Canton：交易要保密，還能跨機構組合嗎？

0:00開場：機構進不了 DeFi，缺口在哪

0:00非常謝謝大家來聽這一場。我想，喔，很好。開始之前，先謝謝人在現場、願意聽的人。這一場要談的，是很多人其實都在想的問題：DeFi 是做錯了什麼，還是做對了什麼？為什麼我們一直沒辦法用大家原本期待的規模，把機構迎進來？這一場就是要把那個必須補上的缺口講清楚。Thank you so much for coming for the session. I think, oh, nice. So before I get started, I just want to get started by saying thank you so much for anybody who is here and listening to this. This is a session particularly about something which I'm pretty sure a lot of people are sort of thinking is: is DeFi doing something wrong, or are we doing something right? You know, why are we not able to welcome the institutions at the scale that we all have expected, right? So this session is just going to talk more about the gap that we need to fill.

0:47我不是要講缺了什麼、哪裡做錯，而是要講那個必須補上的缺口。開始之前，先快速自我介紹。我是 Jatin，Canton Foundation 的開發者關係經理。如果你聽過 Canton Network，Canton Foundation 是服務這個網路的非營利組織，用治理、成長和教育來照看 Canton Network。我做 Web3 超過四年，做開發者關係和成長，從以太坊生態到 Move 生態，現在則是在機構等級的 Canton Network 與 Canton Foundation。I'm not talking about what's missing, what we went wrong, but I'm just going to talk about the gap, which is sort of something that we need to fill. Before I get started, I just want to talk more about, you know, a quick intro about me. I'm Jatin. I am the Developer Relations Manager at the Canton Foundation. If you have heard of Canton Network, so Canton Foundation is the nonprofit entity that caters to the Canton Network, and we basically steward the network by bringing governance and growth and education for the Canton Network out there. For me, I have been in Web3 for more than four years. I have been in Web3 doing developer relations and growth, from the Ethereum ecosystem to the Move ecosystem, and now on the institutional grade with the Canton Network and the Canton Foundation.

1:39所以我有位置可以談一些缺的元件。我想從最重要的這一張投影片開始，而且先把該記的功記上。我們一開始，以太坊選全球狀態是對的。我不想一開口就說這很糟、這錯了，我要先把功勞講清楚。要迎機構進來，得先謝謝 DeFi 革命當初發生了。全球狀態才把 DeFi 帶出來，才有那麼多鏈上交易，才先把世界帶上鏈。So, which gives me the place where I can talk about some of the missing elements. So. I want to start by this one slide, which is the most important. I want to start by giving the credit where it's due. Global state was the right call by Ethereum when we first started, right? I don't want to start by, oh, this is bad, oh, this is wrong. I want to get started by giving the credit. In order to bring the institutions, we first have to thank that the DeFi revolution started in the first place. So global state was the right call to bring DeFi, to bring that amount of on-chain transaction, to bring the world on-chain in the first place, right?

2:27我想提醒大家：預設透明其實是當時做的設計選擇。它不是問題，也不是誤解。那個時間點就是需要這樣，我們也需要它發生。免許可，不用先取得同意就能呼叫別人的合約。可驗證，任何人都能驗證鏈上任何事。還有可組合性，你不必要求對手方先具備某某前提。這些依設計發生的事都很重要，功勞也最大。但這是好事，這是設計選擇。So I just want to remind you that this transparency by design is an actual design choice that we took. It's not a problem. It's not a misconception. It was the need of the hour at that particular point, and we needed this to happen. Permissionlessness, where we don't need permission to call somebody else's contract. Verification, where anybody can verify anything on-chain, you know. The kind of composability that you have got, where you don't have to consider a counterparty to have XYZ prerequisite. Everything that happened by design was important, and I think this is where the credit was due the most. But this was good. This was a design choice.

3:14同一套透明設計，讀第二次就成全面揭露

3:14現在退幾步，把同一句話讀兩次。我們說設計選擇是：所有東西對所有人都可見、都透明。同一件事也可以讀成：它給你的是全面揭露。所有東西對所有人都看得見。機構那些不該被看見的敏感資料，一旦被看見，自己的客戶就可能對這家機構搶跑。現在這些資料也看得見。所以只要把同一個設計選擇讀第二次。But now, let's sort of take a few steps back and read the same thing two times. When we say the design choice was everything is visible and transparent to everybody, right? The same thing can be read as something, as you can see, where it also gives you a universal disclosure. Everything is visible to everybody. So the data, which is a sensitive kind of data that should not be visible for an institution because that sort of makes their own client front-run this institution, right? Now it is also visible, right? So whenever we read one particular design choice twice.

4:01事情都有一體兩面。每個作用都有大小相等、方向相反的反作用。同樣，每個設計選擇都有一體兩面。我們說原子可組合、全面、免許可、一切透明、一切公平。也可以讀成：每個部位都公開給所有人，不管善意還是惡意。這給要上鏈的、規模以十億或兆美元計的機構，開出一塊很大的揭露面。心態上最大的改變是：我們說機構該上鏈，到底是什麼意思？This is the thing where everything has both the sides of the coin, right? Every action has an equal and opposite reaction. Similarly, every design choice comes with both the sides of the coin. When we say atomic composability, universal, permissionless, everything is transparent, everything is fair. We also read that every position is publicly available to everybody: the people with the good intent and the people with the bad intent, right? It opens a big disclosure surface for a trillion-dollar or a billion-dollar institution coming on-chain, right? Now. The biggest change in terms of mentality is more about, like, when we say institutions should come on chain, what do we mean by it?

4:53機構帶來的東西，跟一般 DeFi 不一樣。他們帶著接下來幾筆訂單的規模和方向，也帶著對手方。他們投資哪裡、不投資哪裡，還帶著專有研究。他們來到你的區塊鏈、做交易，資料在還沒對外宣布之前就已經看得見。交易一發生，他們沒辦法同時對外宣布同一件事，等於先結算了某件事，而這些資訊在全球狀態上人人都拿得到。別人就能拿這些資訊，對他們自己的客戶搶跑、開始賺錢。Institutions come with some of the major elements. If you compare them to general DeFi, they come with the size and the direction of where their next few orders are going. They also come with their counterparty. Where they are investing, where they are not investing, institutions come with proprietary research. They come to your blockchain, they do the transaction. The data is available before it is even publicly announced, right? The moment transaction happens, they cannot announce the same thing publicly, right? So they are settling something beforehand. That information is available to everybody on a global state. People can basically, you know, use that information to start front-running their own clients and start making money.

5:43所以我們談過的、據稱要放出數十億、數兆、甚至更大金額的機構，很多都還沒去敲 DeFi 的門。這不只、也不總是「最好的監管策略是什麼」，也總是我們怎麼看這個問題。有時候我們會這樣看：如果一個場所，例如結算，把你的資訊漏到公開，別人就能在全球帳本上用這份資訊去搶跑。我認為這對應到三種三難。我稱之為隱私三難，因為有三件事非常重要。And that's why a lot of the institutions that we talked to, who are supposed to open like billions and trillions and quadrillions of dollars, have not knocked on the doors of DeFi, right? Now, it's not only and always about what is the best regulation strategy. It's also and always about how we are looking at the problem. Sometimes we look at the problem in a way where if a venue, for example, a settlement, they leak your information to public. Somebody will be able to use that information on a global ledger to utilize and front-run themselves, right? Now, I say that this is responsible for three different kinds of trilemma. The reason why I call it a privacy trilemma is because there are three things which are very important.

6:36隱私三難：可組合、保密、擴充，機構三樣都要

6:36第一是可組合性。比方說韓國的新韓銀行要上鏈，它應該能跟同一條鏈上的其他區塊鏈或其他機構組合、一起做事。例如日本有 SBI，韓國有新韓，兩邊就算想保留自己的私有網路，也應該能互相合作。這就帶到另一面：機密性。談到機構，沒有人想犧牲隱私。機密性非常重要，每家機構都要把敏感資訊保得很機密。First is composability. For example, if let's just say Shinhan Bank of Korea is coming on chain. They should be able to be composable and work with some other blockchain or some other institution on the same blockchain, right? For example, you have SBI in Japan, and you have Shinhan in Korea. There should be some kind of composability. They should be able to work with each other even though they want to retain their private network, right? Which gives us the other part, which is confidentiality. Whereas privacy is a very important part. Nobody wants to sacrifice it if you are talking about an institution, right? Confidentiality is very important. Every institution wants to keep the sensitive information very confidential, right?

7:22應該只給自己、給監理機關、也只給稽核者。然後是第三件事：擴充性。問題在於，很多鏈承諾機密性和可組合性，擴充性卻大打折扣。如果你做出很好的解法，就應該能擴充。但結果你得接一堆其他服務、多付很多錢，這就不是可擴充的做法。擴充性不該排第三、也不該事後才想。跟機構談的時候，機構有很多錢，數十億、數兆、甚至更大。他們不會像你的私人 DeFi 夥伴，只要你三選二。Should be to themselves, should be to regulators, and should be only to the auditors. That is the third. And then scalability. Now, here is a problem where we have so many chains that promise confidentiality and promise composability, but scalability takes a major hit, right? If you come up with a really good solution, you should be able to scale it. But then you end up with so many other services where you have to pay a lot more money, right? Which is not a scalable way to do things. Scalability should not come as a third or an afterthought. Now, when we talk with institutions, institutions have a lot of money. Institutions have billions, trillions, and quadrillions of dollars. They will not, unlike your private DeFi partner, ask you to only, like, choose two out of the three.

8:18他們會說：「我們要走能同時提供這三樣的基礎設施。」可組合、機密、擴充。他們要一個解法一次給三樣。到目前為止，我們只能三選二，這就是區塊鏈最大的缺口。我再重複一次：這不是定律，也不是缺陷。這是當時最需要它的時候做成的設計選擇。同一枚硬幣的同一面，你讀一次，會說我們在任何一條全球狀態鏈上都有透明的表面，同一個問題就這樣被設計出來。They will ask you that, “We will go with the infrastructure that offers us all the three at the same time,” right? Composability, confidentiality, and scale. They want a solution to provide them all three. So far, we have to choose two out of the three. And that's where the major blockchain gap is. Now, again, I am going to repeat what I said. It is not a law. It is not a flaw. It is a design choice that happened when we needed it the most, right? It is the same side of the coin, where you read it once, where you are saying we had a transparent surface on any single global-state blockchain, where the same problem came as a design.

9:06我不是說這是 bug。我只是說，現在該談更好的解法，給機構他們要的東西。他們要同樣的擴充性、同樣的可組合性，但不能犧牲隱私，也不能犧牲資料的機密性。他們要確定，在交易還沒公開出現在鏈上之前，沒有人看得見。我認為這才是戰場。這一場我要大家把注意力放在這裡：請花一分鐘想，這些錢到底卡在哪。Now, I'm not saying this is a bug. I'm just saying that now it is the time to talk about more better solutions that provide the institutions what they want. They want the same scalability. They want the same composability. But not at the expense of privacy, not at the expense of confidentiality of their data, right? They want to make sure that nobody will be able to see their transaction before it is publicly available on-chain. And I think that's where the play is all about. And that's what I want you to all focus in my session is. I want you all to take a minute and think about where exactly all this money is getting stopped.

9:54選擇性揭露：驗證者不該看到不該看的

9:54為什麼數十億、甚至更大金額進不來？談到最適解，我們會談現在最好的解法是什麼、現在有什麼、我們能做什麼。我覺得每一條全球狀態鏈都該開始想選擇性揭露。我要說的一件事是：驗證者不該參與它不該在的交易。如果一筆交易只有兩、三方，那他們自己層級的驗證者說得通。但如果你把每一件事複製給每一個驗證者，做出一塊揭露面，風險就很大。該想更好的替代方案了，叫做選擇性揭露。Why are we not getting the billions and quadrillions of dollars, you know? Now, when we talk about what is the optimal solution, right, we talk about what is the best solution, what is available right now, and what we can do. I feel like every single global-state blockchain should start thinking about selective disclosure, right? The one thing which I say is a validator should not participate in a transaction where it should not belong, right? If two or three parties are in a single transaction, then the validators at their own particular level make sense. But if you are replicating every single thing to every single validator, making a disclosure surface, it comes with a major risk, and I think it is time to think about a better alternative, which is called as selective disclosure, right?

10:48這個例子裡有四方：Alice 付給 Bob，Bob 付給 Charlie，還有一個同步器，確保一切可組合。現在換成真正的機構。換成真實用例，你有資產管理人、保管人、還有抵押品場所。保管人兩邊都看得到：錢從資產管理人進來、錢往抵押品場所出去。抵押品場所不該看到錢是從資產管理人來的，資產管理人也不該看到錢進了哪個抵押品場所。同步器只負責讓整件事可組合，同時用加密保住隱私。In this particular example, you can see there are four different parties, right? Alice is paying Bob, Bob is paying Charlie, right? And there is a synchronizer, which is making sure that everything is composable. Now let's replace it with actual institution. When you replace it with actual use case, you have an asset manager, you have a custodian, and you have a collateral venue, right? Custodian is somebody who is able to see both the information, which is the money coming from the asset manager and the money going to the collateral venue. Collateral venue should not see the money is coming from an asset manager, and asset manager should not see the money going to which collateral venue. And the synchronizer is just making everything composable and private by having encryption at the same time, right?

11:41引入選擇性揭露，三個缺的元件才一起進來。只有保管人負責看到資料。資產管理人看不到場所，場所看不到資產管理人，兩邊都不能對自己的客戶搶跑。選擇性隱私把整件事保住。擴充也不該拿去換掉。你為了這個付很多錢、買昂貴解法，但那不是擴充。擴充要付很多錢，那就不是擴充，那是當OK繃用的加價服務。擴充性不該靠額外的加價服務才成立。So this is how you actually bring all the three missing elements by introducing selective disclosure. Only the custodian is responsible for seeing the data. Asset manager cannot see the venue. Venue cannot see the asset manager. Hence, both of them cannot front-run their own clients. Yet, selective privacy is what secures everything, right? Now, scaling should not be the thing that you should trade away, right? You pay a lot for an expensive solution that gives you this, but this is not about scalability. When you have to pay a lot for scalability, then it is not scalability. It is a premium service which works as a band-aid, right? Scalability should not come at a cost of an additional premium service.

12:29擴充性、機密性、可組合性要一起運作，任何一項都不能犧牲。這是區塊鏈上每個生態的基本需求。我不想讓你想成：我一定要建在以太坊？一定要建在別的鏈？一定要建在 Canton Network？我講過 Canton Network，也講過 DeFi 和以太坊當時在哪。我想讓你知道，時間在變。請為真正重要的事而建。機構正在上鏈。機構在找下一波隱私與擴充的革命，而且可組合性要內建。我們看到的例子裡，有同步器、全球同步器，資產管理人、保管人、抵押品場所，彼此看不到對方的敏感資訊。Scalability, confidentiality, and composability should all work together. Without sacrificing any element. And that is a basic need of every single ecosystem on a blockchain. I don't want you to be like, do I have to build on Ethereum? Do I have to build on any other blockchain? Do I have to build on Canton Network, right? I told you about Canton Network. I told you about where DeFi and Ethereum was, right? I want you to know that now the time is changing. I want you to build for what matters. Institutions are coming on chain. Institutions are looking for the next revolution of privacy and scalability, along with built-in composability, right? The example that we saw, where there is a synchronizer, a global synchronizer, and having asset managers, custodian, and a collateral venue, not able to see the sensitive information to each other.

13:24Canton 用選擇性揭露當機制，現場可以再問

13:24這才是把數百萬、數十億、甚至更大金額帶上鏈的路，也是唯一的路：給他們需要的那塊技術。選擇性揭露讓這套技術更完整，所以 Canton Network 用選擇性揭露當機制。如果這讓你覺得有道理、讓你好奇，會場裡可以來找我，我很樂意聊。我再次是 Canton Foundation 的 Jatin，謝謝大家來聽。有問題嗎？非常感謝。謝謝 Jatin 的精彩分享。This is the way, and this is the only way where we can introduce the millions and billions and quadrillions of dollars on chain, right? And this is the only way you can bring that amount of money on chain by giving them the piece of tech that they need, and hence, selective disclosure makes it much more perfect of a technology, and that's why we use selective disclosure as the mechanism on Canton Network. So if this is something that makes sense to you, that makes you curious, you can just catch me around the venue. You can talk to me, and I'm happy to talk to you. Again, I'm Jatin from Canton Foundation, and thank you so much for being here for my presentation. Any questions? Thank you so much. Thank you, Jatin, for your wonderful presentation.