Hashing It Out
Hashing It Out

Episode 5 · 3 years ago

Hashing It Out #5: Zaki Manian

ABOUT THIS EPISODE

We had the pleasure of interviewing Zaki Manian, prolific engineer, Founder of SkuChain, Head of Tendermint Labs, Advisor on the Interchain Foundation, and Executive Directory of the Trusted IoT Alliance. Phew, that's a lot of hats. Zaki gives us a glimpse into what it's like designing a scalable, secure consensus protocol. We speak to him about the philosophy behind scaling solutions, the differences in approach to consensus mechanisms, the role of blockchain as the world grows accustom to trust, and get a really solid look into the history of the blockchain space that led to where we are today.

Now anjoringquick note before we startthe show for episode five, we decided to try andpublish hashing it out through the Biin podcast network. firehose of allnetwork shows that you'd normally subscribe to. So, if youwere hearingthis for the first time this as episode, five of hashing it out and if you'dlike to continue listening to a show like this, please search us throughItunes, spotify, whatever potcasting a you're listening to subscribed to thisshow as well, because we're trying to see what type of growth we can getoutside of the Biguin Pod gast numer fire hose a little bit different on theback end, saying great quality content joy, welcome to hashing it out, poastd orretalk to the tecond, O metters Mi blocked in intostructure anddecentralize that one we dive into the weeds to get at Wyand help tofill thistechnology, the probes e FAC on Eway, I misin Amer. She can joine oeall right episode.Number five today, I'm also joined with our Trussy Co Host Collin sat was up.WHAT'S UPHY: Hey, hey all right calling and then our guess today is Zackananianand overall Badass in the cocurrency block chain space. You want to give usa introduction as to Um, where you come from what you do and while you're here yeah H, so I I've been in thewatchonstis ro right now, a long time, ISS ever started getting interested in sort of late Thosan and thirteen. I Cokounded a company in sort of it was like focused more on sort of the enterprisespace caled Tean back in twenty fourteen, I started a consortion withlike Bosh and Cisco and a number of other players in twenty seventeen andsort of shifted. My fogus in twenty seventeen to trusted Oto Alliance, I'vebeen working with Jaquan and eaten on tenderment D and Cosmos. Essentially,like I met J insonerview, the tenderment paper I think, before Wenbefore the cooriginal genterment paper had from house and reviewed it withthem. Um E'v been friends for at really one timetrying to figure out. You Know Ho how t how D it Get tendermen going, and sorecently I've been. I I've seen Y board. I tenderment, as had agenderment lass,which is uh just sort of like our um sort of building new business mals, Ikind of view Cosmos as kind of like the first product, where we had productmarket fit for genderment Um, but I think we're going to have lots of otherproducts that we can build as a as a as a company and h. You know I'm justreally excited about getting the Cosmos Man Tnat Li. I think it is. I thinkCosmos is going to sort of break people's brains a little bit about whatwere H wh. What block cams will look like over the course of twent Yn,eighteen and tweanty nineteen. My inmy interest has always been at like theInfrastructure Ayer, but in a Londa way s you know nobody was wanting to fundthe Inprostructur eer until very recently and sort of the ladder hot oftwenty. Seventeen is like when funding ot available and I'm just sort ofinterested in working in the next five or ten years in the emparticularer o.You know watciin technology and the broadest scoke just trying to build something Um.That is truly awesome. Um in terms of its power, I have a pretty queer visionabout where we're trying to go, and I'm just it's all about trying to like workwith the entire open source watching community tor to make that vision.Reality Yeah and I'd I'd really like you kno get to that vision, but I thinkthere's some building blocks to get there so you've. You just had a lot ofamazing accomplishments and things you're working on and 'm. It's likesuperprosis like from school Chang, which is it's apply chain Manag. It attemptid,Opunchi Management Um. I guess asset and track and trace Um to the truckteit it alliance, which is trying to create H, trust networks throughInternet thing technology, which of course, would be high volumetransactions. I would assume early a number of devices significantly on ascale that we're not used to seeing at this. At this juncture an and yeah athat way, galability or IOG is scaleabilities, probably the number onebigest challenge.

So I have a general tatement like so M.my current, my purent statement is, is like the block Chans we have today areusis um in the sense that, like ub block, can watchins need boot,scalability and pidacy in order to h Um. In order to be useful. U and wehave, we have a lot of work to do on scate billies. I Din' a lot of work todo on the privacy side. It S. I was a hundred percent creer to me how tobuild scabable private latchains, an as like a mastive amount of work together,yeah, totally Um and cogmos is trying to do withot it at work. Correct Tosmas is so you know we have the advantageand cosmos. You explain I before we get into thattoo much. Maybe we can explain to the general audience a little bit moreabout Cosmos and tenderment and how the two marry together and basically, whatis tenderment and what is Cosmos, and why are we building them? Yeah H, soyou know wha. So it's a great question. So we'vehad the advantage, because so I remember back in twenty fourteenJakeuan was like there is going to be a world wherethere are ginto thousands of public watcects, and at that point I find it twas like Okay Jay, I'm not really sure what you're getting it and I don'treally understate and Um, and that kind of, like just was to me like how herigh Jy was and Alo e the condect about why he was right. Um on this and itsjust always been arec a really likpowerful like memory and likeexperience bout. What how the visionary J is in terms of this space. It's areally J wike. What tenderint was sotentomit or Toc tenemint usuallymeans a couple of things. Ttenin is sometimes what we call the the thecompany Thats H has been contracted by the entertain foundation to Bild Cosmos.The company J started back in twenty fourteen t also goes by te name.ALLTATS JS history, as a foker player sort ofaffected in the and Um Yeah and J has been all in onBatchan, teabilities and tent borgans long before anyone else so won. Jreally is one of the they're really twut people, whore sort of back intwenty fourteen discovere the connection between an academic. Youknow O Bisnestin bultoeconsensus and the block chain seting, and theyrealized that really what we were doing with Prov ofwork was a sort of a reinterpretation and a lotof that academic research m and that that that that was sort of a a deepbeang of technology and ideas that were sitting in thirty years of academicresearch. That could be sort of mind for new ideas o how to work inTheconsensir space and to to the you know, there's probably a few otherpeople that they use to Galy et dominic and J, really the early Pirt on years.In this space, nominic Williams, yeah Domic Williams,ovinity O definity yeah M, and you see that really relected in en,like Wik Jane, wanted t what they wanted todo with Tenermeas. We ere just sort of figure out the details. F. How do youapply Um, like tisyielded research, so he reallywent back to what was one of the first academic Um, a sort of successful research, H, sortof results in the fiel of distributed yet to eleobals, arer itit's, an AlverUNTOL, DLS, Um hhh, and what he did figured out, how he would gapp dls tothe balchane like dl s like the th way Disi above our EARTINC, aTis was sort of conceved was not in the sort of like replicated ward repicatedstate machine setting that we actually need four Qalchins, and so she started working on Lik. How do Iactually create lite? A A modern production, ready, um sort of biable engineerlike from anengineering point of view, implementation of dls? You knowblotchin setting and he he also sort of like Um Hehe a she a a massin sipnification ofthe Algorthm, because all up until...

...tendermn, all of like Lov you knowtheseget Tal wit en a existed ash research projects. None of them were IEVE EA inproduction, use, Um and definitely never ened, upte nes aon strangers onthe Internet Um. So there was this huge neat. SOJ sere so j bills. All of this,and in twenty four teams, like I twenty fourteen eigt ye ha like tetendemateportocall, and at that one time I wasn't really an expert on vt consensus.I'E, probably learned most of the things I know about concensus Fromkay,but the reason I J st sort of believed that worked is he was like he would. Hewould break all everybody else in sort of naive, more niive ideas, um n Conaoexplaine to me why they wouldn't work. That's always a good, a good sign insecurity in distributed systems that you know what you're doing is when youcan bri other people's stuff Um, yes, itio, so th, so that a many more n I kind ofdescribe is la two year period. Were there were lots of experiments and sortof various commercial products? Um- and you know commercial experiences, withwit, Um l in the in the kind of bloc chains that people were fryingher,where the where there was market demand Um for a block, chaind kind of things.The the property is a tenderment O, a bunch of usually adversarial or non cooperativeparties. Sort of running a block chain over the public Internet there's justnot a lot of Um appetite for that Um and sebining twent a sixteen as sort ofthe spring an summer twenty. Sixteen, so I eathing came on board wet teminte,Um e waiht. Twenty fifteen iam still primarily focused on skew Chan. At thattime, uh skuchain was doing like a bunch of banking, pss, etcetera Um. We were, we were looking at and we saw in twenty sixteen and wewere like you know. I think tokens are going to be a thing like you're. Likeyou, Kindo noticed t at like Oh Someboa, like choken thing, an seein people seemto be into them M and we're like well, okay, so people, so we we we've beenlooking at. We, you know J and even and I hade been looking at Blockchainestalability, I sort of almost like a hobby for twinpis for for for two years.At that point and we're like well, we know you know that these systems onscale wo were there when I eureum was being designed. Oin ut his like, likethere was an explicit sort of rejectionab e lot of really importantideas in terms of scaleability. At that point, Um there was UM. You know it might havebeen wives actually Um that yo it was really hard toPioriti Taywal Bii back then MHM mhh. It was like you know, bit coin blocks,weren't full Um uh. Everybody was like. We don't even knowwhat the stuff is useful for. So why why in Investin Staling it well, notonly that, but like sometimes just getting something out is enough tostart a thing to happen, a movement you know,and they did they just got something out. It didn't lose steam, it just gotout and then a lot of problems with like the EBM and stuff like it's outthere. We can use it. So that's good thee! There re there's. There arecertain tat of Dligne decisions back then that, like I deeply regret, I alsokind of deeply regret, not yelling harder about it like on me as much as anyone else,'cause 'cause honesty. I didn't think if you was going to work. I go yowere in with the thing thatfailed and the thing that came after it was the one. The thing that was couldsucceed. That was my bad Um and I you really got handed to h what theaperience engineering team pulled off in twent and sixteen I was EBE gettingto a a stable public network. Um Uh is just, and you also w. We also have tothink whoever tha anonymous person is, who spent thousands of dollars a day,an all twent, sixteen leaving the network. It showed a problem that on o pers wwithout, that without thatperson, we really would not have had of the ability to do what happened withTeberian in terms of the whole like actually starting to use Thi stufp intwenty seventeen m, so anyways yeah I mean I, I think thatthat th, that hero was underappreciated. I also suspect that hero wasdesperately trying to demonstrate that the terium was never going to work and very much to be the Batman of aserium. If you will...

...well like a joke to let e just finish with those cosmasstorys Um is th t. The essence of the Coginar story was related. The reason why you want a world thatenables thousands of blockchanes why that world is exciting his that, if thebarriers, no a go back to the original cicanes white paper, that that theblockstream came out it their life as one a barier innovation is hard horkinga disun existing distributed network innovation is going to be Ruselu. Whatyou really want is a world wher n you cun, innovate on top of block chainsand Innivatan o the potical wearer, the virtal machinewar, et Cetera in a veryfast loosed way, and what t tendernen was the consensus layer of a world of athousand of thousands of blocpchains, the hotler Tu. What we realized and whywe decided to build posmos is that we gre that if every time you spin up anew block chain, you have to go out to like the world of all theAqitingpoliders, the exchanges, the dexes and and get them to connect toyour new block chain. Again, that's like a huge barrier: innovation, MHMAdwit, the Hubb he's basically saying is we'll give you tenderment will giveyou the Cosmus S. DK the building your block chain and Wi'll also use whatwecall ice like this inner blochin channelso that objects arelik objectsin the computer science, sentitem fo the term object. But you know you thinkthe simples object is a token. Gotokens can flow between these new block chains,so the exchanges and everybody all the aquidity, for that is only need toconnect it a Hob and make it access to the lype rich world of intervie of newassects and applications. So it's it's. Definitely a really cratey thing thatwe're building that sevenary afte Gon toeat it works, is rel, OI, sort ofgreat people's brains in terms of o of what black cains will look like intwsentand, eighteen, O thosnd and nineteen toeand and twenty, and that'sjust kind of like step, one tounwalking like an entire new world in inovation,so yeah, I we know in many ways we kind of havebeen have been a huge benefit interms of timing, but it does take years. I think one of the things that peopledon't bully appreciate out. There is just how long it takes between getideas: An engieer yeah. The engineering here is really hard yeah, but just butsome of the hardest engineering, the bigest in the world. U and Everyo,there's no margin for error. I mean you know it's not it's not thatbad, but it's it's truly complex systems, engineering, it's its COMPLAsystems, engineering at the Utterfonier of everything, Andwe don't have. We aredressed in the process of starting to put in abstraction layers, et ceterainto that engineering. I think that the sentiment of blockchain is a bitmaybe past, where it's supposed to be interms of where we are in term of like technology and B t. What you're sayingis that these are very very complex problems and we're just now findingways to obstract on top of them, but we haven't really quite figured out thebase later yet still because we don't know how the very fundamental foundation it lookslike in terms of what a block chain network looks like Wer like Cosmos, isCosmos and tenderment are one answer or one part of the answerwhere you have anaon being one part of the answer and how these thingsinterconnect and move assents across them still as yet to be found out andso to really look and make a comment on the layers that are going to be builton top of it is really really hard to do, and that's kind of where we are asa whole right now, which people don't quite understand. I I am I getting thatright, yeah, I think people people don't. I think the thing that isappreciated the least, is the value of clean interfaces between tints, andthat is probably core engineering value of costmus. Is We are frying to makeinterfaces between systems so that two systems on the seecs each side ofthe Intervang can invall sort of independently like if he looks away theInternet is self no G C ie was this: was this very clean, intervaws beweanhow two different computers send messages to each other and it allowdfor an enormous amount of innovation on topof and Indep and, like you know, interms of Operti, ystem etcetera? As...

...long as everything was willing to speakCS, tooke me in in Tenerment, we've come up with a number of like we haveic as as as affimitive like that, Ye. What we call the BC I intervas, whichis th Tere, btwee, a INSENSISLAOR or application layer, MHM and we're just at the beginning of sort offleshing out what these interfases it are and then building the t. Thevarious components around the different sides of those Ou know vases, but ifyou look get like essentially all o the work, what I've done in thelast, maybe ye R. Eighteen months, it's been bering focused at this interbaselayer, and I I think that's going to be it's like a key piece that has to bebuilt up. So there's there's a problem here, maybe not a problem but M, maybea roal block to piecing into the entire community,because you're, not the only one working on this Ifteriam is trying todo the same thing by becoming the baselayer and having smart contracts B,the interfaces to other offtain networks, the hub is kind of doingsomething similar for Cosmos, and you have other platforms like you, kN W,the quote: Unquote: Serium killers like Eos definity, and things like that thatare also tryot, become this platform. In reality, I feel like there's goingto be nowthezer equivalent to the various internets or Internet thatexisted before we made so one internet that connected them all. How were wegoing to do that? Because we have to find a way to marry. All of thesethings together, so you can have a you know, one standard to build on top.Otherwise it's just no. What are we doing? Yeah Um, so kind of you know, there's always a youknow: History doesn't repeat itself, but it Ryns y world, and you know before we ended up where weare today when we base it. Three operating systems. Ther th andprogrammers basically had two eightrs, almost it's getting to the point wherethere's just one API, because, as as windows, eats more and more the UNIXEigtti service area, all just kind of bepening one, but like in many ways a lot of this stuff andlooks like the nineteen eighties and nineties. Opening System worked at warYao, and I would I ex in like extent with wchis ar and they are autersystems.They have ther. They sort of sit in this weird place between an operating system in adatabase and they're h. You know that's what they look like they're, just aweird setting to run by their plaht. The Way I think about this is sowhatlwhat, actually we're trying to do and what? What bit coin? Reallydemonstrated was that there's this there's a huge economic there's,unexpected economic implications to having a secure computer, be avail.What I mean is byasecure computers of SEC, a computer that is almost implaudably expensive tocompromise its behavior tolike alter its baty. So that's how I think a aboutthe big one watching and tetheory o watching is there a computer. It runsacross thousands and thousands of physicalcomputers and you pay ind copucation is infredibly swell at it. HAS THIS MEE? Iis really powerful property where it is min. bogglingly expensive, like itwould toss tens or hundreds of millions of dollars to change the Executon of the execution of evenlike a single instruction on these computers and that level of Oltlornce,Tom and Lt and Brazilians. It Ha never existed. Thefore NSO, tewer, Lokreading operating systems that now that can take advantage of, and I exposethis property of secure, poputation toe the world an that sort of what this marcontract. Tatforme wars are his life. How that should be done and what formit should be done, and one of the things that I would w I woul I wu,wouldn't think I really like Aot casnos verses, almost every other project inthe space and almost every other project on the space is conceived as aMonoli in the sense that, like everything,everything isjected to everything else from the smart contract layer down o the Consensusara h,everything is is is is connected together in the way. Almost everyoneelse is doding. These things and imposmus were really about O es ar likediscret pieces and like you should be able to mix an match. So can you differentiate a little ofthat? So that actually still sounds a lot like...

...similar to, I would say, the plasmaeffort. Can you maybe differentiate that fore? COSMO SAN working withplasma, an theyre on o Ashe, is working with ANA. I thinkwhen, you know the PLADMO likepap original Bagnawigaver has a lot of flaws in it. I'm Tau to jessing about this annever timesUm. I Wan call that Outso I can know Wut other W. I mean it. It's just L.It's it'. It's it', a bit of amashp of Ecurit, present a lot ofideas as if they wrote all like one compu thing, when they're really just alot of different independent ideas, so I'd say in the communitation aroundPozma, there isn't a really good. There isn't a really good. We don't do a goodjogabng like this is the problem, and this is the solution to that problemand what clads must sort of imagines that there would be like a sicain or subjeve consensus layer thatwould be present in PLA, ilimintation and Wat Os oports w a Zon out, Yeah Andi can't halk aboutthe design space like in many ways, cosmos and Gasma arear are too approaches to solving problems in Sinspace and there verycomplimentary because of the fack of their kind of trying their life. Why plagma approaches the problem ofhavinbill communicating, falttolerance application visited chained in one way, Anvobnas worked onit another way, so a degree, a bunch of primitives thatenabled calable wattins to one of those primitive tesconsensus. So Kenderman isan approach. Thatconsensus CAFTER AG is attendermen inspired, O het reconcensoscases dcblabscaster as attender Mak, ixpired pesat censes, Aug, thresheld,Rway of kids, a different set of trade offs in that space, but where youAkwere, they CNSO like tendernes te lie mostly on correct distributed system.thinking. TBUT has a lot of messaging init like requires a lot of messagingover hem, which keeps t at the the the validator pool needs to be in theordersof hundreds pasper relaxes. Some of the am the Caser eporteres that thepolitanbladder tain are about relaxing sone of the dilike discributed systemsconstraints in relying more on prictor economics and just being like mhm ansome way. The computers in the system are not going to have certainty aboutwhat we're supposed to do next, but they're going to have a sort of aninbuilten economic risk model and they're going to locally reason aboutwhat is the best like West risky choice, so o maximum riscal war voice that theycan make it out. Yo know what what Lo, what torps of the Crain to build on? IfI could stop you, there, there's actually a really interesting point tothat that I think is unique to the theoromeca ecosystem right now and thattheperium has value and without value you can't propose that kind of modelyeah they needed to go from proop of work in order to go to proof o stak Um,you cad't just imp ride into that kind of kind of system. Indeed, so saTitolican Gla have have a stronger preference for exploring forelying very heavily oneconomics to drive consenses and then in Trestel relway peal reway is mostlyexploring t the the space reline more heavily on Pon sort ofsomewhat advanced tripographic primitives to drive consensus, so Li Tenerment is exploring the spase of howmar can you go with, while relying the least on Crotography, an cripoeconomicsOyou can waiter the Bareconomis and afterwards like with flashing and stufflike that, but yo don't rely on it for the very essence of coming to consensus,the caswor efforts are mostly focused on were on Exporin's space, where youdon't need Nobl Cotography, but you lower the messaging overhead Tepesen.This idea of the that the natiglibrium of the economic modelly drive consensusand then threshold reway is really an exporation. What happens if we don'tRelin or the economics as much and don't rely on on tetumid systems as much, but insteadthe line portography to Dret ee? If I can Kano recap that in a in a kind of adifferenta different way, I feel like...

...when I want to get make sure that I getan understanding of what you just said and it drives with Haw. I currentlyunderstand things, and that is casper or you know the proof of stakemechanism for e future atherium is relying on human greed to driveconsensus or K, O human self interest and optimization to to drive consensus. You have costmos forlying on the actual algorithm outside of HumanGreek to drive consensunce, and then you have something like dicinity, whichis derived on novel cryptography to drive consensus, which takes B of thosetwo things out of the situation. Is that right, yeah? That's that's! That'sKindo how I can see of these different systems oywhat orthe limits of each each one of them. One of the PL pros and Pons, so I thinka lot of people are kind of more familiar with the DM Paye, but from thecosmos and the dicinity space like the threshhold relate space. What is the?What is what is the PROEM compinence whate of th the limits on these? Whatis what are we sacrificing to to implement each one so hereis? What I would say, tendermen had turned out to be a verypractical system, um to build from an engineering point of view, which iskind of like why we have a main extertenderment right now or we havetestets for tenderment, that we'd run with strangers and they stay up and they work, and we cankeep u them when they get when they do when tey go wrong, and so as anengineeer you're like okay, like Tenerment, you know, you know, myattraction to tendermen is on one hand. I see he appeal of all of these ideas,but I might wonder you know you know in a world where wigo Ga Chans will bedriven over by shipping stuff. The tendermente approach seems to me to bethe best starting point from an engineering standpoint. On the other hand, like I've spent thelast five years, learning an enormous amount of chotography and lie the sortof more cripto heavy sie of things that that promescial relay represents fielda lot more state of the yard, but life theof. I hes Ito Hade significantengineering challenges because Je 'cause, like the field of of Lik, doingphotography in Engineering Si you like look at like what chrotography weactionmostly use over the Internet there. It's all basically stuff that wasinvented in the seventies. It's halyvetted, I what you're sa well, it's just wo, like we INVENTEDistriaran seventies, we're sort of wer sort of migrating, more aslating, moremodern titography that, as invented in the in theand, H, se yclose to thousandwith Um Elypic er, based if the element, anEliftic Cors, so eter ut, even the primitive of signatures and and tagreement are N, nd Ninetden and seventy seven forit Um. You know if youhome, Inno a like very very early, and I and it's taken a long time to get thelottle of understanding of how do you actually deploy those in a complexdistributed system that Um? No? I don't necessarily think that thatmeans that decial reway is bad or isgoing to take a long time, because ingeneral chotograpy, as he feal has matured enormously, and it is easier todeploy a more advanced Cridet morge of a researching crick like he pathwayfrom research to deployed engineering systems, is getting shorter El. We also have new incentives to doit. I mean th. The blockane space basically gave cryptographers a newworld to play in e active, an an incentive, tovelop e new apflihableCorphonography, a lot of stuff um, but you know like we're. We're also intendermentdoing a lot of research. Abot, O Yo, the sentially tendimate to point out umIC, will probably involve more advanced trotography on tantenterment. One pointI was and difinity is hired the people that you would want to hire. You weregoing to deploy ento novel CICK DOWN INTO A distrigent system like they hirethe people who who woul know how to do that. Um, a Wen Mi, you think about like automated economict Egence, likethe the complexity of like the bit coint mining provocall andlike the Soccerthm, the complexity of going all the way. AU, O foingtoGlaktapfer in terms of the ability of the of the system of reeason about itsown economic self interest. Becaus, yes, wiy Wat drives consensus is that thehumans operating o Ode and CASSESDC wanting optimize thewell and that prevental system amsort ofFallin, apart O forlight how to degenerate, Eqal Oberhim, but someengineers have to actually convert that...

...entire bougt protest into a procedurefor like looking at a bunch of critographic evidence, tof they're,coming in on messages and then making decisions about what messages tossendand the CDC state machine like is terralingly complicated. The tendermanstate machine is already kind of it. Nightnargivog I mean I can give you anexample, like we've been seeing on the test nets that for some reason, if thetestment halts, if you then so, like third of Temonin Power, GES Olarn, thetestment should halt it, his tats. What poperty that you would expect him anySACENES, say: Yea, N, Oi Power, Te ' lie it holds. So that's that's expected.What's unexpected is that when Yo bring two thirds on, you know two theres suchone online. It sometimes takes like an hour for the network to re T, politeactually start generating new box and we're not actually sure why why that isit's sort of a it's ely upbt of an ongoing mystery. Do you imagine like aleader election issue or something or we don't actually now it's like othlit o hundred engineersthend were staring at logs, trying to clear it out. So how can Ihelp you? How can somebody's outside kind of maybe mind this day idndactually try and help that prothat yeah like we produce it like it's pretty,it's pretty cool to reproduce it. It's sort of a pain in the ass because youactually to take a tendermenno like set up like a fornoe tendermint net word, bring two notes down: leave them downto like eavry, two notes up and lead two notes down for a couple of hoursand then bring or set up afor otenderment that, wheretake two dose down, then bring up one of the nose that he kok down and thensee what happened and like just getting more log. Data from that would behugely helpful. But what I'm trying to explain is more like a he things, like the edge cases onthese engineering systems, are swicky hairy to think about and like Gnen and Anton and Zarko, arkind of all of our onsensus expertions ICAM had like the entire tendermentstate machine like in their head Mit' a complicated stat machine, but it'spretty well pretty heavily specified and it'see and people have been endegent. We have a lot of expedients, debugging, it so N. I have no doubtthat we will figure out this problem. I'm just saying, but I you imagine ifdevuging the same kind of problem in depidity requires not onlyunderstanding the distribute system statemihe, butalso like flying to figure out like, what's going on Im various pieces ofadvanced Crompovety, aspitential sources of the problem, and if you lookat castor you're like do we inclement economic lead, thm, the economic lodgitincorrectly, so that Li yo end up in degenerate, gliberay and where no onehas any has a strong economic incensive to advance the system, even though,like you know, Mo correctness provs like if we had implemented this offercorrectly, that, like these, that we wouldn't enter into this degeneraty,glibery and and Hendo find that thougt. So, basically, my sense of a lot ofthese things, its just like emplimentation maturity, inplementationengineering, is a lot of wo is, is, is I's sort of alecent order. Managun harder than portocal designe and tendernit was likea good balance between like a good phrotocol and a thing that can beimplemented and Bebog anpanage of the real world. And it's going to be like. I think everything all the otherapproaches are really cool. I think it's going to be a lot of hard work,figuring out Howo buill tools that actually allow either one of thesesystems in the real world. Uh cus like we had to build a lot of tools thathelpe US understand how to run tendermen in the real world and, as me,get to more real world appointments, but Pasmas Gans laundhing. Now it'sgoing to get fun. You know when you're, when you're testing in an interpisappoigntment or your own small world, and you can get everyone on the phone.If the network goes down, you never test the condition of what happens ifthe network goes down for five houars and then wants to come back up Um- andwe never saw know but like now that we you run without Valitat or communitytestne Alli time we are seeing these educations that, like we just didn'tseeeor yeah, I I I'm curious because, based on what you just said, as a youngdeveloper, getting in the space for those who are listening, who ore younge bilers trying to get into the space?...

What should they focus on in order totry and contribute to problems like this, because they are such hardengineering problems? M EAN YE? Of course. The most naive way is tobasically take Cryptos Zombis Tunono program, smart contracts and startdiving into decentialized application development. That's the in my opinion,easiest way: Fom a developer who's Nacit to the space to jump in and dosomething. But if we're really going to move forward with the entire space as a whole, thereneeds to be some way for people to start learning things, so they cancontribute to proack you joke m. So IT'S R! It's! It's sort of it's areally interesting question that I like struggle with Iu like impacts. Like youknow, w. We do a lot of work with the the Blockshin, an Berkley kids, whichare mostof like a giant tribe of undergrads, the TV that are like otrying, yeah yeah, they're e great good, a theyre, great ant. We hired one oftheir founders. Whe Got TA conince Hem to drop out of Berklay and hes. A sunny is a great example of g of someone going O lie.Essentially, almost no experience to being able to contribute at thepartoall development war like really from Ero to sixty poyt, the fastestrampant I've ever seen had seen Zobar how to forget that, like excep sor he'sso strong ind, a potical design face that. Let him kind of forget that O hesalter has like almost no programming experience and ften doesn't knowanything about crotography. U So n Um! My recommen like this is the layer. If you want to builda career in this stuff where you kinda have to like, where theye's going to beinjuing Balu Um. If you want to make yourself Um, you know right now: SchooRiti developent in secure solidity, development, ture, more contractdevelopment is super expensive and so people who can do it, are sort of ableto make like a reasonable amount of money. Ou know it's a pretty profitableplace to play because I e my estimated that, like deploying like two thousandlines of solidity code on Eteri mean that securely is like half a milliondollars right now, and but I don't expect that to be truemuch longer. I expect that to tart becoming like an order of magnitudecheaper over the course of the next year year and half for variety ofreasons, and but like understanding how these like distributed systems, workand how to deboge them and how to work with them. So you know, if you want tolearn, I would almost be alike the whol. The reason I would takecomeplay with tender men, andset, building and start learning that a BildO onoe tenderment as on one hand, I'm like pushing the company H Company the line, but I've also seenthe real advant, like ited up SUC, a real practical system that you canwearn. It's like a thing that works, so I you change it and it doesn't workanymore. You can thin! You can think about like why did' my chaine break it,it's a lot harder! Then you're, working to read the asystem and Sir Tendermenthis no longer a Green Teo tystem and it's it's. It were really actualallyinviting people en to start playing with it that my my favorite thing abouttenderment and like the bact that it works that, like literally, we tend tofind out that that people had people have our building on top of us whenthey publish a blonpers, an they're like we used tendermen like tig KD, didand there's a bunch of other projects in the pipe oren't. Rarely like likeweek up in the morning, an tere like, oh there's, another bogpotof, a big highpropile projects it just like yeah builts on top, is tenderne because E. apparently we have a new partnershipWhil, they ha no about how we could help them, but like right now, it'slike they're building on top of it cause. They need a thing that solve the a bunch of strangers around a sharecomputer over the Internet probly, and it really isn't anything else m all right. I I have '. I have adifferent, maybe alternate question of like where these things fit in 'cause.Personally, I feel that, like most of the open block chain, space has beendeveloped to solve problems that are needed for completely trust in acompletely trustless environment, which don't necessarily lend themselves wellto private, could slike private enterprise level constraints if you'retrying to deploy enterprisennable block chains, you don't have a lot of theproblems that you have with. You know, creating an open d democratic money,which is what most people associate. Pre lack, chained, based applications.So H, how does like is, does temperament find itself being Um moreavailable to enterprise semitrusted environments,whereas other open block chains bases...

...or solving problems for completelytrustless environments for Ar Iow, Compos and weatly a trustless? I don'tlike you, I don't need to like you I. How does that work out? Is there? Isthere an optimization fom? What tenderment does gendermant slash?Cosmos, dods versus other implementations is on area that evenwithintenderment we do not have a green of Al Um in terms of just like Thi,it's a Philosophon 'cause. We all agree that Tentermin, not good Um. So thefirst question is really: Do you need isn't? So? Do you need Busnetinvolttolerance in an enterprise, ettit and so far? Most of the answers are noand you get such a benefit in terms of running a very simple Olber, then avery high threbot Tan, giving up ult tolerance and only going to a a regimewhere you can decheck and halt on holds rather than try to Awperat ambls wiketenerment is really optimized for the world of a small ish number of entityis economically incentivized to run a program collectively overthe Um, and that that that program, I ab youknow, is a block chain, but it coul be any sort of Wallchain oppencation. Thequestion is, then, is tenorment of good consensus for democratic money, and I think there's there's something of an openquestion about what the optimal consensus is for democratic money. Ithink a lot of the reason why Um, dominant and and lad and Calli a likeexfor directions that are like Mak different trade offes. They havedifferent vews on this Um our and you know, ran working on Shia, hasdifferent views on this Um h. The cadenateam working on chainweb hasdifferent views on this. Like Um, we kind of one of the things that wewanted to do. We we did very queerly with the with the with the design ofposmos. Is We actually sort of implicitly acknowledge that we're notsure that this is the right consensus for democratic money, because we didn'tship focurrency Um in the pormers we are, we are bringing accets from I Iinee furences, and we and the Adom Takin is Merya security staking tokenob, which there will be many in the future as other tenderment Chang Omgets on up and that it sort of provides that security where Um so yeah. I Ithink the question of of what is the rights. An CENSOF PTECHNOLOGY for inthe future of democratic money is, is unclear ut. I do know that for peoplewho want to run secure applications in a Balltrin sedy like DAPS, essentiallythe property h, a bunch of of like a fault accountable, a all contributableenvironment where thousands wer you know couple of hundred networkfunptionaries are running an application and you kn w. Theirsecurity is compromised in any way and we can completely attribute who who failed? Balt is like a reallypowerful, primitive, her building, a sort of scalable, secure application,world and sort of like to protocol of what isthe ideal for the call of Um a securiy distributive money like luckily, thefending environment today is sufficiently rich that weare, trying onfifteen different approaches and will see in my five years. Beton seems to begood man. I could go and very many different directions. H. I'vebeen doing this for five years. I've been like deep in the technical weedson a lot of projects for for five years. MHM. I I have. I know a lot of and it' it's a lot of it's Concernais,I'm going challenge alike how you actually communicate nnexpand the baseof people who know those things in in the exoditious way and, like Idon't want to spend, I don't at time, spend all my time like educating peoplebecause whike I have shit to do right right. This is t the poant of thispodcast and that's that's. The whole charter of what we're trying to do isis hash out some of the problems that exist and what people are doing to saltthem and you've gone over a significant amount of underlying fundamentalproblems in the blockchain space and the different projects that are tryingto solve them and what they're running up against and where they may besuccessful. So that helps people in general, but I mean it's just it's goodinformation to to try and get a handle on the current state of things and whywe're not superadoptive already is...

...because there's a shit load of reallyfundamentally difficult engineering problems that exist. Despite Way andnow you know, I think, what'. What's what what people are like people,people don't realize it to work on these fundamentalengineering problems. There was essentially no funday available. Um, you Knowso, I you know, tendemant wasself founded by Jame for a long time, there was UM. There was. You know,blockstream docways bunny, to do some stuff in this space. You know ze cashraived a little Bor, the money, but like the fundamental engieering, where,if you weren't building an exchange or something else, it doesn't really. Ithas nothing to do with this wayer of the system. There was a certainly nofunding available until summer of two thousand and fixteen, so we're ot ayear in Q, a significant amount of Hundine andwere just starting the pole like new talent, into the space bringing youtown in onboarding them Im skilling them to really understand what all wasgoing on and then and like you know, the thetalent fool is just Hertin Reali, okay, ther's, like isn't et some weird flashin the pand where, like they're, just like, we cheered up like hundred andmillions dollars being raised, and then Cuk e pencepis es crash and Itturnesthat nobody has any money like people are starting to realizethat a there's something here- that's permanete and a talentis starting ship.Over and like the world of people who are doing the kind of work that I'vebeen doing in the last couple of years is growing very quickly. Som reallyoptimistic about how near we are a year from now H, two years and after yearsor now we just Hinda have to by keep running in the direction that wererunning on and at the end, we're going to have something really awesome andamazing for Decuy porhumanity. But it also really, you frequently have to step back fromlike the. What token, like is the future Cardono is the future thousand oif the future of Ez a realized that we're all just writing open sour ofsoftware, we're all moving fundamental engineering fields in potogathe,industrane systems, incrictoeconomics orward everything is good to be opensource and, at the end, there's going to be something like there's going tobe winners of loses along the way, but at the end the end products five to tenyears now is going to be amazed, and I doubt that that's the fig I'mexcited about more than any like today, the project of the moment or h. You know what Turken has appreciatedthe most inthe last week. So I'm going to ask you a question that I I askpretty much everyboy who comes on here. I think I might not have asked Mo onthe last show, but sit's sort of philosophical question onthe design of of these blotching, architectures and stuff, and at you youmentiond, like there's, going to be thousands of block chains, but is theregoing to be one block chain which Kindof aces these central source oftruth and it sounds like you got this concept of a hub, an cosmos which wekind of function is that I would assume, but it sounds like cubs. Can altherGoin a be many hubs? You Know My my question is: Is there going to be onecentral, oracale truth in these block chained systems? Do you feel or arethey're just going to be this mismash of different truth mechanisms which canexchange value, and you can verify that value on that individual truthmechanism? But I has no context in the others Um. The AC is like that's like that is likea big metapyliosophical question, and I think that soap, all of the answersthat I've seen from it, have more to do with U People's prior commitments and theirfundriesin model then like a genuine attraction to addressing that just Soinposnus we do not anticate was not design into the system an we do notbelieve it is necessary for there to be one moot chain that wom Um, we we think sort of an ad hack world Um is going to be, is going to be prettygood for most people. I can see advantages to that rout wayer ofsecurity, especially if there is one especially a blockshaing based systemin its netive assets, become the dominant liquidity rail for the world,and I think the thing that you're trying to build when you're trying tobuild dais very different in he thing that we're trying to build the cosmosand so like, I'm all I'm on board her for for for trying to build that theother than the the sort of liquidity, Raio and Security Rail Er. This is likesort of movin system for the ENHIRE...

...world. Ittuld be really, I think it'sGN B, interesting to see what happens in that system, but I think short term.I would rather I culd corture my thin cosmeses, an even easier thing to build.I think cosmas and five machines are pretty complimentary to each other. Ithink, depending on what sort of starts to work will probably you know Casmusand Pasiae kind of merge and become very similar ideas in many ways, thePatmal obviously positions at theorium at that rute chain and the COSMAS,especially I when e conceived a todmosit wasn't lo like Ta, we can deedthe ponpes on Som Er en sixteen, like ourin network, would go down likeeveryday. We weren't where we were. We werewoling to bed on the security of anyone network at that Poinei time and Plolo.Very much of that, and then we'll see what the market was W ca could. Could you create a consmusnetwork as a plasmachine underneath the rude so they're, not even Ou, say thecomplimentary like they? I mean I've seen people who s actually a nosemodilel for connecting a sequel database as Ey PLA, a kind of chainlike a Stak Sta, I'm not sure how that works haven't looked into the detailsof it but M. I know that there's interest in setting up casca streams isbeing sort of like operating some have the stake mechanism m applied to the the hetransactions onthat? So you know I mean Patia allows for a lot of people to connect to arout chain, which is to me extremely compelling yeah, there's also thesolkodot group and like all these others who are just trying to cake thisInternet of block chains, which is Um. You know hard for me to Kindof defineright now. When I see im I mean Poli Ot. I is just pokanat of Asharvi, so herwaslike an interesting choisacical question and the design of loses that Ihave been trying to actually figure out. How is a Pocadot parriting written withibuilt in using a Wagan state machine on top of Ode, Substri Pi Acognas S,decay, applications running Hu, an independent generment consensus and Primea depinity aftermodel. Smartcontract running on top of the Botcher andnervousy Sto Giv Frik in any fundamental way, ow, don't think that they are ilike. Ithink h, the ine between a dia and for like a smart contract and a batchingapplication ar probably when Aduriam was created, it wasn'treally clear whether or not what the market demand was for whether it wasmarket demandus for provrammable money there, the market demand was for secureapplication platports. I think it's becoming Increasin requarethat there's market demand of both applications, but the technicalarchitecture of Bote popplications should probably be different and soCosnos, Pocedad and depidnity. You know always Amartin bzes work in divinityseem to be like we, we find e, like figure out what TAT distributedapplication space looks like, and basically the programable moneyspaceshould probably stop at, like the CLIMO routine contract as being themost complicated bhing you want at the Programaf Mone. I feel like the designof these systems should be completely agnostic so that the market can demandwhen it wants from the system, because, if you, if you put in rules of the baselayer that canstrain you to what is capable everything O on top of it, then you basically constrain that system towhatever that Tan is allowed to do. And but so the question is: Do we end up ina future where we have a bunch of networks that are single purpose or,like maybe contrained, to a certain ideology ofwhat it's supposed to do? Or can we build systems that are general enoughto do all of the things at once? It's like you know you can have if they'regoing to be Bosh, todecentralize trust layer for applications or and alsoprogramie money, or do we just have kind of things highly specialized to befinely tuned to do a single thing and we interact between them. So we havemaybe bicin be the money while, if mcan be todesentialize application, or wehave this network of networks that do all of these things, I don't know like.I otly o understand when I'm trying to ask there Um and I'm skeptical of the idea to doeverything II'm pretty I I'Mskeptical aboutwill beathing that is...

...economically efficient timaintain. I think that ther, I think, there'sthere's going to be an enormous amount of value to specialization, Um and there's going to be an enormousMont of value to m specialization and interrappability, and I, if you, if youthink that there's like a fundamental bat um that is like sort of the root ofof of the Cosmos Teem ideology about the world, it's that it's that scecialization, aninteroperability. Is S Ththe, key that s is the WATI system will be built inthe future rather than here is the general purpose? Do everything system,because Mann, the general purpose, do everything Sustins colleas right? People like like, Oh like we havetransactions and they manipulate a Mircle tree of state is actually liketen percent of what the system that we need five years frome now to be able todo like the system that we need five years in it. FROR now needsto be n, like Ha like support computations in Jourali that, likebeturns, your Nol Tos computations that ill run on secure, ontades popucationsthat are on insecure multiparty compucation. We need be Oto. Peoplelike economic punishments and availability, groovs and froadcroesfrom a variety of different systems. Um Y Noyou Youe btto have various kinds ofstat machines that run on top of the stop running, like probably a fairlyrapidly ofvolvinsor a kind of Poi primitive, and I think the lesson has been like why Bitpoin is crearly too inpexible Um.To really do any of this UF, even at Erim for Aitpexibility has enabled anenormous amount of innivation, but even eurianspexibility is and like the moreyou go down the road of okay, od, so ifyeure, you know, mickland was'nsitable. Now, soemended, I your if your Ange's not flexible enough, sowe're going to make another thing even more possible. Essentially you get this y. You you getto these systems that are so complicated that no one like how likehow do you practically manage them? Ebuild HEM UM. Where is if there areintooperating system that can focus on different Corse strents Um? We can getto a world Um. We can get to the world where, like allof this stuff works, yeah yeah. So I was wondering you tell me liketell me little more about the trustt alliance, what kind of stuff you'reencountering there Um and what kind of you know, problems you're, trying totackle withocert. So, what's like like to go o like of the Sop, I really I've. seennlikeever itpanding,like there 's a huge amount of onthusiasm and interest and possibilityin the intersection of Iot and bochangs and th use cases are were starting toget a sense of it. Is You know, secure edge computing Wik being able to USEAPokno for her for devices that are deploying the field in order to accessreached, Bik conflect, reachourses Ik power in bandits? Actess compute storethe Dita Etcetera in on nonsor cloud based environment.It's about like securing and authenticating devices as they comeinto the system. I think there's going to be a huge amount of interest intobacking assets that are existing on public block chains, Wi with a pigitaltwin of that asset, so letterit like a prototypical barrel wall or old bars,or for other kind of bloching traded asset like Tiiim, expecting to see avery rich world, Merr Yorset YS, seven, twenty one assets or Yeka evtwenty, one aftet likenontendable Tokens, ait Av back Bo that like had a physical twin in the worldand there's Indata Ean, there's like there's or there'se a keypair that sortof represents the state of that Im data coming out of it. That thing e's, nolonger just pictures of cats yeah. I think that that's that's ther. It's going to be a huge, a huge andrich market and a very important layer of the system to reen we having aliance is. I would tethere's a lot of duplicative worktime that is going on in the at...

...watching spice and at duplicative. Workhas te a at its core, like you kind of, were to break down, maybe the last fouryears of Ioky botterin projects, it all really just comes down to abunch of of all the companies in the stace exploring integration patternstogether, so e sort of independently Av, almost all of them, came to the sameconclusions like it's not like we'v, had twenty companies exploring twentydifferent ways of integrating, or maybe they explore fifty or a hundreddifferent ways, an on IOT device and a block chain to be integrated with eachother, but it probab they pretty much all came to the samely for five answers,and so that was the reas. You know Ande. We came together as a community andpublished the Um Yo Tosaiteoline Parpojectoy, that',sort of incumpasses bat, but Myall with the Allia has bend. You knowjust in terms of like the teem of trying to build infrascupture and we'llsee how we're tey're, trying to see how Harwe can gats in this, but it's Yolife is very much. I an expere is like it's kind of like this crazy experimentand not driving a thing with a trong propet loado like instead trying to getall of these individual companies of various cales that are all written bytheir own popit. What is you stop duplicating each others work and try tobeeld on top of each other's work so that we can get to scale paster M, butit? What does scale mean for you like what s? What is what DOS world demandgoing to look like with the nit world and block cane like what would are wedo you measure it in transaction volume or what do what o e you actuallymeasured m. i Aso care about, like numbers ofdevices connected to Ballcrain is probably the member that I'm thinkingabout, like what scale actually kind o me withthe scale. Actually, Mans is really actually like H, so th the digital money applicationlike the digital goal, application of Wachins, an application that existedscale UH dollars in the value number of market participants, Richis Aega,systeit's completely x, the Um fro, the secure programable Internet FatInternet application. Vopfom is also a thing that exists inbalching. What Iwould say is is what Ilt were not a scale yet because that market categoryhas not goten not to be RM, DBA and there's so many moving pieces from likehow you so got secure keys. How do you deal with the staability challengesthat actually having IA intems action volum? How do you deal with Um the factthat, like you, can't run of watching full no on an Iot device? How do youintegrate his legisy systems? There's all these significant technicalchallenges? Thut are not just Simor Fon The Timesof of technical challenges that we saw in the kinds of woctren applicationsthat I'd mentioned for Hade like reached scale Um. My concern is reallythat all lik my concern is o space. Is that without more cooperation? It'sactually it's like, if not in the interests of it's like, and it'ssomewhat antagonistic to the fundamental interest of of the startedcompany is en the you know, the the protocols that have been lost in thepace to actually truly cooperaten with each other, so I'm like I'm witing,their natural tendenc is to an expet, but but I'm trying to to like convincethem that there is the better world on the other side. Ine You work together,so it's like Theye they're, all struggled everyevery player in the space is like trying to competitively engy each otherfrom everyone else. Ca' do that with watching an it defeats the wholepurpose, all the KINDOF. I go ahead Olgre. This is this. This is not. Thisis not a a a Newsu I watching space, and this is, I tik summed up by aconversation that I heard I think a year year and ahalf ago between all of these storage plays in the space and it comes down tostandards right. It's in order for you to grow, especially in a and when youhave bultimums on multiple implementations of the same kind of coridea need a standard which could be built which, which INFRASTARC can bebuilt upon. The reason why we have such a massive ecosystem in tokens is thatthere came out with the Yarc twenty standard which you could build, whichyou could allow people to build, something that interoperates ith eachother and you can bolt he exchanges that then move throughout the reason.Why? Maybe you see this future in the Er? Seven, twenty one, the nonfungibletoken stuff is ecause one there's a...

...yuse case, but there's also a standardwhich you can build this infrastructure around and with the same situation ofmaybe scaleability or supply, chaind Mani, a or whenever eerseen you'retalking about as to bee ane in IOG Um. But like the the rule of an alliance isnot to leave the market, it's the follow the market, an like we createan environment intincentive incentives in a playground where people can but like it's. It'sfascilnating to me like to me. The business opportunity o standard of howon nft is Linke to Um is linked to wins a physical acset to aBocin Ottat and what the entire lifecycle povisioning nd all that stuffand the value of coming together around that is, is so obvious to me and likeI'll, continue to champion that ince I teliant. At the end of the day, I botthe individual, alines and Tembe just going to figure out and do what's intheir best interests, Oh and and how they perceive their bestinterest. But I still I stume you like a year into the alliance. Um Ins goingwel from a okay like we're, collaborating with each other and like the membership is growing andingagement is epening Um. I would really like to see I I would love tosee hy a segular Hokis by a bunch of companies, but how to Per Tho to market around a standard anerd, and that hasyet tok her, and so I'm still waiting for it Um and I it will definitely it.It won't definitely come UH nowhe, the the point of the alliance was Totris totry. This experiments is to try the experiment of I'm tryingto get these companies to convert and Cooperatig with each other, but Bok meO. There's all these market forces that also foll em in all of these otherdirections. Um! Well, that's! That's it! Maybe, possibly if I could get you totry and put on your your thinkers cap, how, if you were to ruminate what woulda standard for Iot devices, even look like what would be like at least one ortwo primary necessities for a standard to I, when Ye onbothat we could wit uh from the allions alike. What thisactually wo Li um it would'. It would be a s like it would be standards orwhat the like, what the harmor source keys or looks like with a provisioningprocess for those kees into the harrolsite and then how te his is, isand like where you go. Look out Um, whether or not when a device shows upthe way. Lik think about this question hi. I am a gold bar that I want not beoble trade, but Wi want to know as an avalt somewhere and iy Ento in thatball I contrayed the token tetogonized identity of that Colba. Okay, that'swhat you wantid, so the whife cycle of that is how did hat she get like. Howdid this particular keypair get associated withat o ar having it oProfe for doing tat? Then acrossusbetwhich says Oky. Now anyonewho holds that open this is where they go and look up to make sure that theGOI access of the audit law pice compelling evidence by that billbars still in the Balt and then powd that Dad asigned and distributed andwears Itstore, and that would be like a minimum biablestandard. Um O doing it and the question had always been like. Do weinvent that standard inside of the compins out of the alliance? And mysuspicion is? Is that like it? We even if he invented a standard in the Langeand advocated for it Um Market Addoctine Wuld not be there? I thinkthat the the marketed like markets have to converge. Like you know, people a lot of at like RC.Twenty was like the first token standard, but remember tundred party tokens existed tovoiceorse twenty dogans like like we went through so many iterations a tookit of of of Tokens on blockanes an polared coins, the counterparty tokenuntil the market convergon happen, and I I I wouldn't necessarily suggest thatit was like it was just the fact. TAT tetallas like went out and said: Here'sThe r see twenty standard that it happened. I think tetalla certainlyhelps bt certainly helps. I mean Batolic is also sort of verymarket responsive. He was like. Oh, like Bei lived on an eirplane andtalkedto hundreds of people a day, um o you know hundreds of people a month andprobably it was like. Oh I see this...

...common. I see the common pattern ofwhat everyone is trying to do. What, if I just like kind of Lik symplify itdown like winty Irocy, face it's a similar kind of thing. It's like.I want the alliance Totis because I'm hoping Wel celary that process when themarket is ready for it as of widely adopted standard emerging, but the market has to be ready for itand it it's like timing. That question has always been like a super hard hthings. I me to answer like whether or not the timing was right, and so youknow this. On the whole, it been like a low wrisked wave s like Oronthisexperiment, and if the anthe experiment is ongoing and we'll see what see what happenes, but it's hard for meto tell that I would log iat, I think it a huge market. Soard of thatstandards would merge at it. But it has to be the moment of start like think about all the thingthat has to happen like upstartup companies actually have to produce byhardware devices that can be attached to physical goods and scale exchanges have to support the datastandards igueis this enormous market, conversions that has to often be drivenin a cer N- and I would say, and many Yo know not allof the companies most ofo subste of Compani, an they wan to care about thatHAL lot of the counparies in the alines are also pushing ten of their ownagenders, an their own potic alls, an ou visions, and so it's going it itsamecy process. T andthe Oians is tin, O life a place for that messing process to take place, andhopefully, hopefully, something to emerge out of it. That will that willallow the IRT Botchan pace to truly get to ESCAP gapes. That's a pretty awesome awesome goal. Ithink that's Gon Goin to be great when eney starts panding out. I really dofeel like that ecosystem is necessary. You know it. It definitely happens.It's definitely necessary. It's definitely super important Um! It's IT'S IT'S! You know. I think a lot of people havenow thth. There's like you know. People talked about you, the Alli, an justlike Issou a token and raise a bunch of money and go it and drive its own approach likeI'm, I'm I reject to all of that and wh N, I'm I'm really in the position of of ideally I'd like you all to just startto talk to each other everybodyis working this pace to talkg to eachother more and hopefully discover for Yourselvs that your incentives are Aoyeah. I think it's funny too, that the token itself is kind of side out here. Like you know, when theyfirst came out, I think they had some sort of LE level legitimacy. Now? Ulegitimacy, but now it's kind of. When I hear token my my my my alarmsimmediately go up like thereis something else, you know, why are youin sentivizing this like what is going on? So I think it's great that you'restaying away from that and just focusing on like look. This is a socialcontract. We are all working together to build something that I goint tobenefit the world and actually benefit you know each other. So, let's just getin the same room. I think it's great that you're taking thot of PROCH STRwach yeah. That's my approach with the alliece. I think everybody is likeeveryone is trying to be flet. There's a lot of little pressures in the linesthat are life go out and draw iave action from the Alianes, and I'm likethat is the wrong use of the Matopnegabov. The rigte use of thenegatone is to wait for that market to discover, yes like discover what it wants to be,but there a reason why we, we structuredthis thing very differently from both interfisiierymalians and Hyberleger aboth of those cectures afe very, havediffeent sensees and indifferent participate B. I none of them moreright, tor that intersection of of iargy induction. So speaking of iprolledger, you wrote a bog about an to nother sixteene regardings. He kind of the idea of hypermeasure is going in theright direction. Are you still feeling like that or just like what theythought? What you thought they wanted to be was the right direction, but theyended up not going there or is there still place for them Um. So what hyperleter is right now um? Sothere's too there's two questions about like what hyperletery is Ri one is. Thequestion is: What of what fabric is, which is which is sort of the NEESPR iskind of Athat flagship, IBM, Tha, hyolegic project and then there'soverall hyperlegure thing Um. The biggest thing that I would observe isthe centure of Rabb in this in the space is going is shifting, like hasshifted, clearly and probably...

...permanently from in our privatebochings two public baces, and I don't necessarily think that that is, andwhile I was I've, always been hopeful that ha would avefer. It happened,maybe, like H, wh what I what happened in twenty seventeen in terms of likethe center of Gabbing, shifting to a public ball change, I thought was maybeanother three to five years in the future Um when I was sitting there intwenty. Sixteen like it was, it was it's it's sort of been shocking, howbast that movement has occurred, Um and so wile. I think that interprise boxinshave um are useful and a technical merit and Um, and one of the big thingsthat I would observe about all public boching technology today is the otherbig thing. That's going on in public altaring technology today is none ofthose systems are remotely enterprised ready and the work of making themEnterprise Riady is is, is immense and almost know, and none of the companiesthat are funded like essentially, the only people whove ever built anintervise ready, hublic Bochan in the Space Aril Um, and it's a very narrowuse case of it, for which they have achieved enterprise, wedding sotion. Sowhat general purpose had enterpised readiness for Boctin Po Worms is athing that is very good to be solved by traditional tenifie technology members and Hyperlegeris, a really good placefor them to to Woot to gether on that O. Would I,but I think it's obvious to me that, like Natfor, that Te sa the the son of the Bachain solr system isgoing t to public botans and the out th, the ordering bodies of that son aregoing to be Barious, labors of Private Wal, Janes and Enterprise Readyconnecting components to connect enterpris systems, two public altainsAnto private poutines. Is You connected the public auterines, and so I thinkthat as hyeten hae necessity, an hicolugure will exist Um, but we'll conwe'll become inextortly more lang around what's happening in the publicwatching tate and I think it' morestance. Everyone accepts that. Ithink when, when I was writing in twenty sixteen, it was just like theonly technology that, like remotely had a pop to intervise readiness in thespace was fabric Um and it was kind of like well, I wantto R. I want to ship.You know ve solutions today. What else am I going to us? Babert is still byFara the most entervised rading solution. The fabric PABI was aretcetected in a world for a world where this public balk chain pungible assets,componients and it was not as Prinar Um and somehow that attentionhas to be resolved and th that, whether it's adopting technology phon I likethis Bein baters already to a world of Public Waltams like Im andstuer I'vebeen working together, ano other collaborations. That might happen.There's a surprising number of of Youyopo, like I think it's fo peoplewho aren't in the space. There are a lot of new interpine bocing problamsthat are GOIN TA launch H in this year, more than you would suspect, Doyou namea few of them. Well, so I have and as a Overo Bunchofthen so I can't really talk about them. I can. I know that Oracles I and Inowtice completely out of band, so I can halk about it. Is Oracle? Iswanting their own sort of flavor of Abra this year? thisill? Be Interesting?U A but there's a lot of actually ovo vmwer is launching orn Um and there areothers that are, and so I'm I really want big enterprise singers to comeinto this pace. I I think, that's like just wlike, I think, there's certainthings that we really Gill suded to do, even as an extremely well hunded sortof block, chaned, stidup kind of thing on the enterprise. Ige totwerk were illsuited to to solve these problems, and so it will be re helpful to have moreenterpise bocting vendors come in that. I think that future enterprise watcing platorms,really need very clear answers about how the Nen to interoperates, withpublic wauctions, 'cause, CR, 'cause, there's just so much innovation,there's already kind of walked in like the money has been raised, even ifproeperrency crisis as today. No, this mitigation is enough: Te Resportation,ISO, N inenough of those projects and, like I don't as Cieer enteprisesto crash any time in t nears, egur uh...

...like and by crash we're not like basedon where weare today Ou Cun need thet not to crash by like eighty percent likby like ninety five percent Um before you know, the innovation funding kinddisappears. The CATS AT ONA lag they've hatted out ofver that, and so thereneeds to be like. I think, Interpros topons really need to have an Ey toboat, how they answer the enterprise readiness question, but also hardlyanswered interroperability with Public Chans question. If you don't Ha e s,you know public tans are not going to do ar notnyea. I would say, probably like on the Eria Bi at meastconsensus iskind of I s, hes kind of working on t, enterpise, readiness of Yeter imstackin a serious way now and that they have the scales an centary do this otherthan o other than what rippl was alreadyaccomplished, Um, and what what contensus is doing on the eteriumseye, still like from the public from publicautames coming to enterprise? It's going to be a it's, that's a prettyslow process and it's actually a lot easier for interhiz mendors to come towacin like PR enterprise boccades an from intervive watching to to publicbarkames is an actually a lot more plaugible past I ind a PAE Goin to endup being very important and that that' halfway as basically instead of you,know completely generalizing and solving all the scaling solutions of apublic blocktun. You just greeten API ind, your primate block Chan to heinterface with what the public fluction's already doing, yeah and it has to be it- has to have anotion of of how those teh ofes Cos Andrea. Now that makes sense on both sides of thatSyste yeah. A lot of the current solutionsthat are um are being proposed. The TH theire amount everybody having theirown kind of worshim Chang Li like what I would say about Babit, just like. Isure that you all s the biggest sort of designed law in the current terationsof Tabra, pretty much anything in the Hyprolega world right now is that theonly deal in the world of a single, unified trust boundary to death like like what what ineroperabilityexists within Dabric is only possible within everyone. Had the same notion oftrusting each other in by that interoperable space K,you've been run, many blkwe you can. You can have meltable difference USboundaries, um inside even a single BABERC outon through their channelsarchitecture, but there's no inaroaility to cross test. ACTERS, Um and H. th like that piece Clus, theenterprise radigness Pec, is Tointo Gev deet, all yeah. That's! Actually!That's! That's! That's exactly the problems I was NOIC saing. I Cyproleger.I think you put it way better than I did so. I appreciate that thati wholeclear. Some things up, Um, like I was saying a lot of the problems I saw wasthat most of the architectures that are proposed for these coonsorcom chains.They require everybody to be on the same kind of chain Um and you know likeyou, you create ive, everybody has their own node in their own network andit's all for the specific purpose for the specific trust mechanism for thispecific. This is the food industry chain, O this iss a luxury chane orsomething like that, and that sounds great and all, but unless you have someway of connecting that value to a bigger picture, I feel as though Umit's just too early staged to to be of. U I mean it will be tremendously valuable,but I think we still got a long. Ine Ago I mean yeah, like parody as avendor. Is Thinking about this consensus of tevendors. Thinking aboutthis like connecting change across tow crustfoundaries is just what we do.ATTENDERMENT UM kind of L, Ike t itacor value of ours Um. I would you know II'd like to see more from the enterprise side Um, and I think youwill there den like no about they can't talk about that are coming down the pipe there. You know, and I thinkwe also can like to acknowledge that technically IBM was re from a technicalpoint of view. It was almost areally good player to interp, introduce aninterprised Doctom behat. They have the distuted systems, atpets theyave, thecritography expergies. The biggest challenge with them is that they havetoo big of a consulting umbut. A lot of these problems, you're, noteconomically INS, you're, not con like having a big, consulting, aren't likeeconomically disintentivizes you from solving these problems, Um, because ifyou have a big consulting arm, you can and you don't produce a solution.That's interopable across trust, boundarthen. You can sell the samesolution to every trusto like to every...

...piece of the trustpatient, whereas ifyou build an interoperable dusion across trust boundaries, you actuallyare forgoing n consulting revenue, because you can't you don't get tostand up twenty or thirty difference of bicingbothins. You Stan that you built one interoperable sof picane solution thatwent across es valty everybody just connectsed. It was a very interestingpoint that I think is lost on a lot of people in terms of th in the untervisephase yeah. You know I wo LD. I worry alittle bit about that with consensus as well tat. Some senphasis has had theseconsulting as aspects of ther reentas, the company Cantentes, and it's really, I lik ibereay andexcited to see. Potentially companies enterprise companies coming thispas,TAT, Tro WH consulting is not theirpor, not prassional. Services is not to steron and really wants. She wants to shit that paform layer m all right. So I think that's that's anice kind of full circle of UH, a conversation. Why don't we wrap upAtri? This? Is that Wat an pretty much covered like essentially what I've beenworking on, but off Europe? It was all abusin. Well, O rign. I pated you yeah,covering a lot of areas. It's pressive yeah. Where can people find you temonionantwitter is probably h? Mwitter is, I is a good place to keepan Iony it ETA, funny that it's like it' in O, a point where, like I'm, isoipolitical entanglemen, now it's like you can't really talk, unto it heranymore whlike. I not you OIT, like I, I caneto like Kek down. I Ge e, completely broken Projectan wit, her anymore stike,polit faster. I wish that you would. Oh Man do I wish that you would. I wouldlove to inder so much more. If oll did more of TAT ye te closet e come owlikedoing these like cakedowns of project is I'm like in a bunch of DC,waterpolor telegram rooms, M and I'll just be like and I'll just slaughtersomething, but invite me PLA. It's hard for Mbut T, youget, pelitically entingled in toomany things, and it becomes it. It becomes wiser to have mostconversations in private or in like a week, somewhatrestricted populations of people rather than just like through the megaphone tothe enhiher world arigt at the cost of prolonging this conversation, I need todig in this a little further, because I I have a serious problem with thecurrent state of the public perspective, of what blockchain is and the projectsin them and that there aren't enough people taking down projects that arethat simply don't work or scams or or are illegitimate and or you know,misconceiving people, because there are quite a few of them and if we don'thave enough people with the prowess to take them down they're going tocontinue to exist, which they shouldn't yeah. I mean I c Toa certaic thatthat's one thing that I'm kind of I feel good. Where welike it's one thingthat I'm kind of Li e at least feel a little bit better about like how we'rehow like it's. You know, they're downsides to the investings basebecoming more and more N, stranger credited investors, but at least acredited investors can pay. People like me to to Kindof like Analye Dodgetanvelike Wellwhat, is the Likele Ha Ben Tas been being useful and successful,and especially early in the technology of Ajorty andbasically of trying to like turnover theprofessional investors as much as possible. 'cause yeah I get so much money rightnow that Wie in general wit to toget something that is questionable all right. Well, I'll tell you, like mypersonal figure, that you never know where someone's going to wind up of these people Ho Ol teiese projectsget in for different reasons. You don't quite fully understand: hereas Ha follescope, Onso, my Bigtureis, like I'm, going to till of a PPIFIC project. Epes makewantet com e itoddid Al Teaeoaearo Tetieede Mo an he might TA e t Tom at Aolthe time we had today I wasupstaning I peciat to come on the show, and I men for to kin o Brocin hepogrsalgas Yeh Te Really Piit Ata.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (108)