Choosing the Right Oracle: A Practical Framework | Bianca Buzea - Chronicle Labs

[Music] hi everyone hope you enjoyed the first day of the main event I definitely did thank you for staying for the second part of the day uh for me this is my forth at Denver and I always come back with great joy today I want to talk about how to build a practical framework and uh BAS basically walk you through the different choices you should take when assessing an oracle solution so let's get started but before that uh we are here at it Denver an event for Builders so I would like to challenge you with a little question so I would like to know what do you consider when you build a d app what sort of choices do you have uh on your mind can you can we hear some some options from the audience anyone okay well I I wrote some some down uh the first thing that I put over here is chain one of the first considerations when building an oracle you want to uh sorry when building a de app you want to know okay on which chain am I going to build another important consideration uh you might have uh are gas fees and actually gas fees are tily the consideration of G of gas gas fees is probably very uh connected to uh choosing the chain so are incentives you might look okay what incentives do I have for building in this ecosystem versus another one similarly you might look into the liquidity in that ecosystem or if we're if we're talking about a novel uh ecosystem the potential of that ecosystem to attract liquidity you might also consider the user experience in that ecosystem and so on there are like very many choices that Builders uh have to take when building a diap but unfortunately I think there is one overlooked Choice uh and those are oracles unfortunately discussions about oracles are quite rare and there is this misconception that or oracles are the same I think mainly this comes from the fact that uh the name itself can be a bit misleading you can see on the uh right side the definition from maram web and it basically implies that it's an unquestionable source of Truth in reality not oracles are equal nor invaluable and here is what happens when there are different Oracle failures uh the top two Oracle failures are flash loan attacks and Market manipulation and you can see on the right side some of the most famous if you want uh cases in the ecosystem of Oracle failure now the good sign is that the good news I know this might sound frightening especially since oracles are a key piece of infrastructure and you're going to need them for building your D app but the good news is that these Oracle failures can be avoided and in fact you can have truly decentralized verifiable and robust oracles for your application now before I move forward and walk you through the framework of of assessing uh the right Oracle solution a few words about myself uh my name is Bianca buza I'm L Deval at chronicle chronicle is the uh the team behind the first Oracle on ethereum in 2023 it's spun out of maker da and now we offer the same oracles uh for the ecosystem in terms of Oracle flavors uh I want to walk you through three types of oracles push Oracle pool oracles and optimistic oracles let's start with the first one this is also my my favorite one uh because push oracles are proactive they proactively provide data to Smart contracts so uh you might think for example when there is a certain uh trigger for example a condition you might have like the new value uh different by a certain uh Delta from the previous value then you might want to update the the Oracle gets updated or similarly you might might have a condition based on time every X number of minutes let's say every 10 minutes you want the Oracle to update now pool oracles on the other side they are a bit lazy they requireed the smart contract itself to request the data explicitly so in the case of poor oracles uh the smart contract sends a query to the Oracle retrieving and relaying the requested information to the contracts now the last last Oracle type I want to uh talk a bit about are optimistic oracles similar to optimistic l2s those take some uh time to have data finalized so basically optimistic oracles they assume that data submited it it's accurate unless actively disputed and they have a challenge time you can see here um in the picture a Time p uh you have uh the value then you have a certain number of minutes for The Challenge and then the price is considered finalized now before going into the framework itself I also want to talk a bit about the different types of data that Oracle can deliver it's true that uh technically an oracle can deliver any sort of data but the most widespread if you want use cases are crypto pric data non- crypto pric data and ATT testable data so let's take them one by one when we're talking about uh crypto price data I'm referring here to real-time pricing for different cryptocurrencies then uh when it comes to non- crypto price data here you might have stuff like prices for Fiat currencies stocks Commodities Etc and last but not least and probably the one with the most potential nowadays given the growth of real world assets it's attestable data so here we're talking about any sort of verifiable piece of information and here are some uh widespread examples if you want again in when we're talking about a real world assets uh are proof of funds and proof of ownership now finally as I promised you at the beginning I want to walk you through a six dimensional framework for choosing an oracle so which choices you should make uh when you're uh picking an oracle I want to mention from the beginning that uh the order itself it's not that important what is really important is to go through all these steps the first one I put here is operational history so when it comes to Oracle you want to look into how this Oracle has been operating over the time has it consistently uh secure funds over time uh did it had like any hiccups and here of course operational history and the longer the Oracle has been around uh the more reliable uh it is longevity signals uh reli liability another important consideration here is total value secured or TVs this is also considered the golden standard if you want uh when it comes to oracles so here basically we're asking ourselves okay how much value does this Oracle secure you can see on the right side top 10 oracles by TVs from defy Lama and here of course TVs means uh battle uh battle tested kind kind of uh signals you additional um confidence in the Oracle solution next you want to think about decentralization and I want here um I would like you to to think about three factors here the first is having multiple nodes uh then having independent validators and having diverse data sources and I want to take them one by one uh the more nodes an oracle solution has the most the more robust it is but now there is kid here it's not enough that all the nodes are run by the same U entity or um a majority of the entities you want hear independent validators even when the intentions are sound if you have most of the nodes run by the team for example those can be prone uh to different errors or they can be succeptible to attacks similarly here you want to look into having diverse data sources ideally you want a combination of onchain data sources such as dexes and offchain data sources such as centralized exchanges with very good liquidity with an emphasize here on the importance of liquidity next I want to talk a bit about uh validators and uh node operators validators are also called node operators uh depending on the Oracle ecosystem you're looking at so when it comes to validators you want look at who is operating uh who is running the network for the Oracle ideally you want trusted independent uh validators that bring higher levels of data integrity and network security also here it's very important uh the reputation of those validators the higher the reputation the validators have the more incentives they have to maintain that reputation and to make sure that the data that they sign is correct number five I put here data sources so you might want to look into what data sources that does the Oracle rely on again here you want tier one reputable data sources with established credibility High liquidity and a proven track record if you're look if you're using onchain data sources such as uh dexes you want to make sure that at least two data sources are on the Chain uh are on the same chain in order to enable a M uh Bots and finally uh when it comes to data sources you want uh to have trace traceability and transparency across data sources and what I mean by that you want to have mechanism in order to check what are the data sources used by the Oracle and to be able to verify those data sources finally I put here operating costs oracles are gas expensives especially when it times uh when it comes to uh periods of network congestions when the gas prices are high then can can be uh very difficult to cover the gas costs if you don't have the right Oracle solution you might think okay one one way to uh address this issue is to just lower the update frequency but but the problem with that is that you're going to end up with Ste data what you want to have instead is to have mechanism that reduce the gas cost that uh associated with the Oracle without compromising the data fresh freshness for example in the case of uh Chronicle we're using schnore signatures to put together multiple signatures into a super signature uh and this allows us to reduce gas cost up to 68% finally I want to leave you with some final thoughts please remember not or not all oracles are equal they can literally break uh your protocol please keep in mind uh the the different choices you need to make in order to assess the right Oracle solution for your protocol and finally I want to leave you with this acronym no DDOT uh which stands for uh node operators operational history decentralization data sources operational costs and TVs thank you