How Orchestration Makes Web 3 Feel More Like Web 2 (Dean Tribble of Agoric)

Transcript

welcome to crypto and wicker chairs I'm here with Dean trible how's it going man it's great great to be here CE in Wier chairs yeah in wicker chairs very comfortable it's the first time for me um so yeah it's kind of weird you know so um how uh how's it going it's going really well yeah we we uh beginning of the year we all realized how important chain abstraction would be to web 3 going forward and how well suited what we were building was to solving a bunch of the chain abstraction Challenge and uh we rolled out in February kicking off our our orchestration uh activity and have been building for for it ever since

and it is exciting to us and it's exciting to people coming to us wanting to build uh using that technology so it's very it's just about to roll out and we're just can't be more excited great yeah so I mean we're here at Nebo Summit um you guys were the headline sponsor so thank you for that that I think uh I think it's been a great event and we've been really glad to have the agoric team here with us um pushing these very important eras I think of abstraction and orchestration uh it's it's like a pretty popular topic right now so I think maybe like starting off you know what is orchestration and how is agoric implementing it so I'll start with chain abstraction because orchestration you know solves chain abstraction it solves interoperability

it solves for composability the thing about chain abstraction is that that's this realization that you know high level the users matter right it's users want to use assets and services independent of the underlying chain they do not care whether you're running on Google or Amazon or eth or or you know whatever they care about getting their token delivered or they care about taking their usdc over here and buying that nft over there and then selling it on that Marketplace without you know again independent the underlying infrastructure and that's really important because users care about something this is good this is what we need in order to get Market um and to get into the rest of the ecosystem and it's also really good because the crypto economy as a whole is realizing that

they need to deliver that value to users right we've done a lot of wealth creation by speculative you know activities um but we haven't done anywhere near as much value creation and that's what people are shifting to and we're at the Forefront of that what's challenging about that is exactly the making it easy to do these workflows that span multiple ecosystems we've got all these chains we have all these ecosystems connected by Bridges connected by IBC connected by you know whatever it is but you have to hand carry these assets around and orchestration is basically being able to program much more interesting cross-chain use cases so that users can do a single action to approve some something that right now is just

entirely too complicated yeah you talked about this shift between from from wealth creation to Value creation what does that mean you know practically speaking you like I mean what what kind of value should people expect out of like out of crypto and blockchains generally so I think a big part of that is enduser use cases that users actually want right now with like real world assets we're bringing things that that that have value outside that people want and we're digitizing them in order to get the advantages of integrity and performance and international access and all these things that actually increase the value of those existing assets in the real world to people bringing more users into web 3 that's an

example of new value creation as opposed to oh my gosh this token is going to the Moon everyone buy into this token at high price you know there was a lot of value there there was a lot of wealth you know there was a lot of money that transferred that coalesced around that but it was speculating on the future value that this thing might have um and we need to Now sort of as an ecosystem deliver that value so I said Real World assets a good example as people move in and are able to use their their existing assets to collateralize or or to make um uh things that they can currently sell in a local accessible now internationally um if they can reduce the cost of payments so you've got like PayPal is all about moving their crossb payments into web 3 and and using uh a

blockchain as their primary vehicle for doing that lowering the cost to them and lowering the price that they offer that service to you know to to vast numbers of businesses around the world all right let's bring it back to orchestration um because I think one of the one of the kind of programming paradigms that exists in web 2 or traditional programming is this concept of asynchronous right uh asynchronous execution which broadly doesn't exist in web 3 or crypto um and so agoric is bringing that to Market with uh with orchestration so what does that look like as a developer you know like what what does that actually translate into in terms of building applications and

then you know Downstream you users using applications like what will that change in terms of the experience of using crypto compared to you know what people are used to now right which is having to make one transaction on this chain and then wait for that to settle and then make the another transaction and so on yeah yeah so um most existing web 3 things and part of the reason why they don't have very good support for asynchron is they're all about building code that starts and ends within one block right I'm going to do a swap you know arguments come in figure out the swap do the swap and I'm done and it happens and I do you know 10,000 of those or a thousand of those in you know in a 5-second chunk and then I'm done and you know however long it takes to finalize it then we're all

there cross chain use cases right you know even simple ones where I've got usdc on eth and I want to buy you know a goric build and stake it uh you know require multiple actions as you cross multiple bridges that the user doesn't care about right it's very low level they have to sign each one of those all done with you know starting and end within one block do one step come back later you know sign the next step to do a transfer when that finishes come back later sign the next step to do the swap and so in the world of web 2 and then in the world of agoric uh orchestration and smart contracts instead you can have lines of code right you know straightforward programming where that that that expresses a workflow that's

more than one block right it expresses a multiflow action right I want to do this you know again take usdc from one chain and end up buying build and staking it that's just you know take my usdc send it over CDP when it arrives there move it to osmosis create it for build take that to a goric stake it as just a sequence of lines of code in the programming language for us and the programming language that you know 17 million developers on the planet understand which is Javascript the key thing that you get in modern development in web 2 is I can do this take an action over there send a message wait for a response you know async a weit as it's called the next line of code might be a minute later after that guy's done obviously after this block is done might

be a minute later it might be 10 seconds later it might be 3 weeks later because I did a UND delegate my tokens and wait for that to be done 21 days later take those tokens send them over here swap them for something and each one of those is this nice sequence in a recipe where that recipe spans you know seconds blocks days weeks whatever it is but it it's the set of actions in the sequence that the the person with the business flow the person with the workflow the developer understands and can easily map to what they need to do and so it's much simpler it's much more direct expression of what the user wants and it's and it's and it's enabling that set of use cases that users have that that they aren't able to accomplish them with three right

now right so this enables these sort of oneclick actions where exact you you execute one set of actions and one click and then those actions get executed yeah over time over several blocks possibly over several blockchains yes so like talk about the inability aspect of this and how that plays into orchestration yes so we worked a lot in interoperability right you know that ability and the connectivity that allows all these blockchains to to move assets back and forth the way that's largely deployed in the world right now though users have to trigger those individual you know they have to hand carry their assets from one from one place to the other um what what orchestration allows is a smart contract deployed on the

agoric blockchain to reach out and do that for you right and so that means now you can do these these use cases as you say where the the thing that the user approves is not the low-level Network language it is the user's desire the user's you know I I would I would use the term intent but that's come to narrow narrowly mean this kind of I want a complex trade you figure out where it happens in our case and and for the future of the web3 economy right that's got to be the complex thing that lasts over time that the user wants I want to authorize a mortgage not just a transfer of payment for for for for a house I want I want a long liveed subscription I want a season ticket that that I can use every week to get into the same uh the

same series of events that's a longterm relationship with the party that's the thing that I want to authorize that turns into telling a smart contract yes the user wants to do these multiple steps the smart contract onor then reaches out to those different chains and makes the requests on the user behalf controls accounts that it manages on the user's behalf to move the assets around the way the user wants it right and you know one of the things that I think is worth mentioning here is that you know moving across chains people should have uh different assumptions about the the the kind of trust and security guarantees of the chains on the state on which their transactions are being

settled uh I think currently you know users make those choices pretty deliberately so like I'm going to move my assets over here I'm going to like use ethereum or I'm going to use like EVMS or use salana whatever right like and people tend to be pretty opinionated about what block space they want to settle on I will respond to this but yeah go on so so like you know from the abstraction perspective as users get further and further away from those considerations because they're just like moving tokens what kind of user experience um considerations do you know we as developers Builders investors in the space have to you know think about including within the broader ux so that

you know you're not moving your tokens through like sort of a risky path if you will right um is that up to application developers is that up to infrastructure Pro developers like where do that responsibility sit because at the end of the day like users lose their tokens their assets you know even if it's just like their kind of like social media uh Digital Life or whatever you know like right so first off you know users have had to manually make decisions and invoke these underlying things there's an important difference between what I call trust and Reliance right yeah you know a lot of people say so therefore they trusted us like no no no that was the only game in town they had to use you know this ryion a bridge to get from

one chain to another um because that was the only thing they could do and did they actually have judgment about it I mean there are several popular Bridges with vast amounts of vast amounts of volume that are like a you know three of five multisig I mean it's just from a it is the equivalent of crossing a rickety rope bridge over a Chasm and yet people do it all the time because that's the only Bridge you can use to get to the other side now my analogy for chain abstraction and where we need to get to in terms of serving the user is uh uh you know I like the analogy of travel sites right you go to a site you know all these people came to Brussels obviously to come to nebular Summit right um that came to Brussels they go to a travel site they say give me a

flight to Brussels some people cared about am I going on Delta or United or whatever it is some people cared about am I going first class or coach some people cared about making sure they didn't go through peris right now because the Olympics would be would be chaos um there's any number of things that that one might have done you know and some people took you know uh spirit or ryion a some people took you know highend things right and the thing is they have the option to care about their counterparties they have the option to care about which cities did they route through nowadays they even have the option to decide am I willing to take a Boeing 7 you know 737 Max or whatever it is right um and people make decisions on those things and as people care about these things they'll get surfaced in

these abstraction tools but these travel things really are prividing abstraction where I can see those distinctions if I care I can say do that one but once I make the decision right once I make the decision that I want to take this usdc and send it over and get that token that implies I'm I'm relying on certain chunks of infrastructure and in that process I said and I'm going to go over this bridge am I doing Axel am I doing Union Bridge am I doing gravity you know what mechanism am I doing once I've done that I just want to say so make that happen do that I don't want to individually sign each transaction I don't want to differently um uh pay for my United ticket versus my Delta ticket right I don't want to have to think about you know payment for the the you

know all these services along the way I just want to say I've now decided make it happen in a uniform fashion so abstraction is not necessarily ignorance abstraction is providing you know the level of experience the user wants so that they can dig in or back out and then providing consistent ways of invoking and participating in it so that the differences in those decisions are not a problem are not are not are not you know you you don't you aren't bothered by those lowlevel details that that you've already decided about now make it happen right I've done it a subscription I don't want to have to decide each time every month I'm going to pay my rent which payment instrument I'm going to do I just want to have set it up and now and now my rent gets paid right yeah and so it's that kind of

thing yeah I think that analogy makes sense um it makes sense up until the up until the point you realize that like there the the the scope of risk in crypto is much broader than you know taking a 747 or taking a I mean I I guess it depends you won't get blown out the door of you know of your your plane and you know yeah but um that and that that's because like the airline industry is so highly regulated and you know so highly standardized maybe we'll get there at some point in crypto and we might be very far away from that and you know there was an ERA where there were a lot of plane crashes and people died that's right um well so I very much want to I mean I don't want to take that analogy too far the the we as

implementers we as an ecosystem there are a couple things that that I really want to want people to pay attention to as they build out the decentralization is value Val right decentralization gives us a robustness to attack it gives a robustness to to to untrustworthy Partners I mean there's a lot of things about that that are really valuable that are not obvious to the end user right you know that that that there are lots of systems that have you know rickety quality until they're underload and then they collapse under the weight right but man you can sure sell the heck out of them until they collapse under their own weight and so we as an industry do need to keep in mind making sure the protocols are secure making sure that that the bridges are good and having you know uh uh keeping holding other

accountable to that and as you start to provide extraction Services they need to be able to surface and be clear about what those risks are and select the right and upgrade the right underlying infrastructures to provide the the the level of assurance that users expect my my summary of what users expect is in some sense no unpleasant surprises right and the thing is if the market collapses and they're underwater everything might worked smoothly but there was still an unpleasant surprise for them that means someone didn't well communicate the risks to them and they should right but if a if a bridge goes under or they get rug pulled that's an unpleasant surprise right and so and so you you know that that responsibility or rather we should all you know work hard to make sure that

we we keep the level of Technology assurance and the level of safety to the point where you know we are ahead of what Regulators might ask for because this is is you know this is technology that's you know long way ahead of the level that they can that that they will understand the technical details and so it can be very hard to regulate in a way that is not dumb right in a way that is not you know do this particular protocol even though no that was that was 5 years ago we have you know zero knowledge is just better than that don't make me not do zero knowledge um just because some idiot is doing you know is still doing a bad hash algorithm right yeah I'm glad you brought up regulation because that kind of ties into this this other topic I wanted to ask you about so I mean like

you you've been around you've been around like the internet for a while and uh you know I I I started my career pre internet yeah in pre- Internet and you you saw the internet standardized and like web 2 St I mean the the the infrastructure I think standardized somewhere like in the 90s uh in early 2000s the application layer standardized around like 2010 and onwards right with the standardization of what HDML and CSS and like you know I think there's like a clear distinction I was a web developer in in like you know around that period there was like a clear distinction between pre-2010 and like post 2010 in terms of standardization and the industry

standardized uh by by itself you know with things like the w3c and the i e uh but didn't standardize due to regulatory forces right at least that's my interpretation uh with with web 3 and crypto we we have this regulatory pressure uh coming from the top and I I feel not very much uh pressure from the bottom to standardize protocols to standardize things like proving mechanisms interoperability wallet interfaces and things like that I feel like there's a real lack of standardization uh across the industry I me that's that's my sort of interpretation of it maybe you have a different view but yeah I wonder if you think that standardization and you know industry governing bodies like the w3c

uh could be useful in in crypto to help you know create better standards that would ultimately you know help improve user experience security usability applications Etc um and you know how do you balance that with kind of like the regulatory pressures right well so web to had plenty of its own regulatory pressure most of it bad right you know the the I mean there there are elements not you know the the the is an Internet carrier responsible for the content that transmits over the internet right you know if I have happen to send you packets that are about money is Cisco a money transmitter is you know is Xfinity a money transmitter is you know is AT&T a money transmitter right you know and there was plenty of efforts to overturn

all the useful mechanisms that would have killed the internet along the way there was all the attempt you know all of the ongoing iterative attempts by people that are as ill informed as as Warren to try to compromise uh uh crypto Integrity so that you could not do privacy so you could not do sign that could not be forged by pretty much any government on the planet right so so so for those of us that were Cipher punks for those of us that were in those kinds of things man there was a lot of that kind of crap right and and and uh the internet succeeded in spite of it not because of it largely there was the you know remember that whole period where um uh Gore invented the internet you know and there was the the the the lay cable

everywhere and all that sort of that led to a lot of dark cable that eventually you know people lost money on and then it became free cable that enable I mean there's all sorts of crazy things that happened there and we see similar kinds of things now I mean they're not they're not the same but it's that same kind of trying to regulate early when you don't really know what is but you're driven by entrenched interests that afraid of you know all this sort of stuff um uh you know using various things as a weapon I you only in a few places have you heard the cry but it's for the children but you know it's going to happen right that that someone's going to do some crap regulatory thing for the children right um and so that's something that we have to continue to push back on and do

things it's like no that's stupid there's a good idea there and we should do this thing over here and everyone in the industry kind to go yeah that's good that makes sense okay I can do that um I don't think a lot of standardization early is a good idea um there's there's plenty of you know really bad ideas that are inculcated in various things like you know okay what if the standardization was all programming in should start and stop within one block and then you can't do these multi-block actions you can't easily express this you know this workflow that is what the user wants and instead you have to just you know trap the user in the world of expressing low-level Atomic things that's that

would be bad and yet there would be people that would want to do that and so so to me the most powerful thing in web 3 is we get to continue to innovate we get to continue to grow and you know Cipher Cipher punks can write code in order order to solve problems and grow that industry um the advantage of various you know opt-in standards for and well as long as you're going to present an account balance present it this way and then all the wallets can play great that's awesome um oh you can't handle nfts oh I'm you know I guess we're going to need a new standard in order to be able to do semi- fungible tokens and fungible tokens that will then need to grow the wallet thing over time right and so you know we've got you know plenty we've got Decades of

Standards activity in front of us but I think most of it will evolve you know in the normal without a standards organization because because the because the users will demand it fairly quickly right you know that or rather you know you'll get the standards organization is that one's working right HTTP as a standard is God awful it's horrific from the I mean it's it's like let's establish a connection oriented protocol do one request and then throw it away and then the next time you know and and then on top of that we did a session level thing that that reproduces what TCP used to do but htttp you know create this throw it away create this throw it away you know create this tell you about session key then throw it away it's just I mean it's a ridiculous set of Standards it's just that computer

performance has gotten far enough along that we could eat that cost in order to kind of agree on this thing that was good enough we're going to get a lot of things like that and that'll be okay I mean you know we'll end up with with you know serving actual value to users in a way that that that that works in their phones or devices or whatever it is is um but we'll do it with supporting a few standards that all play with each other right yeah yeah I mean I think like once those initial standards were established I think there are some um some calls for http2 that would you know create more of like a streamable kind of data packet thing uh and of course there's like um ip6 which has been like really hard to implement so once those standards are established and you know there's lots of

value and sort of Industry built on top it's it's hard to to update those standards but but for example people were doing media across HTTP and then they gradually evolved to okay that's clearly not working let's do RSVP or let's do these other things that give us better streaming media with the better things and now it's easy to throw out to that when you need it right you know so yeah I think the media thing in h i mean really this is kind of history of the web is really interesting but you know the way that uh for example like media over media standards in on the web you know were very much inexistent up until like 2010 2012 uh and you know people remember that most video was done in Flash or real time player or whatever until Apple you know put put its foot down and said

we're not going to do video on the iPhone and all of a sudden we had we had standards emerge from that decision um yeah it You' be interesting to see if anything like that happens in crypto right where like one protocol sort of establishes standards and refuses to you know refuses to to interoperate with any other standard yeah yeah I mean I actually think like you know nfts were a little funny because that's like anything that isn't money right you know it's like okay nft is a giant category there is not a solution for nfts Right tickets are not the same as po apps are not the same as Bill of liting for Freight right they have completely different privacy requirements you know you know transferability all all that kind of stuff and so that's one where where you know you could easily I mean

I've seen standards for doing nfts they can't spawn all of the things that are not money right you know you're going to end up with this evolving set of things as people introduce new use cases for which the idea of you know an object or a tokenized thing is a meaningful thing and and we will end up with a broad range of standards for for the the broad range of uses of the underlying technology yeah well we're we're goingon to we're going to start wrapping this up here because we want to get lunch uh but um in our wicker chairs yeah uh yeah how can how can developers like start building with a goric orchestration uh products and what's the

best way to do that so we uh they are in the release pipeline now that' be out next week yeah um for the core of it um come to AG gor.com uh SL you know orchestration or just a gor.com and you'll find it the the library is available now you can start to build on it now the important thing and this was the most exciting thing about realizing how critical this orchestration was to this you know burgeoning cross-chain world of applications is that um you know while you can come and build on the agoric general purpose plat platform to to launch new web 3 apps it directly adds value to existing apps so what we see is these existing applications on their existing chains and you know Q3 will will Q3 will be

going out to evm evm chains as well and they have users they have use cases and those users are asking for more stuff as they always do but a lot of the stuff they're asking for is we like what you're doing can you make it so I can use my assets over there as well or can you make it so that it works for those Services as well and so a lot of the drive now because we these we have these isolated stacks on connected platforms is I like what you're doing can you bridge me across can you do a multi-chain thing based on your current business with your current users and so what that so the kinds of people that are coming to us or people that have a business they have a platform they don't need to move they can just add a goric

Orchestra ation to help extend the functionality of their existing offering to their users and so that just opened up the door to you know just like orchestration in web 2 or you know orchestrates underlying microservices to the tune of trillions of dollars a day in the likes of Salesforce and Bloomberg terminals all that sort of thing you're going to get the same thing in web 3 where existing services that have real business already they can benefit from integrating with the orchestration stuff to do a little bit of JavaScript orchestration to do these simple cross chain workflow so instead of just you know yes I can move usdc from one place to another now I can move you know uh Brett over to madic by selling Brett in the local market move usdc by the by the

madic and use these underlying rails in a way that is you know arms length because I'm able to do software I'm able to do programmability on either end of this kind of communication and so those are really exciting use cases to me that's what's necessary to build up the Next Generation of web 3 economy right right now we have this flat I can move stuff back and forth but at the point where you can do these automated things you could start to build larger and more interesting longlived use cases for customers and that's that will get us into the rest of the economy and so absolutely come to a gor.com check out adoric on on Twitter um X or 10 or whatever it is you call it um and uh uh and and come start programming cool Dean thanks so much thank you sir cheers