Hashing It Out
Hashing It Out

Episode 18 · 4 years ago

Hashing It Out #18: Origin Protocol - Stan James

ABOUT THIS EPISODE

In this episode we talk with Stan James, Lead Engineer at Origin Protocol. They are building a mechanism for completely decentralized market listings and purchasing using Ethereum smart contracts. We go over the design, the tools, and the integration pathways for Origin, the challenges faced in designing this protocol, and the layer 2 solutions which improve the service of messaging buyers and sellers in a listing.

Entering work. Welcome to hashing it out, a podcast where we talked to the tech innovators behind blockchain infrastructure and decentralized networks. We dive into the weeds to get at why and how people build this technology the problems they face along the way. Come listen and learn from the best in the business so you can join their ranks. All right, guys, episode eighteen hashing it out. Today we're here with origin, a sharing economy protocol. Stan James, is the standard? What do you hold an origin? We're kind of fuzzy on titles like yeah, your engineer, lead blockchain engineer, basically like engineer who's been around the block a few times. Perfect. That's exactly the kind of person we want to be talking to for this. Why don't you give us a quick introduction? Well, first off, as always, calling, say hello, hello, and I'm Corey Patty. They say what's up? Calling and change it up this time. No, don't change. I change it up all the time. I'm going, we gotta Sizzu, going Hiogny, do it, fine, we'll do it next time. Okay, cool, all right, Stan Ley, don't you give us a quick introduction as to how you got into the space in the first place and what's your background is and what origin is? That's a lot how I got into the space. I guess I'm just lucky to have some people that were in it early. was about about this time last year that I was just kind of became aware of that there was excitement happening in the crypto space. Well, I guess, like everyone like, bitcoin kind of came on my radar back and went two thousand and twelve thirteen ish. It was, you know, first came on the scene and I was actually living in Berlin when there was like the first store that accepted bitcoin payments, and so I'm kind of kicking myself that I didn't sort of pick up on it then. But about this time last year, my friend Josh Fraser, who's the cofounder of origin, you know I've been friends for a long time, and we were just hanging out and he was explaining, you know, basically how etherium works and how there's more to crypto currencies than just paying for things and just, you know, anarchists whatever economics, and that planted a bit of the seed. And then a month or two later, when he said that he and his buddy were actually starting a company, and he sent me just this, you know, flurry of white papers to read on I don't know this like plasma and Zek snarks and I don't know, like IPFs and I can't remember, and I just, you know, had this night of staying up really late and like this, like this is awesome, this is the future. I got to be a part of this and I was practically begging to be the first employee. So I was there on day one. Is the first employeed origin. That's awesome. Like, what is the let's talk about origin a little bit. What? What's what problem do you seek to solve and why have you and how are you a differentiate or in that problem space? Um, well, the problem, such as it is, is that right now, if you have something, I could relatively simple thing of, like you've got a US bicycle that you want to sell, it's like pick which giant corporation do you want to share that fact with? When you as a seller, like you don't really, you know, you don't care about Craig of Craig's list, you don't care about Ebay, you don't care about Amazon or any of the like local classified. You just want to find a good buyer and the the block chaine just such a perfect solution to this problem of you just want to put a fact out into the world and like when I explain blockchain to non technical people, I say it's it's like just database that just kind of lives out there that you can write data to and it's not own money particular company. And so the use case of just a buyer trying to find a seller is seems like a match made in heaven for for blockchain. Cool. So there's a lot of components to that. That seems to spend a lot of common topics among our guests, one of them being is, how do you register assets? How do you exchange assets? Let's let's let's start on that. So I'm origin is using the etherium protocol, right. Yeah,...

...so is this a seven twenty one basic NFT system or no? No, it's it's much simpler, I would say, just on the idea we have listings and people make offers and buy the listings, with some, you know, a little more complications around fractional usage. So things like a I want to rent my apartment for a few weeks or I want to rent my bike for a couple days, but we we don't go down the seven, twenty one round of like trying to like, you know, give your used bicycle on chain identity. It's just there's a listening and here's the description in the pictures and etc. So it's like an open marketplace and you've kind of got a bunch of primitives in smart contract form which essentially enable you to create an auction or a bidding bidding system. Or is it more like in a direct by? More direct by an auction thing could be built on top of us in theory, but we're much more on like natively at the direct the direct by approach, with a little bit of like you make an offer and then the buyer accepts. Cool. And so what kind of volume are you looking to support right out of the box on this? Is this is something that's going to be requiring the post their own contracts or so that they deal with the volume issues, or are you have like a centralized contract that they can interact with? They'll be a set of contracts that they interact with. But at origin we're also very like we're trying to do sort of the full stack solution. So we're not just a set of smart contracts. We are developing our smart contracts and then, in parallel, developing libraries, for now just for Javascripts, like origin JS, n PM install origin, which allow someone who doesn't know anything about blockchain but knows javascript, like your sort of typical web developer. Our goal is that they could get up and running with a market place and just a few minutes. And to that end we're also making an example tap that sort of shows a sort of a you know, a best of greed example of how to interact with our CH and JS library. Got You see what people to integrate this into their own existing system, so that you would be kind of the the origin, the origin protocol, the origin smart contracts would be the point of entry for posting this kind of market, marketplace data, and they would be responsible to maintaining it. But it would be kind of like the truth keeper, the reference to single source of truth and Value Exchange Mechanism for actually paying for these services. Is that correct? Right? Right, Yep, and and just like the easy on ramp for regular web developers. We often talk about that video demo from the early days of rails, like a ruby on rails, where he was like a ten minute video where he created a blogging platform just like in real time. Yeah, and that kind of blew everyone's minds. So I think at least my personal goal at origin is to make a ten minute video in a few months of someone creating a market place like, you know, for use bicycles or something, and just like complete the whole thing in ten minutes and that's that's really awesome. Yeah, that's going to be using if is it you're you're only using a theorium as the payment mechanism, or do you opening that up to tokens on a theorium platform, or you are trying to stick with Fiat and availability option for this type of thing? Or is it like because you're using the technology of Atherium, does that let it your scope to what's being used to pay? We will support any arc twenty tokens. I'm pretty sure we're going to support that out of the gate. So yeah, and the rough strategy for supporting you know, Fiat would be, you know, something like a maker die or like this Nabal coin. There's also third party integrations which could support the origin protocol, such as coin base. Can decide. Hey, you guys are doing awesome work. You know we like what you got cook in here. We want to be able to people to directly, through our USD purchase mechanism that they already have set up covert to f immediately and transparently and go straight into the contracts so that they can use leverage their existing payment mechanisms. There's also things like shapeshift, I guess, which could also possibly integrate with this kind of stuff, so you could use other currencies get it directly to F and directly. There's no reason why this kind of these kind of processes can't be abstracted away and made, you know, transparent through third party partnerships and integrations. Yeah, like our dream would be one to make our protocols open and welldocumented and flexible as possible, that you...

...know someone who wanted to make a, you know, a bridge to Bitcoin or whatever, could sort of do that transparently under the hood, and our end goal is, you know, for that. You know just our you know the classic you know, our MOMS and siblings that don't know anything about tech can use this and, as far as they're concerned, if they know how to use craigs list, if they know how to use Ebay, they know how to use air BEINGB we want them to be able to use origin powered applications. Well, a problem that you may have is a problem of the entire community and that if you want people like that to be able to use the platform, they're going to have to get the token and then use the tools necessary to sign contracts to transact those tokens. How do you overcome that type of on ramping or Induser on ripping difficulty? Well, you're right that, especially like today, it would be really impossible to give that fluid of an experience for some of some of that's just like UX problems of how you present things. I'm not the UX Guy, but I know we a lot of the the UX design decisions. We're thinking kind of try to use metaphors that people understand, like you frequent fly our miles or loyalty points or something, for when they have to interact with with a token of some sort. Or, like you said, maybe it's like a direct transit transition of like you have a credit card layer that directly converts it to die or etherium under under the hood. Or maybe in a few years people will be more comfortable just, you know, having some etherium or other earctwenty token. It's yeah, it's hard to say what the future will be like, but it it. We know there's a lot of people working on making this on boarding experience easier and we're going to do our part from the application side and of course we hope that things like you know, like we're getting better and better sort of wallets on the mobile space that integrate and give a good user experience their coin basis doing all I can pay pal type experience for check out. I think I'm pretty confident that in a year, say two years, we will be able to achieve a pretty seamless level of use for the average user, and that's going to be a central for any sort of adoption. I think that's one of the big barriers that a lot of these cryptocurrencies have. Is Not just current currency, but anything in this decentralized land is very relegated to the people who have sort of this technical knowledge. Being able to make it virtually transparent and yet still provide a incentivization or even benefits model over what is currently offered by centralized system is really the challenge I think a lot of people are facing. Well, another you asked earlier about kind of like what what makes us different or why? Like hy origin, and I think that origin, or at least this approach, has some advantage in that the market that we're targeting, you know, so like an air being be host or someone who sells a lot of stuff on Ebay, they are highly sensitive to the cut that the centralized services take, you know, as air beingbe takes a bigger cut, ebay takes a cut. If if we get the user experience good enough that they can start to use it and realize, hey, we can save or we can I can make fifteen, twenty percent more money on my bookings by going with origin, that will be a good motivator to get over that hump or for them to even encourage their buyers to go or that Hump. So I think if we can get it to a certain point, just pure monetary ands innovation will maybe get them over that last ten or twenty percent of friction. Yeah, because basically you're cutting out the middleman, all these exactly, you just don't need them anymore, which lowers prices overall, which makes it competitive. It's you know, it's a really, really good interesting in fact. You know, I don't hear that quite often enough, I think in these conversations is, you know, I think original. Like two years ago I heard it all the time that, oh, the competitive model will be we're cutting out the middleman so we can be more competitive. But it kind of like that language kind of disappeared over the past well, since the tokens kind of took off, I haven't heard it quite as much. So I'm glad to hear that is that is what you're seeing. Is the value had there a direct value add Um. So there's a couple things I want to talk about. Let's just start with this one. I remember a while ago you guys had a identity kind of playground that I saw using Fabian Vocal stellars. So your C Five. Yeah, your s seven hundred and twenty five implementation. What are your experiences with that so far,...

...and are you integrating this with the protocol or what we what was what came of that? We're still really happy with it. We're still using it in our demo application on the test nets now. It's proven to be really flexible at hasn't, you know, boxed us in in any way. We're adding more and more verifications so that, like right now, you can you can, you know, prove that your profile, that you have a facebook account or your twitter account or I think even now might be pushed out live, maybe not quite yet. You like, you know, your air BEINGB identity, etcetera. You can make those attestations to your ear. See Seven hundred and twenty five identity and well, as anyone you know in the share in economy or marketplace space knows, like, reputation is everything, and so identity is going to be a kind of a key piece of making this all work and you'll feel a lot better about buying something from someone if you can see like, oh, they're a super trusted air BEINGB person and maybe you know coin base attested that their credited investor and their bank because attested their credit scores whatever. That's that's the future that we want to get to. Yeah, so let's talk a little bit more about the implementation details of that. If you don't mind, you can speak to it. I don't know who specifically works on that at in your team. You know you've got quite a team. The the the thing that I've I've noticed about the seven hundred and twenty five. Well, how about this? Why don't we start backing it up a little bit and maybe four our viewers, explain a little more about how it works and what the actual smart contract does and how you see that fitting in directly with your current mark, you know, orgin protocol. So it's right. So I'll confess I'm not the guy that did the implementation of seven hundred and twenty five. That was a nick who was awesome. But essentially, with your she seven hundred and twenty five, your identity on the blockchain, so to speak, is a smart contract that that you control, and other other smart contracts or other people on the blockchain can make claims about your identity, such as that you know this person is so and so on facebook, or this person's phone number is this, etc. And then you can accept those attestations of like, Yep, that's me, I that's true, and then anyone else who wants to sort of know about you can query your smart contract and see like Oh, it's you know, so Andso says that your phone number is this, and if you trust that Soanso, you know, might be twilly, oh, might be some service, then you can feel good about it. Right now, of course, most centralized services don't have ear C seven hundred and twenty five at station services. So you kind of have a bit of a proxy where, like right now, origin runs what we call a bridge server, kind of this bridge where you log into facebook using our bridge server and we can vary, we can get the token back from facebook that proves, Oh yeah, this person did log into facebook and they are this user, and then we will sign the attestation back saying the origin ID server certifies this person is person facebook. So are you trusting that? That make is at a trustless mechanism? Are you integrating that with the blockchain? It all in such a way that we can actually trustlessly verify that origin is operating in good faith? Or is there a trusted mechanism there? There's well, it's not ideals. So the code is open source, so anyone is free to run their own bridge server and of course we can inspect the code of what we're doing. We imagine in the future other people could run their own and if for some reason you don't trust origin or in some situation where that doesn't work, anyone could do that and of course, hopefully in the future you'll get you know, you know company. That might well be ideal. would be like new US government a test this person is this person, or your bank would actually sign directly to attest your credit score or something. The way, the way I picture what you're doing here, and I think it's how you did it, is you basically like when you create some type of Webap, you're a they have they have hooks in which you could use these third party services like Gethub, facebook, Google, so on and so forth, then log in as a user without having to implement that and your own Webap, and what happens is usually it sends you to a portal to their site and which you log in with your own credentials of their site and it sends back some type of Jason Webtoken and its success, and what you're doing is intercepting that and saying, yes, this person is...

...the person they logged in as, and then you can then craft a transaction with those things. That, that is the attestation that someone would normally do. For seven hundred and twenty five. That correct exactly, exactly. So we're kind of playing the part of you know, imagine that it was just like a regular website facebook log in, except our application is really, really simple and dumb and all it does is once you log in, we're like, Yep, you are who you say you are, and we write to the blockchain a saying that we have we have seen the token from facebook and we certify that you are who you say you are. It's cool because I was kind of a one of the questions I had about for any two sided market place, reputation is a big deal, especially in digital in the digital world, I mean there's a reason why Amazon pumps so much time and development into the user rays or that the item rating system right, getting reviews and so like. Reviews have become almost the way people's shop nowadays, which is artificially inflated through paid for reviews and so on and so forth, but like it's like it's clear that's how people shop and so and when it's when it's online, and you want to know who you're dealing with because in like, you know, the creepy craigslist world, in some cases may be dealing with some somebody that's, you know, just going to steal your money or trying to pull scams and so on and so forth, and you need to know that like there's some type of trustworthiness or reputation around somebody in the digital world that you're going to buy from, so that you're not renting some beautiful condo somewhere in a vacation place you're going and ends up and it's like doesn't exist. Yeah, it's kind of a real shame that cratings list has come to dominate. You know, it's great that it's free, but because they are centralized and just are the de facto monopoly and they've resisted all attempts at any kind of reputation system to get integrated or people tried to do, you know, build on top of it. So yeah, that's like a side goal of mine, is to be able to get some reputation brought to the simple, you know, craigslist style transactions that we all do. Now there's there's competitors, I think in this space, even within the blockchain world, and that's something I think swarm city was used to be arcade city, then moved to swarm city and they wanted to kind of generalize. It's too two setted market place. You. Are you differentiating yourself from them? Are you working with them? Are you aware of them or other people doing the same thing aware of them? I'm probably not the best. I briefly have looked at their code in the past, it seemed like development had slowed down a bit and, as I recall, they were not especially doing the fractional usage side and making being that general, and it was much more of the you know, x's for sale, someone buy Zex, not the you know for time showers, nife things, time sharing. How do we? Can you explain that a little more on the time sharing aspect of how origin works? Yeah, it's just sort of basically adding some richness to the protocol. Where when you in fact, for working on this right now, especially on the Ui side of you know, when you create a listing, depending on the type of fractional so like for a home sharing, it's typically by the night, you can just specify when it's available. The that calendaring information is stored in IPFs, along with all the other sort of metadata about it, and so when the buyer comes along they can see when the house is available, etcetera, and the seller can then approve or Dian I a quest to book. And Yeah, it's basically it okay, cool there. Are there any other competitors out there that you've heard of or thought of, like I think Slockett is doing something similar with their locks and microwaves and things like that. But it's it's I wouldn't call it a competitor, but like along the same lines of buying something someone else has as a service. Yeah, they seem, again, I'm not an expert, they seem a bit more focused on the the IOT angle of, you know, using the blockchain to unlock the door, etc. That's been my rough impression and they seem again, from my memory, a bit more focused only on that side. So then they don't cover the case of the craigslist selling a used bike. I think we're trying to make a very general protocol that handles all of these type of sharing economy market places, everything from, you know, the craigslist I want to sell my used whatever, to Airbnb when I'm at my house and then eventually all the way to Uber lift style I want to get to ride in the next you know, five minutes, or even just like Wi fi sharing. Like...

...you know someone has wi fi and there, there in their network you can maybe buy a, you know, some time or data from their network by automated automatic that CON transaction exactly. I mean I really hope that we will know where we've succeeded. If people are coming up with of all kinds of things that could be rented and sold that we haven't thought of that because the friction was too high before. So you know something like, yeah, like WIFI sharing or just, I don't know, access to two hikes or you're a good your tour guide or, I don't know, like the skies. The limit really on the other side of that note then, because the guy's limit, like what type of problems have you run into and trying to do this, like where we're worked out of catchalls do you have? We're trying to scale US out. You had to see anything in the future that's going to be a massive road block for you as you build out more and more on the technical side, you mean like like on the representations? Technically, yeah, technically, I mean the big roadblocks are the ones that are common to everyone in this space. Of course, you know, scaling, the confirmation times, etc. And the on ramp. More specific to us, there's at least a challenge around the Schema gainer representations. Right. We don't box ourselves into a corner in the way that we represent things in we're using Jason Schema sort of specify this is my apartment for rent, is this Mani apartment has this many rooms, etc. We don't want to unintentionally box ourselves in too much. So yeah, that's probably the big as technical challenges, apart from obvious just the ones that like if you deploy a bomb contract and in we somehow don't have a weight upgraded or fix it, it's probably the biggest risk. That's a big risk for everybody. Cot Com disputes. Oh, yeah, that's that's also key in this space. Yeah, I was just reading through a smart contract and I see you have a dispute mechanism in there and I kind of was hoping maybe go over some of the process and problems surrounding that because obviously that's when you have a mutability disputes become kind of a problematic issue. So if maybe you could talk to me about how how that's manager how you envision that will be managed going forward. Yeah, well, we're still investigating a lot of options in this space. Right now it looks pretty good for using ee IP seven nine two, I believe it is, is the Claros proposal for arbitration, and that is that's for handling the kind of the worst case scenario where and a third party arbitrator has to get involved when there is a dispute. Much like with centralized systems, we're going to push people to as much as possible resolve that amongst themselves. Of you know, like Hey, I got I checked into this your apartment and it's filthy. Here's some pictures. I don't think it's right that I pay the cleaning fee or whatever or whatever, and hopefully people can settle that amongst themselves and we're making sure that we have those methods available on the smart contract so that you can kind of change things even after the bookings made and then when it does kind of come to neither side is happy and it's just like we got to bring in something else. That's where we're probably going to use that seven, nine two protocol, which is kind of a nice just an interface basically for arbitration where you can plug and play different arbiters into it. Initially we're probably going to launch with a very kind of standard centralized arbitration system which you might just be customer service agents employed by Oracle or by origin, who not by Oracle, who are who are handling it, and then then we can decide either on creating and truly centralized arbitration system, which is what Clearos is trying to build, and there are a few others in that space, and we can kind of basically kind of watch the developments and not have ourselves box into a corner and, you know, basically pick the best of breed to be that ourbitation system. Or, and what might actually be the ideal is if we keep plug and playable, the people creating the market place applications could choose their own arbitration system. So, depending on the on the depth that you use, could dictate which arbitration system that you're using. I want to guess maybe back step a little here, because that the conversation that we're having sounds like origin is becoming the on stop shop application for this type of...

...thing, which isn't necessarily the case. Right like this is. Even looking at your website, you're it seems as though you're focusing mostly on the protocol level type stuff, like your current and corner of the market of making two seted markets work and that along probable to build applications on top of that to make the types of decisions like arbitration in the first place. As long as you're general enough, anyone can build an application with their own design decisions on how they do this type of stuff. In right, you may be the first entrance into this market using your own protocol by dog fooding it, but it's still free for other people to use and incorporate into whatever their else is building. That that's the key idea here. Exactly exactly, and it's kind of a fine line that we're walking with. Amazing having the proof of concept being out there, but making it clear that we want other people to build on us, which is why we have a lot of partners that have committed to build on our on our protocol. So they got so. Actually, this kind of brings up a question that I have with a lot of these decentralized systems. Is Business Model. Now, I know that you are speaking to the engineering side of things, but you brought up the arbitration side, which kind of things my by business, you know, not strong business acumen, but some business acumen. I'm like, okay, wait, that's the value add here and that there needs to be some sort of there always needs to be some sort of trust identity. If you're going to arbitrate something, you can do that in sucially decentralized manner, but so many needs to still manage the reputation of those people and and do some sort of independent audit of the the actual arbitration system itself. And you know they're they're reputation needs to be staked in that process being true in some manner. And I'm kind of wondering, well, first off, do you have a business model already set up for this and, if not, are you considering maybe the arbitration rout is being sort of a source of generating actual revenue through the orange of protocol for origin? Yeah, as you said, I'm on the engineering side, so take anything I say about business models with a grain of salt. That's probably fun. Yeah, I don't at this point. Arbitration does not play in super strongly to our business model, which is more around using origin tokens in the listing creation process and a sort of that side of the equation, not the arbitration side, but believe they we certainly thought about the arbitration incorporated it to arbitration. Okay, so your business models actually revolving around the origin token. So can you explain to me a little more about how that how that's used and how you actually receive income from it or how how is there? Do you get like a percent of everything that's that's posted, or your listings go a little above and beyond what the actual cost to directly listits with us that would be? Or we we definitely don't take a cut. That's kind of like the whole promise of not being the new middleman and trying to think. We're still settling on exactly what the details will be of how the token is used, but the bottom line is it will be you know, needed to transact on the network. Kind of going back to our discussion around user friction, I think we figured out a pretty good way to make it just so the seller is the only one that needs to be aware of origin token and the buyers can kind of transact seamlessly. But beyond that, I will say it's in flock. So I don't want to commit us to too much of what we're actually going to do and there are smarter people than me on the team that are working on that. Now. That's fine, don't worry about it. You know. That's just something that I like to touch on a little bit. Sometimes. The the the the big problem that a lot of these systems have is like what is the business model when majority of the business models surrounding all of these systems historically, throughout all of history, revolves around the centralization. So we're trying to come up with new ways of incentivising people to start and run businesses that in Roms, this technology that also still add the value at to benefit, but without becoming the next pay Valet S. Yeah, yes, I mean, like the basic idea is we want the value of the token to go up. That's how origin gets paid. We don't get paid by taking a little cut of every transaction. So, which I think is pretty exciting. Again, not as an economist or a business guy, but I like the idea that the the the terms basically under which we get paid as origin are set in stone, you know, on the immutable blockchain, and it's not like a centralized service like an airbnb...

...or an Uberus set are where, once they get into a good position, they start cranking on down the screws and like let's extract maximum value. And now that we've got to where we are, it's it's nice that, you know, in some perfect future world where origin is now, you know, like eighty percent of all transactions are going to origin. We literally can't go back and, you know, like say like he had. Now we're on top, we're going to like turn now, we're going to crease our cut the ten percent. So that the ideal is to me really loves that side. I mean it's it's a it's kind of the idea of token nice protocols in their places to find a problem inert like disintermediate it, create a tool that's really useful within that problem space and then saying you need this token to use that tool and say go and you hope that you then anybody who uses it as an incentive to get other people to use it at that grows the utility of the entire of the entire protocol or networkers or and so forth, which it's kind of like this, like cinners, just a self feeding cycle of utility, hopefully right this that's the goal, I think, in a lot of this type of stuff, is to build things that are useful, that require the use of the underlying token, and then the value of their token is usually goes up with the people who add value to the network. Exactly. And I like the idea that the payment is a bit more connected to the work that is done. Like well, it could have. Like, speaking technically, I'm sure any of US could make a fairly decent craigslist clone, you know, weekend. You know, we could probably make a pretty decent airbnb clone in a week or two. But even of what we created was like ten times better than the centralized services. We we could never break in simply because of the network effects that these existing centralized services have. So I guess I like the idea that would you are usually we spind lister. They they were like a little like it was a it was a nice like share your bike service. Originally heard of it. It's like sharing economy by and they just shut down recently, a few months ago, and it's kind of sad because I use the service a lot and we basically was like this niche just didn't make enough money. You know, it wasn't an Airbean be it wasn't an Uber. They weren't able to attract big VC's to come in. But I like the idea that with origin, someone who just likes bikes, could you know, follow our ten minute how to make a market place tutorial and get like a rent your bike out service made and, you know, ship it and going to people around the world and are able to share bikes. And and the guy that like wrote the thing, you know, he's not looking to be the next, you know, Uber, you know, Travis whatever guy. He just wants to share bikes and I love that vision that, you know, someone who just has a passion for stuff. You could just like write the APP. That goes out there and it goes around the world. Yeah, Hobbyist AIRBNB. How Weird is that? Like Hobbyist, like, Oh, I don't know, something's like Brilli into model trains and they want to set up the next Bo so they list like, you know, like a I creating a market place, so you could bring your I don't know, rent out space on this or whatever. You can create your own sort of like because of that, it doesn't really require the weight and the scale and the necessary, quote unquote, network effects. You can actually create market places for market places or advertisement systems for market places, all the bay the same protocol, and then people who are interested in a certain topic can follow the topic and not necessarily the specific website that they don't know about that's buried in a needle in a haystack of other other websites which are all super niche. Instead, you can create sort of like a reddit for market places, which is kind of an interesting like. I'd like the I think the term here that I think that you're trying to get to is, like I'm like, incentivizes micro communities, whereas beforehand, when you have centralized services that have massive network effects, those the mom and pop shops that have good service but can't get into the cat break into the network, cannot survive economically. This is potentially a way in which they can, because their success isn't tied to their own space, it's tied to the all all of the two seted marketplace space. So I could say you have one large, massive mover within the origin protocol, that the success of that one massive mover rises the tide of the success of all of the other smaller market places within the entire ecosystem. Exactly. So let's say that there's like some dominant I don't know what it's going to be, home sharing or whatever, and a going back to earlier conversation, you build up this like great reputation and now you want to go rent a bike on Joe's bike service that he made...

...last weekend. You already have your cellar reputation to go with you, like it's the the market place is already ready to go. You have to wait for people to build up reputations. Yeah, I love that. That network effect. It basically, I guess, the goals with these centralization we can make an even stronger network effect than the current sort of Monopoli, like the many network effects that exist within the centralized services. Yeah, this is something I personally've always been around. I think in two thousand and five I started. I bought a domain and s tried to try to play with something called, I try to criticize called online Karma dotnet, because I'd I noticed the same thing. If there's all these these reputation systems which are all across all these other websites, and if there was some way I could use like open ID, but I was trying to actually kind of like make them all talk to each other. You know, that'd be great. But no, the actual value add a lot of these central systems is that they hold on to that reputation that's there. That's their system, their way of doing things. There, metrics is there. So there's no way to kind of like cross it with the block chain. That just makes it so. I was going through your smart contracts again. So, as you can tell, I do my research a lot on the call. I start digging through code as ARY time. Two things that kind of popped out at me. Rolls are fixed to buyer and seller. I can definitely foresee market places that require more abstract role models, roll, roll, base, access control system. So I'm kind of wondering if you have any thoughts of expanding that particular aspect of how you're running the origin protocol. Can you can you first give an example of what you're talking about, like an agent or something? It literally has an enumerator in there to call buyer it. So I'm talking about a good example of more abstract roles. Sure you might want to assign an arbitrator or an arbitration system in the actual purchasing mechanism right up front. You might want to have an external auditor. You might require, due to international trade law, any shipping to well, this is this goes in. The other question I had, which is regarding states, is that they're not dynamic. You might require in and a traitorite law might require that certain things sit and holding and go through testing before they're allowed to enter the country. That kind of notification could be clearly updated in a blockchain and also trustlessly exchanged in that manner as well, meaning that, although I'm thinking several years down the line, at the moment it's kind of baked into in a hardcoded paint in the corner kind of way, where the fire and seller are fixed and enumerated values and the and the actual status of particular thing is thing I'm calling a thing is going through. is also a fixed list of states as well, and I've seen tons of different ways of configuring that particular aspect of it and I feel so that's one of those things you kind of also abstract out now. For now, I get for adoption, you want to kind of stick with what's easy and people can understand. I'm just warring. What are your thoughts on how you could grow the protocol so things like this would not be fixed, and why aren't you kind of doing that now, and why did you go with the decisions that you currently currently going at? Or do you see what I'm saying? Yeah, I think, I think I do and I think will one. Like you said this, we want to get out soon. But I think, more importantly, when to keep it as simple as possible, like you know what's the heart of the market place and trusting that. We're not trusting but also counting on that people will build on top of it and extend it. So, for example, you know it's very possible that, like right now, the the seller address, that's just an etherium address. It's quite possible, it's totally possible, that you could create a service which, you know, allows you and maybe some other trusted agent to both of you control share access to a to this key which represents both of you, so to speak. So basically, a venue holder, you are running out tables or something out of venue. You're the purchase particular person run in the market place for selling these tickets. The venue holder still owns the seats, so you both need to kind of sign off on the factory seats are allocated. So they have a stake in the game. You have a stake in the game. You're doing the selling, but they also need some sort of way of actually being participantss or contract. I G see that is being sort of like a situation that might follow the model but still not actually match what's currently the right. So when you see buyer and seller, I mean those are roles. It doesn't have to be individual people and that could be smart contracts, that could be whatever you want, and even things like notifications of...

...like I think you were saying, like you know you need something. You know, International Agency needs to be notified at the autumn was shipped or something. You know, there's nothing. It's a public blockchain, so anyone can set up kind of like a Bot to like monitor the contract, to see when the status changes and send an email or whatever. So I think there are goals. Is kind of capture the heart of what is a market place and not make things so obscure or, you know, convoluted as so as to block someone from making tools on top of it. I think would be our part, you know, kind of like SMTP. It's a very simple male protocol. There's a million weird things you can do with email, but the goal of SMTP is just to capture the heart of sending messages and let other people build fancier protocols on top of it. That makes sense and and some corey brings up a lot supposed be crazy. Is it is? You know, his ideals, and minor line as well is that, you know, we want to keep things it's thin and is bear metally essentially bear essential as humanly possible when it comes to this stuff, because you never know how it's going to be used. It's the only one to have the essentials and and part of the point of me bringing these kind of things up on these interviews is because this is counter to how we currently build centralized system mean, as centralized systems we tend to shoot for the moon and of you know, ultimate flexibility. You can do everything in anything in our centralized system. So we're building the whole kitchen sink for you because if we don't have that particular thing, we lost a customer, whereas with these decentralized systems, the onus is genuinely on the implement or to actually utilize the protocol in the most flexible way, and the only thing the protoclass to do is is provide a capability for them to do the things that they want to do on their own and they don't have to specify that necessarily, which is kind of why I did actually bring up the numerlation issue, because I kind of felt like that could be something that could be over written by a smart contract in and of itself. And but that again from the bare bone aspect. All of the things you've listed there, for even the state. The state has awaiting payment, shipping, pending buyer, pending seller, pending in dispute, review period and complete like. Those are all things that every single market place will have like, or it least kind of generalized its general yeah, yeah, so it was. Those are the state. It's that I think this work contract must care about. And anything else, like you said, is pretty interesting to me is that there's there's some sort of you could you could make a smart contract as the seller, you can make a smart contract as the buyer, and hell, those could even probably reference other marketplace things themselves. I guess I don't see anything really stopping right now. But you mean another another much lef you can chain chain transactions, so as things go through a supply chain, things exchange hands. This kind of protocol could actually do in a sort of like a supply chain selling mechanism, meaning that you can actually have things go all the way across and and with this tariff stuff and this waiting holding thing. That's could be a sidecase for once smart contract and once it gets out of that, then it goes to the open market place within the country of origin that you're worried about, and so you could partition things anyway you really want or need to. It's all about how creative you are as the implementer. Arly, that's what's up. I think the term here is is that that's kind of I guess the do Meta that people are talking about is the cryptoeconomic primitive. You want to try and find the smallest social contract that people have and then you want to write you want to generalize it while keeping it as small as possible in terms of the actual social contract, the thing that you're doing with another person, and then it's you have the smallest piece of something that's token ezed and written into a contract, and then you can test that based on that particular functionality and then build up using either up that along with other cryptoeconomic primitives, so then build larger and larger and more complex systems, but you don't have to worry about the functionality of each individual piece. But this is is the cryptoeconomic primitive of a two sided market place, one person selling something, one person buying something, and all of the MECHANIIC mechanism design involved with doing that type of behavior. Yeah, and what's what's interesting, because that if you were to compare this to a centralized system again, there are things that tend to happen as states change and you would need to implement those triggers. So you need to go, okay, this is gone, this is now in shipping. Okay, so that triggers the UPS ups protocol. HMM. You don't want to do that. With the way that this works and the inclination a lot of people have is to build that kind of thing. You don't want to build that in a decentralized architecture because...

...it's an open, shared database. So all you need to do is tell them I want you to track this particular market, this particular sale process, and whoever needs to monitor that builds their own triggers, integrates their own thing into it, and then we've become completely agnostic of whether or not they're using sap, whether or not they're using some logistics system. We don't care anymore because it's up to those particular systems to be aware of the protocol in a very genetic, generic, thin way. And then add the complexity that they need specifically by monitoring the blockchain and monitoring how the transactions proceed through the blockchain. So we've completely eliminated the need for all this central architecture that would trigger off an event which sends a notification to X Y Z systems, which we support and are constantly adding and removing supporting systems. Those supporting systems now are pull mechanisms and can and that changes the whole way that we we build stuff. That's not a very uncle word. Yeah, and the duty is like, and that's as an entrepreneur, it's like in every one of those like a business opportunity to you know, you know if this takes off, you know there's a there's a market or there's a business need for that ups integration. Also, someone writes that you need the like third party agent to manage your listings. Maybe you know you make someone writes that service. And it's because we're not trying to, like you, think of every possible use case and bake it in from day one. It's like, no, we're just trying to handle the very little thin base case and you know, let let a ton of other people build on top of it and that we'll control of your job script libraries, so you don't need them like necessarily. If somebody else wants to build custom stuff to interact with your spark contracts, they can, but you have an out of the box solution which people can then implement and actually integrate in their systems. Which is another problem that people are having a blockchain now is tooling. So I'm kind of wondering what does your tooling Road Map look like and what how do you see enterprise integration working with something like well, like I said, just basically, we're going to launch with the origin JS library, which is just to change the primitives to market place primitives, so you don't have to think about blocks and smart contracts and gas prices etc. You just, you know, create listing pass in the the JSON that like specifies the listing etc. So our goal there will be to make that as easy as we you know, we look at like the Stripe Apis that are of just like like very simple. Anyone who knows job skip can drop it in and use it, and that's that's our goal out of the gate. Shortly they're after a probably two thousand and nineteen some time, get python libraries and maybe other languages and yeah, that's great, cool, of course. That's a great ray to start to wrap this up. As there are there any questions that we didn't ask you that you hope that we would? We've done some really exciting work around messaging, which we're I'm pretty excited about because, for example, you know, in any market place there's often a bit of a back and forth of you know, hey, what what color is that? Or is that negotiations exactly? It's okay if I bring my dog or with your not the price by, you know, five dollars if I didn't do this or whatever. So whatever engineers you pan has the design this incredible messaging system which is a it is a private you know, so it's like just between the two of you but at the same time completely auditible so that if it goes to arbitration and you need to prove to the arbiter this guy's said that my dog was okay, then you can basically reveal a key to the arbitrator to reveal a conversation, and so that's been a really fun technical challenge that I'm really proud of the work we did on that. You know what protocols are used for that, because that's basically like, you know, encrypted with selective disclosure. Like what, how do you? How do you do that? Using IPFs, pub sub and the orbit DB is what's used for basically the the day store and back and forth. And then what is it? Web ARTC for the sort of browser to browser communication, and then the complicated series of each side does a one time signing operation in Meta mask or whatever with their private key just to sort of initiate and prove who they are, and then from that there's like a shared key that they use for the ongoing conversation so they don't have to sign each message to sag along. Essentially a state channel exactly exactly. And it's kind of...

...funny now that we've built messaging just as a funny aside, all the other engineers, including myself, or now like thinking of all these other use cases for it of like, you know, like in arbitration, of like Oh, I want to offer a ten dollar refund. Oh, we could stick that in the messaging system because that's like that's system works for everything exactly. Now we're like, Oh, now messaging has become our like bide channel. But yeah, it's fun great, all right, well, how can people get a get ahold of you, learn more, contact you? Well, obviously origin PROTOCOLCOM. We have open engineering meetings every Wednesday at one PM Pacific Time between involved by any engineers who are curious about origin. Two hang out in all of our developments done the open. So those are really fun and you see some amazing demos like that messaging thing I just described. And for me personally, I'm standard origin PROTOCOLCOM and wandering stand on our discord and on twitter. All right, great, well, things come on the show to calling just something there saying thank you. Yes, great, thanks for having me. If are your listeners out there, if you like this, then click the like button, but subscribe, show your friends, share it on twitter and find me at court patty on twitter, calling it at Colin Cuche Cell I, N SUSC and in the podcast that hashing it out, pod on twitter, websites, hashing it out dot stream, and that'll be it. Things coming on. Thanks very mean. You said great here.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (127)