announcement about that in Jinja it's an Uganda and we are exploring things like supply chains and property business registration now we always view Enterprise blockchain is the gateway drug to get people into the decentralized world and of course our preference will be to connect people to Cardno we will not actively build bridges to other systems because Cardno is awesome about other systems okay we'll 5g cause Mastiff population via DNA corruption yeah that's way outside of my ballpark I've seen the videos you've probably seen super smart on YouTube that some people do have concerns about 5g and if it has health consequences some actually come from credible universities and researchers and it is an interesting question does the existence of this technology given the shortwave and high energy caused some sort of health consequence and the answer is I don't know it's not my domain of expertise but it does give me a little bit of concern I noticed my brother who's a doctor tends to use a Bluetooth headset and doesn't like talking on the phone straight to his head he's convinced that cellphones may not be the best thing in the world for health but the proof is kind of in the pudding there where if we see a huge uptick in brain cancer or skin cancers or things like that that I guess 5g is probably somewhere to blame all right Dante says Charles Matthias said something about to be paid from recipient any update about that this concept of contingent settlement and we are exploring contingent settlement so basically the idea is that you have Alice and Bob and under a push transaction system Bob pushes a transaction to a known address of Alice now Alice can be a pseudonym or anonymous it can just be numbers that but it's some sort of payment address so Bob just pushes it in now here's the thing Alice did not consent to that that was just received the payment for the more Bob had to pay the transaction fee to push the value to Alice so when you take a step back and you deconstruct this you'd really rather have the ability to do transactions where you have some sort of interaction between Alice and Bob and process so for example at the base level you'd like perhaps to capture the commercial intent between Alice and Bob so Bob isn't just pushing money to Alice for just the sake of sending money some people do that we see that with dust transactions with Bitcoin for example but for the most part bob is buying something or bob has some commercial expectation or bob is making a donation but in all these cases there is an understanding between Bob and Alice that relates to the transaction at the moment that's met - that's abstracted from the system so the base layer doesn't record that so while the player saying and event happened because the nature of the event terms of the amount of value and the actors involved it's going to timestamp it and that records immutable the contract the social dynamics of that event are not captured in the system which means they're subject to debate so give me an example of this let's say that Bob goes to an ATM and withdraws $300 of value out of an ATM now that's a transaction now let's say Bob did that next in an Italian restaurant on on his birthday and all of his friends happen to be there you would say oh it looks like Bob's pulling money out to pay the check or some way is connected to this event he's a celebration there's people there it's expensive and let's say it's at 12 o'clock it's lunchtime yeah Bob can then go move 30 to 300 dollars from an ATM so the same type of transaction but now I've changed the metadata let's say it's a 2:00 a.

m.