Canton：一筆交易，為何只給部分驗證者看？
Private vs. Public Chains: The Key Decision Axes for Institutions
影片
影片與逐字稿來源：ETHTaipei
場次摘要
Martinet 主持的座談比較 Canton、IOTA 與公鏈在機構業務中的取捨。Canton 的 Teagan 說，交易可只向相關驗證者揭露，適合不願公開部位的金融機構；IOTA 的 Ko-Wei 則以貿易文件為例，說明只把雜湊上鏈的做法。兩者都須在隱私、資料可用性、節點成本及去中心化程度間選擇。
值得記下的 5 件事
- Canton 的示例只讓交易相關驗證者看到資料，隱私來自限制資料的分發範圍。
- IOTA 的貿易資料架構把文件留在鏈下，只把簽章與雜湊上鏈，收件者再核對內容和憑證。
- 雜湊能偵測文件遭竄改，卻不能找回已遺失的鏈下文件，儲存與復原仍須另外設計。
- 機構先問上鏈後能跟誰交易、降低哪些資金成本，生態中的具體用途是採用關鍵。
- Canton 開發者須自己管理驗證者、資料與服務可用性，這種控制權也增加營運責任。
逐字稿
開場：這一場改用英文，三條鏈自我介紹
好，大家午安。我們決定這一場用英文，因為現場有不少非台灣面孔。這樣也好，因為不知怎麼的，我英文比中文好。我也是。至少加密貨幣這塊是。對，術語太多了。開始之前，稍早 Oskar 講了他在以太坊機構隱私上的做法，現在我們要跟不同鏈討論，因為……All right, good afternoon, everyone. So we decided to make this an English session since we're seeing, like, quite a few, you know, non-Taiwanese faces here. So very glad to do that as well, because somehow my English is better than my Mandarin. Same. At least for crypto, I guess. Yes, exactly. Like, so much terminology. So I guess, like, to kick-start—well, before we start, like, earlier we had Oskar talking about, like, his solutions around, like, Ethereum privacy for institutions, and now we're, you know, discussing this with, like, different chains because...
我自己很偏以太坊。ETHTaipei 我主持四年了。但我也很喜歡從不同生態聽到的東西。他們可能試過不一樣的路，那是實驗。我們想看能學到什麼。所以很榮幸請各位來聊。正式開始前，先請各位短短自我介紹。好，很高興認識大家。我叫 Teagan。我在 Canton Foundation 負責策略合作，人也在台灣。Me, I am, you know, I'm very ETH-aligned and all that. Like, I've been hosting ETHTaipei for four years now. But what I really like is also, like, there's a lot of insights that are coming from different ecosystems. They may have tried different things. It's, you know, experiments. And we want to see what we can learn from those. And this is also the reason, so I had the honor of, you know, pleasure of having you guys over to have the chat. But before we really start on the panel, maybe we can do a short introduction of yourselves, and yeah, go from there. Awesome. Well, great to meet everyone. My name is Teagan. I lead strategic partnerships at Canton Foundation, also happen to be based out of Taiwan.
很高興在這裡。大家好，我是 Ko-Wei。我是台灣人，人在新加坡，每次回老家參加 ETHTaipei 都很好。我從 2017 年就認識 IOTA。2018 年起做 IOTA 華語社群。2023 年加入基金會。所以這條鏈的好、壞、醜，我都知道。好。很多人應該認識我，我是 Martinet。在這一側，你們知道我是 Quantstamp 的開發者關係負責人。我也參與過、近距離看過 L2 被做出來。因為我比較偏以太坊，所以會多從以太坊這邊評論。So great to be here. Hi, everyone, I'm Ko-Wei. So I'm Taiwanese, but I'm based in Singapore, but always great to be back to my hometown for ETHTaipei, and then I... I know IOTA since 2017. I built the IOTA Mandarin community since 2018. I joined the foundation since 2023. So, like, I know all the good, the bad, the ugly of our chain. Yeah. All right, and, well, a lot of you probably know me. I'm Martinet. On this side of things, so obviously you know me as Quantstamp head of developer relations. I was also associated with building and help—well, building and seeing close of building up the L2. And because I'm more ETH-aligned, so I'm gonna comment a little bit more on the ETH side, you know.
對，我代表以太坊跟 L2 這一側。就是這樣。既然碰到這題，也許先介紹鏈。我不覺得，我相信大家都聽過這些鏈，但不一定知道有意思的設計，或你們自己覺得特別值得講的。也許是專為機構設計，或者剛好很好用。先從鏈的介紹，以及它吸引人的地方開始。我來自 Canton Foundation，是 Canton Network 上面中立的治理層。網路有超過一千個驗證者，主要是機構在跑，每個月處理超過 9 兆美元的附買回。But yeah, that's the ETH side and L2 side. So that's me. And yeah, since we're touching on this, maybe we can do an introduction of the chain. I don't think— Yeah, I believe everyone has heard of the chains, but they don't necessarily know, like, the interesting designs and maybe something that you'd like to highlight about the chains that you consider very interesting. Maybe it's specifically designed for institutions, or, you know, it just happened to be very useful. I think let's start from that, the introduction of the chain and its appeal. So I'm from Canton Foundation, which is the neutral governance layer over Canton Network, which represents over a thousand different validators, namely institutions that are running, and processes over $9 trillion a month in repo.
Canton 每月九兆美元附買回，IOTA 走貿易資料
再講一點 Canton Foundation，也就是治理層。我覺得這很有意思，也是機構採用這麼多的原因：共同主席是 DTCC 和 Euroclear。DTCC 在美國的子公司 DTC 是中央證券存託機構，在美國保管大約超過 100 兆美元的證券。這就是我們網路要對準的規模。中央證券存託機構在傳統金融市場負責證券保管，也負責結算與所有權移轉。Now, to talk a little bit about Canton Foundation, which is the governance layer, which I think is very interesting and why we have so much institutional adoption, is we're actually co-chaired by DTCC and Euroclear. DTCC, their subsidiary in the U.S., DTC, is the central securities depository— in the U.S. they custody over $100 trillion or so in securities. So that's the scope of what we're targeting at the network. And a CSD, they essentially handle custody of securities in the traditional finance market, but also the settlement and transfer of ownership rights of these securities.
有意思的是，我們怎麼跟 DTC 及其會員客戶合作，買方與賣方，像 BNY、J.P. Morgan、Goldman 這些，把擔保品、擔保品流動、證券借貸、附買回這整塊市場做得更有效率。尤其我們想走向 T+0 的金融狀態。對，我在 IOTA Foundation。IOTA 是很老的一層。我們 2015 年公平發行，2017 年主網上線，沒拿過一分 VC 的錢。然後我們是第一個在歐盟柏林登記的 Layer 1。So what's been very interesting is how we work with the DTC and its member clients, the customers on the buy side and sell side, like your BNYs, your J.P. Morgans, like your Goldmans, etc. to help bring better efficiencies in collateral and collateral mobility, securities lending, and repo, and this whole market. Especially as we try to aim towards being a T plus zero financial state. Yeah, so I work for IOTA Foundation. IOTA is a very old layer one. We did a fair launch back in 2015. We did our mainnet launch in 2017, and we never got a single cent from VC. And then we are the first Layer 1 registered in Berlin in the European Union.
所以 2017 年 IOTA Foundation 成立。2022 年有瑞士基金會。2023 年 IOTA 是阿聯酋阿布達比第一個受監理的 Layer 1 基金會。我們一直相信，要跟政府合作，必須是中立、可信的基金會。IOTA 一直強調兩個信念。第一，我們相信加密貨幣不只區塊鏈。我們是這個領域最老的 DAG，以前叫 Tangle。早期很 cypherpunk。2015 年就做三進位、無手續費交易、後量子密碼，那時很狂。2021 年做了比較現代的設計，清技術債。So back in 2017, IOTA Foundation was set. And then we got, like, 2022, we have the Swiss Foundation. 2023, IOTA is the first regulated Layer 1 foundation in Abu Dhabi in the UAE. So we always believe, like, we need to be a neutral and credible, like... foundation to work with government. So IOTA, we always stress two beliefs. The first is we believe crypto goes beyond blockchain. So we are the oldest DAG in the space. It used to be called Tangle. In the early days, we were very cypherpunk. So we did, like, ternary, feeless transactions, PQC in 2015. That was very crazy. So we did a more modern design in 2021 to clean our technical debt.
但我們仍有代付交易，因為我們知道機構應用不能叫他們直接付 gas。也有帳戶抽象，人們可以用 passkey 或自己想用的方式簽名。第二根柱子是……我們相信加密貨幣要大規模採用，得靠政府與機構合作。聽我們的名字，大概還記得早期跟 Jaguar、跟台北市政府做物聯網移動相關的事。但 2020 年之後，我們比較聚焦貿易物流與貿易融資。所以 IOTA 現在跟 TradeMark Africa、世界經濟論壇、Tony Blair Institute 合作，想把整個非洲、英國、歐盟的貿易資料上鏈。But we still have the sponsored transaction because we know for institutional dApps, you can't ask them to pay the gas fee directly. And we also have, like, account abstraction, so people can use passkey or however they want to sign. And the second pillar we always have is, like... We believe crypto will go mass adoption through government-institution collaboration. So, given our name, you probably remember in our early days we were working with, like, Jaguar or, like, Taipei City government for all the IoT-related mobility stuff. But then after 2020, we more focused on the trade logistics and trade finance. So basically, IOTA, we now work with TradeMark Africa. World Economic Forum, Tony Blair Institute, to really try to bring the whole Africa and bring the whole UK and bring the whole EU trade data on-chain.
IOTA 從 Tangle 換成 Move VM
因為比起 DeFi 和 VC，一碰到人的資料就有很多 GDPR 之類。但若處理的是貨物與貿易，比較接近。所以……IOTA 長話短說：我們想當全球貿易的信任層。我其實想問幾乎同一題，但因為這是 ETHTaipei，今天有點混。它核心曾是開發者會議、研究會議。所以我得以宅的身分問：這條鏈技術上的創新是什麼？技術上什麼會讓人興奮？Because compared to, like, DeFi and VC, if you involve, like, human data, there's a lot of, like, GDPR or whatever. But if you try to deal with the goods and the trade, that will be more approachable. So... That's a long story short for IOTA. We try to bring, we try to be the trust layer of global trade. I would actually want to ask almost exactly the same question, but because it's ETHTaipei, ETHTaipei is, like, more, you know, today it's kind of like a mix. At its core, it was a developer, you know, conference, research conference. So I have to ask, as a nerd, like, what is the innovation of the chain on the technical side? Like, on the technical side, what will be, like, exciting?
因為那件事，某種意義上我中立。什麼技術讓我興奮，我就興奮。我不在乎它來自哪個國家或生態。我是技術宅。請給我 chain-pill，我不知道這該怎麼叫。好，我先講。講到 IOTA 的 L1 或生態，我們有林迪效應，在這裡很久了。但技術上我們一直在改，因為我們知道資訊科學一直在演。當年 2015 年，Tangle 當第一個 DAG，想法很精彩。Because that's something, you know, like, I'm kind of, like, in that sense, neutral. Whatever excites me as technology excites me, right? I don't care if it's from specific nation or ecosystem, whatever. I'm a techno nerd. Please, you know, chain-pill me, I don't know how it's called, but yeah. Okay, so let me go first. So when we come to, like, IOTA's L1 or our ecosystem, we have a Lindy effect. We have been here for very long. But in terms of tech, we always keep improving because, like, we know, like, information science is keep evolving. So back in the day, 2015, Tangle was an amazing idea as the first DAG.
但我們知道現在很多人也用 DAG，像 Avalanche、Fantom、Kaspa，甚至 Sui。所以。當年我們確實有機會把 IOTA 2.0 放出來，那是我們自己的 Tangle。最後我們決定，如果 Mysten Labs 已經做出好很多的基礎建設，那就向他們學。所以我們把 UTXO DAG 升級成以 Move VM 為基礎的 DAG。因為你也知道，UTXO 不好做智慧合約，而生態與合作需要智慧合約。如果你只是比特幣或 Zcash 那種價值儲存，也許不需要智慧合約。但我們真的要有實際採用，就需要智慧合約。But we know a lot of people, they also use DAG now, like Avalanche, Fantom, Kaspa, and even Sui. So. Back in the day, we did have a chance to really release our IOTA 2.0, so that's our own Tangle. But in the end, we decided if someone else, like Mysten Labs, already built a much better infrastructure, then we learn from them. So basically, we upgraded our UTXO DAG to a Move VM-based DAG. So because, like, as you know, UTXO is hard to implement smart contracts, and we know smart contracts are necessary for ecosystem, for partnership. If you are just a store of value like Bitcoin or Zcash, maybe you don't need smart contracts. But given we really want to have real adoption, smart contracts are needed.
所以 VM 我們只有一次機會選，最後選了當時最先進的設計，現在仍這麼認為。Move VM，或物件 Move，等於把帳戶模型跟 UTXO 最好的部分混在一起。如果你的物件單純，例如我轉東西給 Martinet，它不必為了排序進共識，因為沒有真正牽連。可以有比較鬆的因果。但如果我們都要跟同一個 Uniswap V2 風格的 DEX 交易，就需要全序。這叫共享物件。所以在 IOTA Foundation，我們相信從 Facebook Libra／Diem 到 Mysten Labs 這條路很精彩。So we have one shot to decide which VM we use, and in the end we choose the most advanced design at the moment, and we still believe that's the case. Move VM, or object Move, is like you basically took the best part of account-based and UTXO-based and you mix together. If your object is simple, you do a transfer, for example, I transfer something to Martinet. It doesn't need consensus to order the transaction because it doesn't really involve in. You can have a loose, like, causality over there. But for example, if we both need to trade to the same Uniswap V2-style DEX, then it needs a total ordering. Then this thing is called a shared object. So in IOTA Foundation, we believe, like, the thing from Facebook Libra/Diem to Mysten Labs is amazing.
Canton：只有對手方參與共識、看到交易
所以我們向他們學……我們借用他們的 VM，仍保留自己的共識。我們相信機構、政府在意韌性勝過延遲。如果你很 degen，想把迷因幣交易做得極快，對，延遲該是 0.2 毫秒而不是 0.4。但對我們來說，最壞情況應該更好，不是只改善最好情況。所以我們的延遲不是頂尖，但最壞情況某種意義上比競爭對手好很多。長話短說：我們一直在演進一層，這在區塊鏈圈很怪。我同意，因為聽起來像不可變，然後你永遠有技術債。So we learn from them to really... Like, we borrow their VM and we still keep our own consensus. We believe institutions, governments care more about resilience than latency. If you are very degen, you want to be very fast about your meme coin trading, yes, your latency should be 0.2 milliseconds instead of 0.4. But for us, we believe the worst case should be better, not just like you only improve the best-case scenario. So for us, our latency is not, like, state of the art, but our worst case is much better than our competitor in some sense. So that's a long story short. Like, we keep evolving our layer one, which is very weird in blockchain space. I do agree because, like, it sounds like immutable, and then, like, you always have technical debt.
但我們已經大概升級兩次。主網 2017 年上線，2021 年清掉 2017 的技術債，2025 年選定 VM。就是現在這樣。那是 EVM？現在是 Move VM，跟 Sui、Aptos 一樣是 Move。對。很厲害。所以，Teagan？對。其實延續 Move 的 UTXO 底，Canton 的共識語言從那裡拿了很多靈感。但我們的共識叫利益關係人證明。我相信 Martinet 能體會我們想做什麼：讓機構能在鏈上做交易，同時有隱私。But we already kind of, like, upgraded twice. So our mainnet launched in 2017, we cleaned our 2017 tech debt in 2021, and we chose our VM in 2025. That's what we have. And that was the EVM? Now we are the Move VM, which is the same as Sui and Aptos, which is Move. Yeah. Amazing, amazing. So, Teagan? Yeah. So actually, going off the UTXO base of Move, Canton as a consensus language takes a lot of inspiration from that. But what our consensus is called is proof of stakeholder. I'm sure Martinet can appreciate what we're trying to build here, and that is to allow institutions the ability to do transactions on-chain while having privacy.
那是我們最大的焦點。讓機構隱私成立。怎麼做到？我們說，當我發起一筆交易，只有這筆交易牽涉的對手方會看到、會參與共識。例如我把 USDC 轉給你，我是銀行，你有 MetaMask 錢包，資產是 USDC。第一個進來的是 Circle 的驗證者：嘿，我轉出去的這筆 USDC 確實是我的密碼學簽章。第二個是銀行這一側，你可以想成 EOA；我們叫 party ID，也就是託管那筆 USDC、正在送出的那個 party ID。That's our biggest kind of focus area. It's enabling institutional privacy. How do we achieve that? What we say is that when I originate a transaction, Only the counterparties that are involved in that transaction actually will see and participate in consensus. So, for instance, if I'm sending USDC to you, and I'm a bank, and you have your MetaMask wallet, right, and obviously, the asset being transferred is USDC. So the first person being involved in the transaction is Circle's validator saying, hey, this USDC that I'm sending over is indeed my cryptographic signature. A second would be from the bank side, whichever—you can think of it, EOA; we call it party ID—whichever party ID that hosts that USDC that is making that send.
用 HSM 在鏈上簽核，如果不同驗證者有不同政策引擎，他們也會參與共識。當然你這一側用 MetaMask 簽名：對，我要收這筆。然後在 Canton 上會做成一筆原子交易。要注意的限制是：儘管網路上有超過一千個驗證者，這個例子裡實際上只有五個驗證者看到那筆交易。其餘的，資料連碰都沒碰到他們的驗證者。這對金融機構很關鍵。第一，假設在台灣，本地法規說碰到這筆交易資料的驗證者、節點只能在自建機房，還要滿足 GDPR 等合規，我就可以說，在 Canton 上交易符合這項義務。Do the HSM sign-off on this on-chain, and if there are different validators that have different policy engines around that. So they would also participate in consensus, and of course, your side, you would, using your MetaMask wallet, sign off like, yes, I wanted to receive this. And then that will create one atomic transaction on Canton. The caveat, and the important thing to notice about this in this example transaction flow, is despite us having over a thousand validators on the network, only, in this example, five validators actually saw that transaction. The rest, that data did not even touch their validator, which is crucial for financial institutions because, one, let's say, in Taiwan, if I am under the local regulation and it says the validators, the nodes touching this transaction data, can only be on-prem and, you know, have to satisfy GDPR, et cetera, compliance, I can actually say, yes, transacting on Canton fulfills this obligation.
機構採用：穩定幣、貿易資料先上雜湊
如果我在韓國，韓國金融委員會說 STO 發行必須用私有區塊鏈，這套共識也滿足。所以我們能跨法域工作。所以我們能做日本到美國的跨鏈附買回，日本公債對美國公債，也是我們想把台灣與更廣的亞太帶進 Canton 網路的方式。很好。我有一個技術的、宅的問題想馬上問，但我先忍住，也許觀眾會問。我從以太坊這邊，一直理解交易、最終性這些。If I'm in Korea, and the FSC in Korea says, well, it has to be a private blockchain to do STO issuance, this consensus satisfies all of that. And that's why we work across different jurisdictions. That's why we've been able to do repo cross-chain from Japan to U.S. between JGBs and, you know, U.S. Treasuries, and that is how we're aiming to bring Taiwan and broader APAC onto the Canton network. Awesome. I, I have a technical, like... you know, like, nerdy question right after, but I'm going to hold it, and maybe, like, someone in the audience will have it. I'm, like, from the ETH side, like, I've always understood, like, you know, transactions, like finality, et cetera.
我習慣交易是最終的，也習慣有一堆驗證者在看。你剛才說只有五個，對吧？IOTA 在這件事上模型也很不一樣。所以我有一堆問題。但我也想講採用這一側，因為你們提到不同種類的採用。以太坊我覺得最清楚的例子，就是鏈上放了多少 USDC。實際數字我忘了，但當時是 Solana 的十倍。而 Solana 是非 EVM 裡排第二的。所以那時覺得，喔，這很大。我今天又查了。I'm used to transactions, like, being final, and also, like, the concept of, like, we have a bunch of, like, validators looking at it, and in your case, you mentioned, like, only five, right? And I think IOTA has, like, a very different model on that as well. So, you know, I have, like, tons of questions there. But I do want to get to also, like, the adoption side of things, because you guys mentioned, like, there's, like, different kinds of adoption. Like, for Ethereum, I think it's, I think the clear example would be, like, just the amount of USDC sitting on the chain. I forgot the actual number, but it was 10 times Solana. And Solana was, like, the non-EVM, you know, second one in there. So that was like, oh, okay, that was pretty big. I looked it up today.
我記性很差，仍只記得是十倍。十倍還是很好。但那就是……至少穩定幣這一側，以太坊在做的事。某種程度上你可以說那就是機構採用。但我想問，你們也跟不同專案合作。你們在機構採用這一側看到什麼？對我們來說，多半從貿易物流開始，因為供應鏈追蹤若不牽涉金融資產或金融機構，起步容易很多。所以我們從肯亞再到英國。基本上是幫單一貿易窗口把資料雜湊上鏈，然後你有信任，有授權手段，像。I have a very bad memory, and I still only remember it was 10x. Still, 10x is pretty good. But I was, so that's kind of like, you know. At least on the stablecoin side, that's what Ethereum has been doing. In a way, you can argue that is institutional adoption. But I wonder, like, you guys also work with, like, different projects. Like, what are you guys, what do you guys see in the institutional adoption side for you? So for us, we mainly start with the trade logistics, because if you do supply chain tracking involving no financial asset or financial institution, it will be much easier to start with. So we start with Kenya and then UK. So basically, you help the Single Trade Window to put their data hash on chain, and then you have a trust, like, you have an authorization means, like.
公司可以簽名，把這筆資料上鏈，然後公司的憑證由例如肯亞的單一貿易窗口來簽。所以我們先處理貿易資料，過去幾年就是這樣做。我們會把這件事帶到整個非洲，有個東西叫 AfCFTA，非洲大陸自由貿易區，我們跟 AfCFTA 有一個叫 ADAPT 的計畫。未來十年，到 2035 年，我們會先把超過 50 個或超過 30 個非洲國家的貿易資料上鏈。但我們知道，真正有賺頭的肯定不是資料生意，是金融生意。所以下一步是什麼？The company can sign, like, they put this data on chain, and then the company's credential is signed by, for example, Kenya's Single Trade Window. So we deal with the trade data first, and then that's how we do in the past few years. We will bring this thing to the whole of Africa, so there's something called— AfCFTA, basically the whole African Continental FTA, and we work with AfCFTA to have a plan called ADAPT. In the next 10 years, till 2035, we will bring, like, more than 50 or more than 30 African countries' trade data on-chain first. But we know, like, the lucrative business is definitely not the data business; it's the financial business. So what's next step?
IOTA 明文不上鏈；下一步是貿易融資
就像現在貿易資料已經上鏈，而且很中立。信任的不只是我們，還有 Tony Blair Institute、TradeMark Africa、世界經濟論壇。下一步呢？肯亞的咖啡農或酪梨農，文件已經簽好，雜湊上了鏈，可以驗證。但貨已經在奈羅比倉庫。新加坡的買方、英國的買方，不想在酪梨或咖啡還在奈羅比時付款。所以有人可以預先付款，因為這筆貿易資料在鏈上、可驗證，而且是政府簽發。我們今年就是這樣開始做貿易融資。Martinet 也提過，衡量鏈使用最重要的指標之一是穩定幣發行，或者說 TVL 與量。Just like now we have the trade data on-chain, and this is very neutral. It's, like, trusted by not just us, but also Tony Blair Institute, TradeMark Africa, World Economic Forum. So what's next step? So some, like a coffee farmer or avocado farmer in Kenya, they already get their documents signed, and then they put a hash on-chain. This can be verified. But this thing is already in the warehouse in Nairobi. But the buyer in Singapore, the buyer in the UK, they don't want to pay when the avocado or coffee is still in Nairobi. So someone can pay in advance, given this trade data is on-chain, can be verified, and also issued by the government. So that's how we start with the trade finance this year. And as Martinet already mentioned, one of the most important indicators for the chain usage is the stablecoin issuance, or you can say TVL and the volume.
問題是我們需要接上一個主要穩定幣。所以 Tether 這個月底應該會上 IOTA，談了很久。我們是非 EVM。這給我們很多……我們知道 EVM 很好，但技術債很多，所以選 Move。但這也讓跟所有基礎建設整合難很多，不管是 Tether 這種穩定幣發行方，還是 Zodia 或 BitGo 這種基礎建設。非 EVM 永遠讓人頭疼，但我們覺得值得。對，順著這題，我很想知道 IOTA 怎麼處理那些貿易資料的隱私，因為，對，我們不上貿易資料本身，因為那永遠有隱私疑慮。The problem is, like, we need to onboard one major stablecoin. So Tether should be on IOTA by the end of this month after a very long conversation and so on and so forth. We are non-EVM. It gives us a lot of, like... We know EVM is great, but there's a lot of technical debt, so we choose Move. But it also makes our life much harder to integrate with all the infra, no matter you say stablecoin provider like Tether or infrastructure like Zodia or BitGo. Non-EVM always leads to a headache, but we think it's worth it. Yeah, maybe based off that, I'd be very interested to know how IOTA handles the privacy for that trade data, because— Yeah, so we don't put the trade data itself on-chain because we know that's always a privacy concern.
所以我們一律上雜湊。基本上，如果我們跟肯亞合作，奈羅比某個政府官員……他們的 AWS 也許有一把金鑰來簽，這件事提供信任錨或框架。例如我是咖啡農，我可以用金鑰簽名：鏈上這個雜湊是我公司簽的。所以這個信任，到最後……所有資料本身都在鏈下。你只能看到有人簽了某樣東西，紀錄，也就是 PDF 的雜湊，在鏈上。最後如果倫敦有人收到這份 PDF，只要核對雜湊是否符合 PDF 內容，再看這個雜湊是肯亞一位農夫簽的，而那位肯亞咖啡農的憑證，是奈羅比、肯亞單一貿易窗口簽的。So we always put the hash. So basically, if we are, like, we work with Kenya, so some government officer... in Nairobi, their AWS maybe have a key to sign, or whatever, and then this thing provides a trust anchor or a framework. So, for example, I'm a coffee farmer. I can use the key to sign, like this hash on-chain is signed by my company. So this trust, at the end... All the data itself is off-chain. Only you can see someone signed something, and the record, like the hash of the PDF, is on-chain. So at the end, if someone in London, they receive this PDF, they just check whether the hash matches the PDF content, and then they see this hash is signed by a coffee— a farmer in Kenya, and the coffee farmer in Kenya, the credential is signed by the Nairobi, like, a single trade window in Kenya.
就是這樣運作。我們也不把任何資料本身上鏈，因為區塊鏈本身不可變、又透明。我們就這樣做。我能不能問，雜湊存在哪？PDF 呢？就在那裡嗎？對，PDF 現在是鏈下，完全鏈下。我們也知道很多法域要求，例如你的資料不能出境之類。IOTA 差不多有幾十個、接近一百個驗證者。驗證者遍布世界，所以……如果你把明文上鏈，肯定無法滿足某些法域的要求。只上雜湊就做得到。基本上我們就是這樣做。That's how the thing works. So we also don't put any data itself on-chain, because blockchain itself is immutable and it's transparent. So that's how we do it. And can I ask, for the hash, like, where would that be stored? The PDF? Is it just in there? Yeah, so the PDF for now is off-chain, just totally off-chain. Like, we also know a lot of the jurisdictions, they ask, like, your data cannot, for example, leave the country or so on and so forth. So given IOTA, we have close to, like, dozens, or we are close to a hundred validators. We have validators across the world, so... If you just put this plaintext on-chain, it definitely just cannot fulfill the requirement of some jurisdictions. So if you just put a hash, then that's doable. So basically, that's how we do it. Yeah.
DTCC 把美國公債測試交易上 Canton
Canton 採用這一側呢？對。我問那些問題的原因是，例如 7 月 15 日，DTCC 正式做了把美國公債上鏈的測試交易。他們選 Canton 的考量之一，是必須真的能把資料帶上鏈，中央證券存託的資料、所有權紀錄的證明與紀錄。誰擁有這些不同資產？你必須帶上鏈。為什麼必須上鏈？因為假設我只在鏈上指一個雜湊，對應的 PDF 或貿易資料在鏈下。DTCC 資料庫裡會有一份。How about Canton on the side of things, on the adoption side? Yeah, yeah. So, so the reason I asked those questions was because, for example, on July 15th, we had DTCC formally do their test trades of bringing U.S. Treasuries on-chain on the network. And as a part of their considerations of choosing Canton, they had to make sure that you can actually bring data, the CSD data, the proofs and records of ownership records. Who owns, you know, these different assets? You have to bring that on-chain. Why do you have to bring that on-chain? Because let's say, for example, I'm just pointing to a hash on-chain, and the corresponding PDF or trade data off-chain that matches to that hash. DTCC would have one in their database.
台灣有人自己那份也有一份。兩邊對不上怎麼辦？鏈下紀錄走失怎麼辦？因為鏈下資料對不上，對誰擁有什麼有爭執。你就得對帳，那就是傳統金融一樣的痛。所以才有那些後台作業。這就是我們為了把 DTCC 帶上鏈必須解決的。所以他們選我們當第一條進去的區塊鏈。也是為什麼 Broadridge、Tradeweb 這些金融市場基礎建設提供者，他們處理的成交後結算是兆級交易量。他們放心把買方客戶帶上鏈，那些客戶代表 J.P.、Goldman 這個世界，因為 J.P.Someone in Taiwan will have one in theirs. But what happens when there is a mismatch between that? What happens when your off-chain records go astray? There's a disagreement around who owns what because of our off-chain data not matching. You have to go through reconciliation, and that's the same pain points we have in traditional finance. That's why we have all of the back-office operations in place. And this is what we had to solve to bring DTCC on-chain. That's why they chose us as the first blockchain to be a part of. And that's also why Broadridge, Tradeweb, these financial market infrastructure providers, they handle post-trade settlements in the trillions in trading volume. They are comfortable bringing their buy-side clients, where they represent the J.P.s, Goldmans of the world, bringing them on-chain because J.P.
Morgan 知道，我在 Canton 做附買回時，競爭對手看不到。對這些買方公司，不管是 Citadel Securities 還是 DRW，這非常重要。我有一個問題。所以你，我沒完全懂你說你有完整的，你在問他，那邊有一個交易雜湊，兩方各自拿著文件，然後對不上。那一段我沒懂，因為檔案不同，雜湊就不同。基本上不太可能。如果你用的是密碼學上安全的雜湊，以我的理解，機率上不該發生。Morgan knows when I do a repo on Canton, my competitor will not see that. And for these buy-side firms, be it Citadel Securities, be it DRW, this is very important to them. Yeah, I have a question on that. So you— I didn't quite understand the part where you said you have the full—well, you were asking him, like, there's a transaction hash there, and there's, like, two parties holding the docs, and then they have a mismatch. That part I didn't understand, because if the file is different, then the hash is different, basically. Like, it's very unlikely on the— If you're using a cryptographically, you know, secure hash, then my understanding, well, that shouldn't be possible, like, probability-wise.
所以我一定誤解了。我想問你，你確切是什麼意思？對，假設我有貿易資料，是一份 PDF。我大概有自己的資料儲存放那份 PDF。雜湊是用來指向那份 PDF，但兩邊最後會各自存在自己的資料庫。那如果我的資料庫被駭或壞掉，PDF 丟了，或紀錄被改了呢？對。如果被改了，交易雜湊會不一樣，對吧？所以，一邊，假設兩方。A 方被改了、被駭了、壞掉了，怎樣都好。So I must have misunderstood that part. I want to ask you, like, what do you mean by that exactly? Yeah, so let's say I have the trade data, and it's a PDF, right? I probably have my own data storage that stores that PDF. The hash was created to reference that PDF, but then both sides would end up storing it in their own database. So what if my database got hacked or got corrupted and I lost that PDF, or something around the records got altered? Right. So if it got altered, then the transaction hash would be different, right? And so, one side, let's say there's two parties. Party A altered, got hacked, altered, crashed, whatever.
鏈下檔案遺失怎麼辦，資料成本差在哪
算出來的交易雜湊對這個人應該不一樣。所以我們知道他的資料是偽造的。我覺得問題是資料遺失，對吧？被改了我們知道。但問題是，例如某人的資料就消失了。這怎麼解？目前我們的情況，資料存在主權實體自己的資料庫，他們有完整所有權。未來永遠可以探索 IPFS、Arweave，或其他去中心化儲存協定，像 Walrus。所以……我們也很感謝 Mysten Labs 做了很多酷的東西，可以學。但至少此刻，資料本身在鏈下，只有雜湊在鏈上。The computed transaction hash should be different for this guy. So we know his data is forged. I think the problem is, like, data loss, right? Just, like, if it's altered, we know. But the problem is, like, for example, someone's data just disappears. How to solve this problem? So, for now, like in our case, the data is stored with the sovereign entity's database, so they have full ownership. And in the future, you can always, like, explore, like, IPFS or Arweave or some other decentralized storage protocol like Walrus. So, like... We also appreciate Mysten Labs building a lot of cool stuff, so we can learn from. But at least at the moment, the data itself is off-chain, and only the hash is on-chain.
我們這樣才放心，合作夥伴與其他利害關係人也這樣才放心。那我完全同意。對不上比較容易偵測，但若沒了，就是沒了。Canton 的情況，你基本上可以相對便宜地帶上去。理論上所有區塊鏈都能存，只是很貴，對吧？所以後續問題：成本，依你的理解是多少？對，因為……我們做共識的方式，你可以有自己的資料庫綁在驗證者上，因為你在網路上跑驗證者。That's how we feel comfortable and how all our partners and all other stakeholders feel comfortable. Yeah, that I totally agree with. Like, mismatch, it's easier to detect that, but, like, if it's gone, then it's gone. In the Canton's case, like, you can basically bring it on in relatively cheap. Actually, in theory, all the blockchains, you can store it. It's just very expensive, right? And so maybe that's a follow-up question. What's the cost, like, from your understanding? Yeah, so for... Due to the way we do consensus, you can basically have your own, like, database tied to your validator because you run a validator on the network.
所以成本就是實際交易資料的負載。在我資料庫那份很大的 PDF 裡，我要送給你驗證的是哪一塊？可能只是那份更大資料裡的一小部分。對，技術上我想問的太多了，但我們該繼續。我會說，我們的情況簡單很多，因為以 Move 為例，Aptos 用 HotStuff 共識。Sui 用 Mysticeti，我們用一個叫 Starfish 的。之所以叫 Starfish，是因為海星不只一個頭，很難砍死。And so the cost is just what is the actual transaction data payload. So within this large PDF that I have, you know, in my database, what am I sending over to you to do verification on? It could be just a small part of that larger data that I'm bringing. Yeah, I have so many questions I want to ask on the technical side, but we should continue. But I would say, for our case, it's much easier because we just, for example, for the Move, like Aptos uses HotStuff consensus. And then, like, Sui uses Mysticeti, and we use something called Starfish. And because it's, well, the reason we call it Starfish is, like, starfish has more than one head, so it's very hard to chop and kill the starfish.
比較有韌性，如果你控制三分之二的質押，基本上就能寫任何東西。長話短說，這個意義上我們是典型區塊鏈。對。下一個問題：跟機構合作時，不管是民族國家、政府、金融機構，或其他傳統企業，你觀察到他們真正在意什麼？你們已經提了一些，像這個政府不准這類資料上鏈。所以，對。你必須是中立的非營利組織。It's more resilient, and if you control two-thirds of the stake, then you can write basically anything. So, long story short, we are a typical blockchain in this sense. Yeah. So the next question I want to go to is, like, when you're working with institutions, be it nation-states, be it government, be it, you know, financial institutions, or, like, you know, some other traditional businesses, right, what are the concerns that you have observed that they do really care about? You already mentioned some, like, hey, this government doesn't allow any of this kind of data to be on-chain, etc. So, yeah. So you need to be a neutral nonprofit organization.
中立基金會、隱私，以及壽險個資上鏈
所以 IOTA 或 TWIN，也就是全球貿易資訊網路，不是第一條想解國際貿易供應鏈的區塊鏈。你可能聽過 IBM 跟海關合作。你可能聽過 IBM 跟 Maersk 做一條許可鏈叫 TradeLens。所以我們不是第一個想把貿易上鏈的。但問題是，如果你跟這個領域的某一大玩家合作，所有競爭者都不會進來，然後。如果你沒真正解大規模協調問題，為什麼不直接用 AWS？為什麼不直接用 SQL？保持中立、保持非營利很重要，尤其你必須在岸。你不能說你是某個很怪的開曼群島之類。So IOTA or TWIN, the Trade Worldwide Information Network, is not the first blockchain. Trying to solve this, like, international trade supply chain issue. So you might have heard IBM work with customs before. You might have heard IBM work with Maersk to do some permissioned chain called TradeLens. So we are not the first one trying to bring the trade on-chain. But the problem is, like, if you work with one of the major players in the space, all the competitors will not join, and then. Like, if you don't really solve a large-scale coordination problem, why not just use AWS? Why not just use SQL? So staying neutral and staying nonprofit is very important, especially, like, you need to be onshore. You cannot say you are some very weird Cayman Island or whatever thing.
你必須是正當的在岸基金會。我們這樣取得信任，因為在這個領域十年了。所以我會說，IOTA 的技術一直在演，但林迪效應來自品牌，來自過去十年的累積。然後這不只是我們的。這不只是我們的東西。所以。TWIN，全球貿易資訊網路，IOTA 只是董事之一。所以是 IOTA Foundation，然後世界經濟論壇、TradeMark Africa、Tony Blair Institute 等等。我想把這些大名都拉進來。基本上 TWIN 是一套軟體，程式是 IOTA Foundation 寫的，但背後不只是我們。也有其他大名在背書。You need to be a proper onshore foundation. So that's how we get trust, because we have been in the space for like 10 years. So I would say for IOTA, our tech keeps evolving, but the Lindy effect is from the brand and from the legacy we have for the past 10 years. And then it's just not ours. It's just not ours. It's just not our stuff. So. For TWIN, the Trade Worldwide Information Network, IOTA is only one of the board members. So IOTA Foundation, and then World Economic Forum, TradeMark Africa, Tony Blair Institute, so on and so forth. So I want to get all these big names on board with us. So basically, TWIN is a piece of software, and the code is written by the IOTA Foundation, but— it's not just backed by us. It's also backed by all the other big names, yeah.
所以答案有點打太極，但我整天都在講：就是隱私。再講深一點，例如我們有 Zinnia，美國的壽險提供者，他們是壽險管理平台，用區塊鏈協調壽險與年金，尤其在美國，你把壽險從一家轉到另一家時。有很多資料紀錄要結算，非常私密的資料紀錄，可能包含你的社會安全號碼。可能要 30 天、60 天，可能要六個月。我們其實把人們的壽險、健康資訊、社會安全號碼放在 Canton 鏈上。So, a bit of a cop-out answer, but what I've been talking about all day: it's privacy. And to go a little bit deeper into that, for instance, we've had Zinnia, which is like a life insurance provider in the U.S., where they're a life insurance admin platform, and they utilize the blockchain to do coordination of life insurance and annuities, especially in the U.S., when you transfer your life insurance from one provider to another. There's a lot of data records that need to be settled, you know, very private data records that could include your SSN and so on. And that could take 30 days, 60 days, it could take six months. We actually have people's life insurance, you know, health info, their social security numbers on-chain on Canton.
他們對我們網路的信任到這個程度。我覺得這很能說明我們在夥伴信任上做到什麼。所以快速整理：你說政治這一側，比較是基金會怎麼設。那是你的點。你的點是隱私，但我還……記得稍早另一點。我記得的另一個是要能把資料託在鏈上。還有別的點嗎？好奇。我問是因為我顯然是以太坊派來的間諜。我想學。所以，當以太坊的人，我們確實知道。That's the degree that they actually trust our network with. And I think that speaks volumes in terms of what we've been able to achieve in terms of partner trust. So, kind of like, quick summary: like you're saying, like on the political side, more foundation setup. That's your point. Your point is privacy, but then I also... I recall the earlier point. The other one that I remember was to be able to host the data on-chain there. Is there some other points? Just curious. The reason I ask is I'm a spy from Ethereum, obviously. I want to learn. So, as Ethereum folks, we do know.
機構真正問的是能跟誰交易
隱私很重要，對吧？這不是新資訊，我想要有意思的。資料要完全上鏈，當需求來說有點意思。顯然，他們以為資料在鏈上是什麼意思，以及實際上怎麼證明，那是想像的問題，也許再來才是技術問題：技術上怎樣算上鏈。那裡大概有一整場辯論。那其實也是我本來想追問、想宅下去的。但除了隱私、除了資料上鏈，跟他們討論時你還觀察到其他需求嗎？Privacy is important, right? So that's not new information, and I kind of want something that is interesting. Data to be completely on-chain is kind of interesting as a requirement. Obviously, like, how they think the data is on-chain and how to actually prove that, that is another question of their imagination, and maybe second is the question of tech, like, technically, what does it count to be on-chain, et cetera, right? There's probably a whole lot of debate right there. That was actually part of, like, the follow-up question I wanted to ask, like I want to nerd out. But have you observed other, like, requirements that come up during the discussions with them, aside from privacy and aside from the data on-chain?
我可以接。我覺得最大的是價值主張。嘿，我如果上鏈，如果把證券、資產帶到這個網路，使用情境是什麼？跟誰交易？能做什麼？需求在哪？買方在哪？所以生態系效應，我覺得其實最重要，在台灣我通常也先講這個。就是：台灣怎麼進全球市場？我們怎麼把資產，帶到 Canton 上其他對手方所在的全球市場？或者我怎麼讓台灣機構買美國公債更容易？Yeah, I can take that. I think the biggest thing is just the value prop. Hey, if I come on-chain, if I bring my securities, my assets to the network, what's the use case? Who do I trade with? What can I do? Where's the demand? Where's the buy side? And so that ecosystem effect is, I think, actually the most important thing and what I always lead with in Taiwan, for instance. It is: how does Taiwan enter the global market? How do we bring our assets, you know, into the global market with other counterparties that are on Canton? Or how do I make purchasing of U.S. Treasuries easier for Taiwan institutions?
我怎麼因為接上 Broadridge 的日內流動性，把美元資金成本降下來？那是立刻能用的情境。立刻，我甚至可以先講，嘿，你想不想跟 Goldman 在鏈上交易，然後發新聞稿？亞太很多人會很有興趣，對吧？因為一切是講他們的語言，講區塊鏈實際能歸到你什麼，不管是損益，還是我可以用來往上爬、拿成績的潛在使用情境。對，完全同意。所以我們確實要解真實世界的問題。貿易融資是大生意，是兆美元級。How can I lower my USD funding rate due to being able to tap into intraday liquidity on Broadridge? That's a use case immediately. Immediately, I could even lead with, hey, do you want to do trading with Goldman on-chain and do a PR? And that would be, like, something that a lot of people would be very interested in, right, in APAC, because it's all about speaking their language, speaking what does blockchain actually attribute to you, be it P&L or be it, you know, potential use cases that I can use to climb up and get credit for, right? Yeah, totally, totally. So, like, definitely we need to solve real-world problems. So, like, trade finance is a huge business. It's like trillions of dollars.
金融危機之後，法規讓銀行越來越難幫非洲或其他國家的小農、礦工做盡職調查。所以資金缺口很大。一旦我們能把貿易資料上鏈，貿易融資可以跟上。例如在 IOTA Foundation，我們有一個生態專案叫 Solos。他們已經有盧安達的鉭、銅或鉺，一些怪的稀土金屬，然後一貨櫃可能值幾百萬美元，然後你得到這個。你在鏈上拿到收據。你有物聯網鎖，有衛星監控，然後可以做貿易融資放貸。After the financial crisis, the regulation make the bank harder and harder to do the due diligence for small farmer, miner in, like, Africa or other countries. So they have a huge funding gap. So once we can put this trade data on-chain, trade finance can follow. So, for example, at IOTA Foundation, we have an ecosystem project called Solos. They already have, like, tantalum or copper or erbium, some weird, like, rare earth metal in Rwanda, and then you put, like, one cargo of this thing can worth, like, a few million dollars, and then you get this. You get a receipt on-chain. You have the IoT lock, like you have the satellite monitor, and then you can do lending as trade finance.
貿易融資是主菜，架構沒有免費午餐
所以我們為咖啡和酪梨做貿易融資，也為關鍵礦物做。我們的想法是，如果銀行做不了這筆貿易融資，我們可以開始幫忙。那就是我們這裡的價值主張。就我們所知，政府通常比較樂意先從貿易資料開始，但那只是第一步。我們一直知道主菜是貿易融資。所以我是從以太坊的人或社群成員角度評論。這很有意思，因為我覺得當以太坊的人很自然會想，這條鏈的吸引力是什麼？So we do trade finance for coffee and avocado, but we also do it for critical minerals. That's how we think, like, if bank cannot do this trade finance, we can start to help. And that's our value proposition here. And, like, as far as we know, government is always more, like, happy to start with the trade data, but that's the first step. And we always know the main dish is the trade finance. So I'm commenting kind of, like, from an Ethereum person or, like, community member perspective. It's very interesting because I think when— ETH folks kind of, like, naturally think about, like, hey, what's the appeal for the chain?
我們很自然走到鏈本身的技術性，對吧？共識、最終性這些。我聽到的，我試著從他們身上學，跟企業、跟政府的對話裡，真正吸引他們的是這些，真實應用。所以這些鏈、這些基金會，其實自己在做應用。顯然，對錯要看你怎麼想，我不認為一定有對錯。我是在呈現我學到的：那基本上是你們的立場。而不是：這是平台。這很有意思。Like, we naturally go to, like, just the technicality of the chain itself, right? Like, you know, the consensus, the finality, et cetera. What I'm hearing, and this is, you know, I'm trying to learn from them, like, what they find intriguing with the conversations. With the enterprise, with the government, is that these are the conversations that actually intrigue them, the real applications. So the chains, the foundations are actually doing the apps themselves. Obviously, there's, I don't think there's, there could be right and wrong, you know, depending on how you think. I'm presenting, like, this is kind of, like, the learning I'm getting, like, that's your stance, essentially. Instead of, like, this is the platform. It's very interesting.
我覺得，有時是運氣；有時有東西勝出。看不同人的哲學很有意思。現在我還是想回到技術，這比較是技術會議。架構的取捨是什麼？因為我們是工程師，知道沒有免費午餐，對吧？Canton 不可能在資料都在上面的情況下做到那麼好的隱私，而完全沒有一點犧牲。以太坊也為了我們認為很好的那些東西做了很多犧牲。IOTA 也是，對吧？I think, you know, sometimes it's luck; sometimes something wins out, et cetera. You know, it's interesting to see different folks' philosophy. Now, I do want to get to, because it's still, again, the technical side of things, this is more of a technical conference. Like, what are the trade-offs for the architecture? Because we are engineers, we know that there is no free lunch, right? There is no way that Canton can achieve that awesome privacy with all the data on there without any little bit of sacrifices, right? Like Ethereum also made a lot of sacrifices in order to achieve a lot of the great things we consider really good. IOTA also did that, right?
所以，當工程師，我們知道自己在做取捨。你們發現自己在做哪些取捨？所以第一個是，到最後，區塊鏈節點跑在物理世界。我們不是活在真空裡。所以得有人跑節點，跑 IOTA 節點絕對比跑以太坊節點貴很多。所以第一，比較中心化，這是肯定的。另一件事是，我們的共識確實限制節點數量。也許可以到 100、150，但肯定到不了 1 萬。所以去中心化要多少才夠，永遠在辯。And so, like, as engineers, we know we're taking the trade-offs. What are the trade-offs that, you know, you find yourself, like, doing, and then, like, yeah. So the first is, like, at the end, blockchain nodes run in the physical world. We don't really live in a vacuum. So you need someone to run the nodes, and to run an IOTA node is definitely much more expensive than to run an Ethereum node. So first, it's more centralized, that's for sure. Another thing is, like, our consensus does limit the number of nodes. So maybe we can go to 100, 150, but we for sure cannot go to 10K. So it's always a debate about how much is enough decentralization.
IOTA 節點更貴，Canton 不適合車庫開發者
我們知道一個肯定是中心化，大概 1 萬個肯定算去中心化，Solana 大概有 1,000，多數 Cosmos 鏈也許 100、150，我們也知道BSC 大概 27 或 45。但到最後是光譜。所以我們相信這確實是取捨。但不是每一個都只是取捨。有時精彩的想法真的給你第一性原理的改善。就像我們說，區塊鏈賣的是區塊空間，但不是所有交易都需要排序。所以對 IOTA、對 MoveVM、對 DAG，我們發現如果你不必為某些交易排序，只要走共識被核准，那就。你省下很多資源。We know one is for sure centralized, and we know probably 10K is for sure decentralized, and we know Solana probably has 1,000, and we know most Cosmos chains maybe have 100, 150, and we also know maybe Binance Smart Chain has like 27 or 45. But at the end, it's a spectrum. So we believe this is definitely a trade-off. But not every single trade-off. Sometimes, like, brilliant ideas really give you first-principle improvement. So as we say, like, blockchains do sell blockspace, but not all the transactions need ordering. So for IOTA or for MoveVM or for the DAG, we found if you don't need to order some of the transactions, you just go through consensus and get approved, then. You save a lot of the resources over there.
只有當這些交易在搶同一件事，像 Uniswap v2 池，你才需要排序。所以我會說 MoveVM 或 DAG 提供的是第一性原理的改善，不只是技術取捨。但我們也同意，至少節點這塊，硬體成本比跑以太坊節點高很多。對。我覺得 Canton 最大的取捨，是對 DeFi 開發者有多友善。我覺得以太坊漂亮的地方是，我可以從車庫把合約部署到這個網路。那可以用來在全世界規模化移動價值。Only when these transactions, they are competing to the same thing, like a Uniswap v2 pool, then you need to order it. So I would say the MoveVM or the DAG do provide first, like, a first-principle improvement, not just technical trade-off. But we also agree, like, at least for the node, you need to, like, get a much— higher hardware cost compared to running the Ethereum nodes. Yeah. I think for Canton, the biggest trade-off we made was how, like, DeFi developer-friendly it is. I think the beautiful thing about Ethereum is that, you know, out of my garage, I can deploy a contract onto this network. That can be utilized to move, you know, value at scale across the world.
Canton 上你也能做同樣的事，但要管的多很多，責任也重很多。為什麼？你管自己的驗證者。你管自己的資料。講到你部署到網路上的應用，你不是廣泛部署到一組驗證者。你自己就是驗證者，你負責那份基礎建設，自己應用的可用時間。我會說那是我們做過最大的取捨。那是機構比較喜歡的，但大概不是想快速生出 Uniswap v3 分叉的開發者想要的。好，時間差不多了，但我想請各位給現場最後幾句。With Canton, you can do it the same, but it requires a lot more management, a lot more responsibility. Why? You manage your own validator. You manage your own data. When it comes to the applications you deploy onto the network, you don't deploy it broadly onto a set of validators. You are the validator yourself, and you are in charge of that infrastructure, the uptime of your own application. That, I would say, is the biggest trade-off that we've made. It's what institutions prefer, but it's probably not what a dev, you know, trying to spin up a Uniswap v3 fork would want. Okay, so I think our time is almost done, but, like, I think I'd like, you know, for you guys to have some final words, final words for the crowd.
像，為什麼應該，這條鏈贏的部分是什麼，或任何想講的。也許我先。我講話習慣看聽眾調，台灣對我很近，我媽是台灣人，我現在住這裡，所以我也講中文。我真的希望看到我們，台灣資本市場，成為亞太先驅、走上鏈。而台灣目前法規允許的只有 Besu 這種私鏈。你必須全部跑在自建機房。多數金融機構根本不准雲端軟體。Like, why should— What's the winning, perhaps what's the winning part of the chain, or, you know, anything that you'd like to chat about? Yeah, maybe I can go first. So I always like to tailor my talks to the audience I'm in, and, you know, Taiwan being very close to home, my mom's Taiwanese, I live here now, so I speak Mandarin as well. I really want to see us as, you know, Taiwan capital markets being a pioneer in APAC and coming on-chain, and what Taiwan currently allows under current regulation is only private chains like Besu. You have to have everything running on-premise. Most of the financial institutions do not allow cloud-based software at all.
結語：台灣可自建節點，IOTA 慢慢做政府案
所以我希望在場各位帶走的最大重點是：Canton 是開源的。我們可以讓你把驗證者部署在自建機房。我們可以讓你跑自己的基礎建設，擁有完整主權。我覺得這會改變遊戲，這個區域很多人還不知道 Canton 能這樣。對。所以。講到 IOTA，我想說，IOTA Foundation 一直知道我們在台灣有很強的社群。我只是社群負責人，但這也來自所有社群成員。然後不只是 2018 年跟台北市政府簽的合作備忘錄，我們也有，例如 Turing Certificate 做台灣學校與公務員的教育證書與 DID。And so the biggest takeaway I hope everyone here understands is Canton is open source. We can allow you to deploy your validators on-prem. We can allow you to run your own infrastructure and have complete sovereignty. I think that's a game changer and something that a lot of folks in the region do not know yet about Canton. Yeah. So. When it comes to IOTA, I want to note, like, for IOTA Foundation, we always, like, we know we have a very strong community in Taiwan. I'm just the community lead, but it also comes from all the community members. And then not just, like, the MOU we signed with the Taipei City Government back in 2018, we also have, like, for example, we have Turing Certificate also do all the— education certificates and DID in Taiwan for the, like, school and for the government officer.
我們確實想跟政府合作，但知道不會快。但 IOTA 這種老 Layer 1 的好處是，我們知道這要時間，我們有耐心。例如歐洲區塊鏈服務基礎建設 EBSI，我們花了四到五年才拿到商業化前採購。甚至還不在生產。在歐盟做這些已經花了四五年。所以我想帶走的訊息是：IOTA 也許慢，但我們會到，我們有毅力把事情做完、交付。對。從我這邊，你知道，我們在以太坊社群這裡。We definitely want to really work with government, but we know it will not be fast. But the good part about an old layer one like IOTA is, like, we know, like, it takes time and we have patience. For example, for the European Blockchain Services Infrastructure, for EBSI, we— took four to five years to get a pre-commercial procurement. It isn't even in production. It already cost us four or five years to do, like, the things in the EU. So I think the take-home message is, like, IOTA, we may be slow, but we will get there, and we have the persistence to get things done, to deliver. Yeah. And from my side, you know, we're in the ETH community here.
我邀請的原因是，我們做機構日，原因有幾個，然後在機構日有幸跟其他鏈聊。我有很多朋友。我會覺得自己相當加密貨幣原生，某種程度上相當去中心化 maxi、也是隱私 maxi。但我在那個光譜上也有朋友更極端，他們連談都不想談。我是技術宅，所以只要夠好我就中立。我也愛看不同視角。也許不只技術的事。那就是我想向他們學的，因為他們某種程度上是成功的，至少繼續活著，在這裡，新的、進來的等等，對吧？The reason I'm inviting—well, there's a couple of reasons that we're doing this Institution Day to start with, and then on Institution Day, to have the, you know, pleasure to chat with the other chains. I have a lot of friends. Like, I would consider myself quite crypto-native, quite decentralized-maxi in some way, privacy-maxi as well. But I have friends also on that spectrum, but in the more extreme, they don't even want to talk. I'm a technical nerd, so I'm neutral as long as it's great. I'm also a perspective nerd. There's more things maybe than technical stuff. And that's what I want to learn about them, because they are, you know, in some way successful, like at least continuing to survive, and here, and new and coming, et cetera, right?
所以我覺得至少有東西可以學，對知識有好處。就這樣，座談先到這裡。謝謝留下來的各位，也謝謝講者。謝謝 Teagan。謝謝 Ko-Wei。謝謝。好，謝謝 Martinet、Teagan 和 Ko-Wei，帶來這麼完整的座談。So I think there could be something to learn, at least, like, good for the knowledge. And with that, I'd like to wrap up the panel. Thank you guys for staying, and thank you, panelists. Thank you, Teagan. Thank you, Ko-Wei. Thank you. Okay, thank you, Martinet, Teagan, and Ko-Wei, for bringing such a comprehensive panel.
