Lightning intro to Chain Agnostic Standars Alliance | Juan Caballero (September 2023)

[Applause] so uh the meis stute uh sounded really cool uh I I like the introduction as a collaborative and not consultant thing that's trying not to turn into a consultany uh C do the same thing it it's a couple years old now it's been a community documentation specification effort for some time I'm going to try to explain what we do what we're trying to do how you can help um and yeah uh I tried to the slides don't skew too political but uh because of uh the topic I I can go into more detail um oh okay that could look a little better oh uh I'll email out the slides or something that can be attached to the recording if it helps uh so um basic mental model I use when I'm uh talking about cost as specs or or talking about the scope of Kaza I try to think in the most general terms about crypto systems not just blockchains lots of them are gags dhts other decentralized architectures for data transactions information identities um one generalized term uh I didn't find it on the slide is D pki decentralized public key infrastructure fancy way of saying actors are identified by cryptography usually almost always um asymmetrical cryptography private public keys so um wallet I'm here to finding kind of mly I'm wearing a account instraction shirt so you know I mean it in a more uh complicated way than just EAS in the ethereum world but uh if you think of a wallet more on the uh software stack uh think of it as a user agent the which is like the sort of technical term for browsers there's a lot of you know someone who uh Kyle from BR who's very active in Kasa always says uh we should think of wallets as the browsers of a web 3 platform which is sort of like a n joke among w3c people uh because browser people always talk about the web platform as the sum of standardized interfaces that make a uniform open web possible so if we want an open web 3 wallets have to sort of uh uh level up a little bit in the standardization department and have common interfaces that fundamentally is for me what why I got involved in Kasa what I think the short-term goal of Kasa is is to is to level up the standard ization of a web 3 platform and platform in the software sense has a pretty different veillance than platform in the economic political economy sense and we can maybe come back to that in Q&A or something but yeah uh wet is the actual agent of the user uh which in many cases they trust more than their browser lots of people have five browsers two or three wallets they trust the browser with their money and passwords a little less uh so thinking about that um gaps I think we can all agree with those are those are things that help you form interactions with a blockchain or other data substrate nodes I try to use the most general term uh like clients kind of complicate things ethereum doesn't really have like clients but lots of cryptographic systems have tiny light clients that can run in a browser extension you can run ipfs locally so like the node you have to think a little more abstract uh to get the whole category there contracts uh I had my favorite metallic tweet here uh regretting the term smart contract U we still call them contracts and CAA so keep keep them around for legacy terms but try to include anything that is uh a decentralized automaton a piece of script uh that's like a cockro that will survive the death of its uh creators and companies uh and yeah State uh ler chain so um this over here is like the classic EIP layering the you laid out in the ip1 and sort of canonical to ethereum um this diagram is like a little more uh it's like a uml diagram so each of those circles is a interface a variously standardized interface um not even this layer is 100% standardized as the etherum uh execution API team will tell you there's still little discrepancies between the data models of the different eth clients very slightly like having to do with type systems of languages uh and the top two total chaos and I think the top two are way less standard ized than the bottom two for a lot of reasons including political and ideological ones like a lot of people just think of the the DAP layer and the wallet layer as something that just needs 10 more years to even form there's this like uh faith in the efficiency of competition and Innovation uh I sort of think into blockchain culture so far uh but that doesn't mean you can't specify and pre standardize and collaborate and coordinate across competitive relationships to build ecosystems like the difference between an ecosystem in a market or a platform is just how much you're willing to collaborate with competitors uh and that's that's a big thing with CA uh part of what makes all standardization political is that you kind of need to collaborate across relationships of competition in Zero Sum games to standardize anything uh there's still plenty of soft power and there's still plenty of collusion and coion but uh you create a public option when you standardize things it's sort of like uh you know in Josh's book there's this communism talk that science is communism um you you create a public option if you document things in the open you open source as you go you design in the open uh that really sort of speeds up the needless competition phase of capitalist Innovation uh in my opinion so uh okay so that's the laring model those are the definitions again if see them again helps this be less opaque um so yeah I actually spent uh Monday through Thursday of blockchain week in Seville talking to browser people uh at at the w3c annual conference uh and fighting about ITF stuff there so um I I think that when we compare this set of software that is running all of this uh funny money to Conventional software what two software what open one software what one software um you that you sort of uh feel the different relationship to competition that's in the DNA of it and in the culture of it so far and I do think that a lot of these interfaces that are slow to standardize or are standardized like opaquely or by brute Market Force uh those sorts of patterns are happening because of the relative maturity of the infrastructure and I mean that not in the sense of like infrastructural software but infrastructure in the sense of like server clusters and legal arrangements and configurations of capital and like Clarity on which clouds are going to let us run what and how soon AWS will reverse engineer all of it um so in that sense like the it's hard for software to be more standardized than the economics behind it is uh so comparing web 3 software and standard to web two self ear standards I think the the not to not to be all uh base super structure about it but different configurations of capital just have pretty obvious effects on who wants to collaborate with competitors and who wants to build uh public options and uh neutral Bridges but um yeah um the this is just a list of complications in general of standardizing interfaces uh and uh you know hardening a stack as they say uh networking P Arms races sort of touched on that but you know like the the fact that all of this is running on video video CPUs uh gpus I mean it has a lot to do with uh those sort of competition with the conventional software industry um what else going to say governance collusion oh uh in honor of uh uh the topic the event today I I do think that um my lightning lightning argument within a lightning talk about Venture funding would be that uh we talk so much about decentral I in and like the layer on which we are the least decentralized is the funding like the funding creates incentives that turn each blockchain and blockchain system into a centralization point and a single point of failure in many cases and the collusion is you know uh pretty pretty onchain pretty obvious pretty easily documented but um so yeah I do I do think that it takes a a pretty deliberate and conscious and usually ideological coordinated effort to say like let's all take pay cuts and not work for vend Venture bash projects if we want this to go faster in a public goods Direction um so that's uh that oh and the one thing I will also say uh hand inand with uh huge amounts of venture capital and its usual conditions on progress uh there's also this thing that happens in web 3 and in most Domin Venture funded corners of the software world where once you have a project once you have sort of recurring revenues and a sustainability then you open source Your Design protocols convince everyone else to use your thing sort of like standard after effect like standard once you've captured the market um my experience coming from w3c I uh 5 years of working on uh identity Primitives um and open open ecosystems is that uh people who've already dominated a market don't necessarily Garner Buy in from their Challengers to standardize something it doesn't really work that let it very well in most cases so I I I often encourage people to think critically about when to standardize things and if the is really standardized uh if a market already has a winner that you're standardizing um but okay so chain agnostic uh gets gets thrown around a little I want to clarify a little the scope of Kasa because it's a standards Alliance and standards folks La a clearly defined scope uh where OCD nerds um no uh no Maxi talk is allowed on Kasa calls and public events we try to discourage arguing about which chain is better uh than which chain like all chains are a set of trade-offs optimizing for different variables there's strengths and weaknesses to any architecture and any piece of software so uh yeah Nazi Talk's not allowed even if you know this is in this is an ethereum beat up uh some of my best friends are Maxis but you know uh your your personal preferences or your ideological and political preferences for which set of tradeoffs you prefer software to make should not uh limit the conversation uh and objectivity thereof so yeah we that's one of the only rules of cost events is please don't show your products please don't show your favorite L1 um we try to keep things flexible and uh dis we discourage sort of specs written aasa from uh skewing towards a choice of network or a choice of system uh that's kind of impossible to do in practice it's sort of aspirational more than it it it's neither um falsifiable or testable uh nor enforcable but we aspire to not in write EV evm reference into the things we write for example um and yeah we're we're trying to abstract out web 3 development from specific blockchain system development like most people went three super complicated super esoteric super new no one learned about it in school except very recent arrivers um so most people sort of learned one blockchain system first maybe if they're lucky learned a second and a third but they're we're very very early for having a chain agnostic understanding of how these systems work they're so emergent there's like what's cool about blockchains is uh we're still finding the words for so one of the goals of the chos standard Alliance is to sort of differentiate what's cool about ethereum from what's cool about blockchain in general like finding a web 3 style of engineering and and architecture uh um so yeah we we write specs and specifications to help all developers public option design in public uh that includes non-technical stuff we have a a working group on ux the chair which is here somewhere there is a chair Ryan from Vision uh you can find him if you want to talk about ux uh we're actually doing Kasa is organizing a uh wallet unconference in Istanbul which is is in coordination with a ux event that Ryan's also working on um what else can I say um so yeah we read these chain agnostic Improvement proposals which are like eips but do not assume ethereum or Bitcoin or anything else like they their Improvement proposals for this abstract concept of web 3 engineering whatever that will look like someday in retrospect um and anything evm specific or or system specific sort of gets profile down so like each Cape we write there's each sort of blockchain system has to constrain that to be valid in their system so it's like a you know we have like a generic concept of an address and then a Bitcoin address is a subset of that and an ethereum address is a subset of that um so if you're familiar with that kind of technical writing like their their profiles um yeah and we do work outside of ux and outside of crypto systems like authorization and data Plumbing um we even had a proposal recently for a cape about uh this is really interesting the way nodes when they're trustless handshaking and discovering other nodes finding out which version of the software they run someone was like this is a shared problem I see across all l1s uh and so we may have a cape about that soon which is to say we hope to someday have tapes at all four layers not just the top two but the top two are sort of like the easiest place to start because that's really urgent people are trying to build cross chain chain agnostic daps and wallets so that's sort of an urgent need to figure out how we can efficiently develop those without having to start over every 18 months um okay so um yeah and one little note I I often find myself um asked like is IBC chain agnostic or is xcn like the polka dot cross chain thing chain agnostic and um so in in CA we talk about name spaces which are sort of like broadly defined systems like all of evm is the EIP 155 name space Bitcoin in all its Forks is the BTC name space Thea and like uh polka dot xcm is a is a uh runtime agnostic way that customized runtimes can communicate um between them but they're all block they're all polka dot uh run times and they're all polka dot addresses and polka dot RPC calls sort of like they have most of the stuff in common so like xcn is a cross-chain protocol within a system of many chains and multiple coordination chains in their case and similarly like IBC standardizes on the networking layer within Cosmos and Cosmos compatible chains so it's like um half of what uh CA tries to make it possible to abstract across they don't have to worry about it's already equal um it's already like interoperable so like I'd like to say that the the use cases we want to work on in Kasa are way harder like we want an ethereum chain to be able to spin up a Bitcoin private key and become a Bitcoin wallet as well in the same wallet I we want daps to be able to accept right at the different kinds of coins tokens signatures wallets like we we want to dream bigger than everyone just use xtm or everyone just use iidc like we're we're trying to document the basics so that people can build these cross system systems um and yeah benefit from the scale uh the economies of scale created by standardized interfaces um okay so I'm not sure I actually delivered a lightting talk I think I just talked really fast uh so Q&A it's totally valid uh for the questions to be like could you just rewind and explain the slide better um any anywhere you want to go thank you very much Jo hey uh thanks for the talk how does one get involved with the chain agnostic okay dang it I have a slack for that um yeah we uh but it it's just get Hub like you just find GitHub CH agnostics so you you can uh look at what we've already written um the I'm going to post these links actually all these slides are already on our name space in the Gatherings repo but um you know we publish capes and namespace documents that help you sort of like add a totally different kind of crypto to your daap or wallet um and all of it is just on GitHub and like if you find a typo you just open a PR and then you're a contributor there's there's lightweight IP for everything only a few working groups have make you sign a sort of patent release before contributing substantially um for the most part this is just in the open public good awesome aome um what um what's the process like for getting a proposal approved right because many eaps are just like sitting there for ever um is there a difference with this is being recorded I will not say anything about the EF process um no the the the process is actually pretty um uh we we try to make it more horizontal so like I I'm Ley and I edit everything and uh if we don't notice anything logically or typographically wrong and multiple competitors from an ecosystem all say this is accurate we're like sound sounds good to me things go in as soon as people that know what they're talking about have agreed there's nothing erroneous or misleading about PRS so if you're documenting a name space you're documenting something I know nothing about like eoso or something good example because we need that stat actually someone needs that and I can't write it so anyone who knows EOS um find two other people that know EOS that don't obviously work for the same people and review each other's PRS and it goes in cool lastly if I could ask one more um what kind of culture do you want to create in terms of like the uh the the standard body or like the contributors to uh you know the the props that one getting adopted like are you looking for certain things are you like nerd swiping just the maximum nerd swiping no I'm kidding uh no what what what do you mean by looking for certain things yeah I guess um how do you balance kind of like uh you mentioned before um the idea that if multiple competitors agree then this is a good candidate to uh you know Implement um so on the other side like um these competitors right or these like different actors they have kind of different levels of let's say like clout and so on right and then there's like uh other considerations around kind of like the arguments right there's all kinds of considerations for why standards might get adopted or not so like what do you place value on and like what do you kind of like intentionally uh you know sort of signal to the people in the community to place values okay oh that's a great question thanks for putting it I didn't understand it first um so I think transparency is the main thing uh in terms of cloud you know a lot of times when people talk about standards politics they mean politics in the sense of like soft power they mean in the sense of like uh unconscious subjective influence and you know cloud cloud stacking um we like if I don't know if super big name like theal comes in open to PR and then other people like I don't think that's accurate that doesn't work in my use case here's a corner case I can point to Links like I can document the corner case um I'd like to think I would be like sorry pic like that don't flat here I I'd like to think that there's not a cult of personality at all in fasta I'd like to think there's the oppos it already but that's what we strive for like it's like it should be non-commercial and it should be non relative you know as far as possible not Cloud power that that wouldn't fit us very far um honestly what a lot of times like we encourage people to take coauthorship credit on capes if they point out like a breaking Corner case or something like like a any significant contribution should be on the same level so like if a famous person opens something they end up with like five co-authors like I think that helps champ down the hero worship thing the influence I think yeah hopefully uh less collusive than w3c and less Nery than ITF which is very little hopefully a little faster than the EIP although the EIP process um is getting better and like I've been making a conscious effort to contribute more proactively to eips lately because it does feel like there's sort of a c change happening like there was a PR on E1 recently or well not E1 but on the informational e that has the process stuff in it so I do think that eips are speeding up and getting a little more um like overt and user friendly to P to openers of yeah pprs um there's been talk of possibly a uh cuz like the EIP oh sorry so in this layering diagram uh the the networking for eips are like customizing their process a little bit to to work a little more hand in glove with the all core devs and we've been talking about whether interfaces and possibly more interfaces and ER could also work more hand glove with the all wallet devs to sort of get the um sort of tailored process that would keep lower the number of eips and PRS just timing out or just going stale for lack of editor action the hard part is that um Kasa like I work pretty much half time it cost them like I work 20 hours a week and there's very few PRS the eips get 100 times more hours and have like no one who is even 20 hours a week devoted to them so like it's sort of just a guard variety labor problem like I would I think maybe the best way for uh our community as a whole to to uh Elevate the status of e editors would be like for companies to sponsor them or co-sponsor like it would be great if people could say like I want to take a year off from development will any four companies each sponsor me fractionally to just be an nip editor fulltime for a year like that's I think that would help a lot U across the different task forces like I think the core EIP process has been volunteer for a while and like ca doesn't need to be because we did the pass Hat Thing early um yeah and we have a this ch. e if anyone wants to donate uh to keep to keep the keep me half time uh and to keep events uh lovely but um yeah does that answer your participation yeah don't thank you okay see you soon I hope on good Lu yeah um you may have a comment about uh PC funding and got a few death stairs um can we pause the recording yeah I'd like to have a job for b um I I worked in VC for some years and I actually really agree with you that it's good for spotting these patterns um I was wondering if you had in a constructive way any any thoughts maybe like it seemed like there were some standards thinking around funding there's patterns that you're spotting is there anything there that uh that is worth almost I almost imagine one more layer I like Sol that um but but it's seems like you had some thoughts on that I was wondering if um I mean I think I don't know how quickly it can happen but we just have to evolve I think a vocabulary of exit Community I mean part of it kind of all comes down to transparency because um like I don't know where he went won was here earlier won gave like two talks this week about basically educating yourself to be a critical reader of coin distributions like tokenomics is uh a somewhat accelerated High version of classic uh like S8 disclosures and stuff and I do think like everyone in the space needs to self-educate on like what a good coin distribution is is or like what a fair distribution is or what a plausible or even feasible exit to community looks like mathematically because like if we don't have authorities like neutral authorities to give thumbs up and thumbs down to each new token launch pre-mine whatever we kind of just keep fall for the same over and over again because not not enough of us are economists not enough of us worked in VC to like spot it and not enough VCS have worked in VC long enough so there's some people under 30 around yeah um it positions of huge Capital Authority Capital allocation Authority yeah I mean uh I don't know there's a lot of there's a lot of EC that I love um a lot of smart brilliant Fair people in DC I really think it's just about supporting transparency or just like building up trans transparency as the norm that we call people out when they fall short of it cuz I don't know too much of the stuff is De uh I have a question in the future perspective like uh it is based on two Factor first factor is like uh if you consider the etherum Improvement proposal for example like 4844 for the DAT comp yeah and the fact is like the IND the AI industry like we have other large scale models to on the community Hardware so like uh that means they will have like what to say uh uh they will be able to run like a hug program on theing computer I don't know like the is oriented to the gas Feast do you think in the future like the gas Feast disappear like if you reach the level of how to say running any kind of program on his to she I mean uh okay so if I if I question it's like a Mo's law question like how many years before we stop have to woring about gas I I think that it's like gas is just translating um capital on one layer to Capital on another layer so like in a sense there's already gas-free l2s because they're being subsidized like uber was for 10 years or Airbnb right like it's it's all a matter of scale so if there's enough uh speculative funding and financialization in a sub Network that's sub Network won't really need gas and probably over time compute will come down faster than storage in the next 20 30 years but I do think like the fundamental problem is still uh political not technical like it it I don't think it'll ever come down enough that no one has to pay for gas it's always just there will just be a smaller amount that someone is paying to have an information advantage on someone else sorry that was an extremely cynical take but um I should end on a higher note than that but uh yeah I I do think like um the compute thing yeah it's it's it's going to come down a lot in five or 10 years and I think it'll really change the economics but I don't think it'll change the politics too much with my personal prediction which I hope is the way you were asking for okay thank you very much give it big Applause to J