Now You Know

A Canton contract is never edited. Move part of a holding and the old one is archived, while new ones are written for the receiver and for your change.

What would you build on a ledger that never overwrites a step?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A tokenised Treasury can settle against a stablecoin in one step. Both legs sit in one transaction, so both sides finish together, or neither does.

Which asset would you settle this way next?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A daily repo figure adds every loan that settles that day, and a short loan that is rolled is counted again.

Which figure are you holding, the loans that settled or the bonds still out?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A Canton contract needs every signatory to create it and every signatory to archive it. If the holder signed, the issuer can't end it outside the contract's own terms.

Which assets would you hold only on those terms?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A new Canton Super Validator seat opens only by vote. Two sitting Super Validators must back the written proposal first, then the operators vote it in.

What should a newcomer show before the keys turn?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A new Canton Super Validator can start with its weight hosted on another operator's node, rewards held in escrow. Past a weight of 2.5, it must run its own node within six months.

What should it prove first?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

On Canton, the node running a wallet and the party owning it are separate. Both sign to tie up. The owner picks the rope. Watch, confirm or submit. LG CNS has announced an MOU with the Canton Foundation.

Which rope would you hand over?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A reserve-backed token works like a cloakroom ticket. The coat stays on the rail in the back, and only the ticket changes hands. The Litecoin Foundation has backed a plan to bring Litecoin to Canton this way, as cLTC. That plan is announced, not settled fact.

Settlement is the moment ownership changes. On Canton, the ticket and the cash can move in one transaction, so both sides land or neither does. Nobody hands over a ticket and then waits for the money.

The ledger has a limit, though. It proves the ticket moved. It can't see the rail. Whether the coins are still there is a job for reserve reports, and for whoever checks them.

If you were taking the ticket, which proof would you read first?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Lloyds and Visa say they settled $750,000 of payment obligations in USDC during a seven-day live pilot, with funds arriving in under an hour, including over a weekend. Visa settled on a public chain. Lloyds ran its own node on Canton.

On a public chain, a transaction works like a postcard. Every computer on the network can read every line, which is how it checks the rules.

Canton cuts a transaction into views, seals each one to the parties it concerns, and delivers each party only its own envelope. Even the synchronizer, the courier that puts transactions in order, only ever carries the sealed versions. A bank shares the network, not its books.

Should a regulator hold a key to every envelope, or only to the ones it asks for?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Canton prices its fees in dollars but collects them in Canton Coin, so every fee needs a conversion rate, and the network doesn't read one off an exchange.

Every Super Validator submits the rate it believes the market is at. At the start of each ten minute round, the network publishes the median of those submissions, and that single rate converts every fee in the round.

One operator submitting a wildly wrong number is simply outvoted. But a median can't tell a wrong consensus from a right one. If the submissions drift together, the published rate drifts with them.

Should the network's rate follow the market more tightly, or is a median that trusts its own participants the safer design?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Every ten minutes, Canton opens a fixed batch of new Canton Coin for that round. Nobody is guaranteed a share just for showing up.

Apps and validators claim only what their recorded activity earns. Each share is filled in proportion to the activity actually recorded that round, and only up to a cap. A quiet round leaves room unclaimed.

What the caps leave over isn't burned and isn't handed to anyone. It drains into a pool of unminted coin, and only a two thirds vote of Canton's Super Validators can decide what happens to it next. The published curve is a ceiling, not a promise.

As usage grows, which fills faster, the pools that get claimed or the reserve that doesn't?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Three very different rooms can host the same trade. An order book lines up public bids and asks. A request for quote is a private conversation between a few dealers. A pool trades against a formula with no counterparty in sight.

Whichever room finds the price, the money still has to change hands safely. Under Canton's token standard, CIP-0056, each side sets its asset aside for that one trade, and a single settlement step swaps both legs together, or returns both if the deadline passes first.

The room decides how a price is found. It stops deciding whether either side is exposed while the trade settles, because the lock behind every door can be the same one.

On a ledger where only the parties to a trade can see it, does the room even matter anymore?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

When a big name joins Canton's governance, the headline weight is a ceiling, not a grant. The new operator's own node joins at zero.

The seat climbs one step at a time. Each step needs a confidential milestone confirmed, and the reward behind it waits in escrow until two thirds of the existing Super Validator operators agree to release it.

Reaching the ceiling doesn't end the work. Under CIP-0105, holding full weight means keeping more than half of everything the seat has ever earned locked away, on a schedule that eases only over years. Let the lock slip and the weight comes down with it.

Is a seat at Canton's table earned once, or rented continuously?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A bank posts the same security to borrow cash, to clear a trade and to back a position. Usually it does one job at a time, and whichever desk has it first, the others wait.

Tokenized, the security never leaves the vault. Only the claim on it moves, reassigned to whoever needs it next in one settlement step instead of a chain of confirmations.

In DTCC's July production event, a single tokenized Treasury was pledged as collateral, lent to another desk, delivered against cash and posted as clearing margin. Four jobs, one object, never physically touched.

Does that mean firms end up holding less collateral, or the same pile working harder?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Two tokens can both say tokenized Treasury and not be the same thing at all. The difference is bridged versus native.

A bridged token is a stand in. The real asset is locked on another chain, and a bridge mints a copy here. You hold an IOU, and you are trusting the bridge to honor it.

A native token is issued on Canton by the issuer itself, who signs off on the asset directly. It is the asset of record, not a copy pointing elsewhere. DTCC's tokenized Treasuries sit on this side: issued on Canton with DTC as custodian, not bridged from another chain.

When you hold a tokenized asset, are you holding the thing, or a promise about it, and do you know which?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

People say DTCC, Visa and the rest are all on Canton. But Canton is not one chain. It is many independent apps that share one thing, the Global Synchronizer.

The synchronizer is the shared clock. It puts every transaction in a single order and stamps it in time, so separate ledgers can act together and never disagree about what happened first.

It orders, but it does not execute, and it cannot read what is inside a transaction. That is what lets two apps settle in one atomic step, both legs or neither, with no bridge in between.

If one shared service does the ordering for everyone, what happens to the network the day it stalls?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

On most public blockchains, every validator sees every transaction. That full visibility is how the network agrees nothing was spent twice.

For a bank, that is a wall, not a feature. You cannot run payroll if every salary is public, or trade if every position shows. Visibility for everyone is the reason institutions stayed off chain.

Canton splits a transaction into parts, and each party sees only the parts they are in. The synchronizer that puts every transaction in order cannot read any of them. The ledger stays consistent, and still private.

If privacy is the thing that brought these institutions on chain at all, is it the feature, or the foundation?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Changes to Canton do not just ship. They pass as CIPs, Canton Improvement Proposals, the formal way anyone can propose a change to the shared network.

But not everyone votes. The ballots belong to the Super Validators, the operators that run the core of the network. A coin balance is not a vote.

To pass, a proposal needs a two thirds majority. If it changes the chain itself, it is only final once two thirds of Super Validator operators actually run it. And weight is not one flat number: it is set by governance, earned and capped.

If the operators who run the network are the ones who vote on it, what keeps that aligned with everyone else holding Canton?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Every claim about Canton sits in one of three boxes, and telling them apart is most of the work.

Confirmed you can check yourself. It is live on the network, filed with a regulator, or written on chain. Announced means a team said it is coming, but until it runs in production for people outside the project it is a plan, and a pilot is not production. Speculated is a read, useful to think with, never to hold as true.

The DTCC story runs through all three at once: a partnership announced, production trades since processed, and the full service still ahead.

Which box was the last Canton headline you read in, and did you check the primary yourself?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

An AI agent can now pay a Canton app directly. No account, no invoice, no human clicking approve.

It is not Canton's protocol. x402 is an open standard Coinbase built around the old HTTP 402 status code. Canton is simply one more network that now accepts it.

The mechanism is a four step handshake. An agent requests data, the server replies with a price instead of the payload, the agent signs a payment, a facilitator settles it on chain, and the content unlocks.

It is built for machine speed: small, repeated purchases that credit card rails were never designed to clear profitably.

Whether that becomes real commerce or stays mostly test traffic is still an open question industry wide, and Canton's own volume here is not sourced yet.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Canton settles trades in dollars, but not through one dollar.

USDCx came first. USDCx has been live on Canton since December 2025, issued through Circle's xReserve.

USD1 went native in August. World Liberty Financial's USD1 went natively live on Canton, issued and redeemed by BitGo.

Native means minted directly on Canton, so the cash leg and the asset leg settle in the same transaction, either both complete or neither does.

Which rail a given trade actually used is worth asking every time, since the two are not interchangeable in who stands behind them.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Ask what Canton actually is and you'll get four different answers depending on who's talking.

A company, then a foundation. Digital Asset is the company that built the Daml language and wrote the original protocol. The Canton Foundation is a separate, nonprofit body that governs the network today, coordinating Canton Improvement Proposals and the Super Validator set.

A network, then a synchronizer. Canton Network is the live, shared infrastructure all of that produces, the thing actually settling institutional transactions. The Global Synchronizer is one specific service inside that network, the part that lets separate privacy domains settle atomically without exposing each other's books.

A company builds it, a foundation governs it, a network runs it, a synchronizer executes the settlement. Collapsing any two of those into one noun is how a governance vote gets reported as a product launch.

Which noun a headline actually means is the first check before deciding what it implies for anything you hold.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

When a project announces it joined Canton, the headline rarely says which of three separate things it actually did.

Visa holds two of three. Visa is a Canton Foundation General Member and, under CIP-109, a Super Validator with a voting weight of ten. Neither makes it a Featured App.

Each track does something different. A General Member gets a seat at the table. A Super Validator runs the infrastructure that keeps the network live and votes on every CIP that changes it. A Featured App locks Canton Coin and earns issuance for building something people actually use.

An institution can hold all three, one, or none, and each confers a different kind of leverage over the network.

Which combination a project holds is worth checking before treating a headline as a single fact, and coverage does not do that consistently yet.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Trading on a blockchain usually means letting a venue move your funds, and often you approve that again for every single trade.

One lock, many settlements. Canton's Token Standard V2 adds a committed allocation: you lock funds to a venue once, and that single approval is allowed to cover many settlements. After a match, the venue names the exact transfer legs and settles against the locked funds, with no fresh approval each time.

Committed, but cancellable. Until a given trade actually settles, you can still pull the allocation back. That pairing is what lets a real order book run on chain rather than through a custodian.

Whether that combination scales to a full order book's cancellation volume without becoming its own bottleneck is the open question.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Canton's burn/mint ratio gets quoted like it proves something on its own: two sides of one number, apples over apples, so a rise in it reads as unambiguous progress. It is not unambiguous, and the reason is structural rather than a knock on the metric itself.

Mint is set by a schedule. Every round, the network caps how much CC becomes available to earn. That cap is denominated directly in CC, defined by the minting curve. It does not move because the coin got more or less valuable.

Burn is priced in dollars first. Traffic fees are set in USD, then converted to CC at whichever onchain rate holds at the moment of payment. That is the same mechanism an earlier issue used to show that a raw burn count conflates usage and price. This issue asks the next question: does dividing by mint fix that. It does not.

Divide a price sensitive quantity by a price blind one and the ratio inherits the sensitivity of the half that has it. A rising BMR can mean more real network usage, a weaker CC rate, or both moving together, and the ratio alone gives no way to tell those apart.

Which force actually moved is the open question, and only the dollar fee total, not the ratio, would settle it.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A CC transfer isn't instant. It's prepared, it waits behind a gate, then it executes. That gate used to snap open in ten minutes. As of CIP-0107, it holds for twenty four hours, and the reason isn't speed.

Ten minutes was inherited, not designed. It came from OpenMiningRound, the mechanism that prices and rounds transactions elsewhere on the network. A person signing off, sometimes more than one, doesn't fit in that gap.

The new window runs on its own contract. ExternalPartyConfigState holds the day's terms, stays active for 48 hours, and refreshes itself every 24, cut loose from the mining round entirely.

Scope is narrower than the framing suggests. The window covers token standard operations, preapproval creation and renewal, and traffic purchases. Reward minting stays on the original ten minute cycle, untouched by this CIP.

One network is now running two very different gates. What happens to the older one when a traffic-based reward model, already named as the next step, extends the newer one further?

I hold $CC. Narration is synthetic. The script, the claims and the sign-off are mine.

Any change to Canton starts as a text file anyone can write. That's a CIP, a Canton Improvement Process document, and an editor assigns it a number. Getting it adopted is the harder part, and it runs through two different two-thirds majorities that are easy to collapse into one.

Nothing moves without a sponsor. One Super Validator has to sponsor the proposal and a second has to endorse it before it's even eligible for a vote.

The first two-thirds is a vote. The sponsor sends it to a private ballot with a public archive. Designated representatives of the Super Validator rights owners get a ten day window, and two thirds in favour makes it Approved, announced publicly in cip-announce.

The second two-thirds is adoption, not a vote. Where the change touches the chain itself, it only reaches Final once two thirds of the actual validator node operators adopt it on their own infrastructure, a separate and much larger population moving on its own schedule, reported through the Canton Coin Scan API.

A proposal can sit Approved for a long time before the network actually runs it. Which of those two thresholds should move first, the vote or the adoption?

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A validator runs a node for you. A Super Validator runs the part everyone agrees through, and it votes on the rules. Two things get flattened when people talk about them, and both matter if you are assessing the network's governance.

The electorate is narrower than people assume. Votes on Canton Improvement Proposals are cast by designated representatives of the Super Validator rights owners, and approval needs a two thirds majority. Where a proposal requires on-chain steps, it becomes final only once two thirds of Super Validator node operators adopt it. Coin holders and ordinary validators have no direct ballot. That is not hidden, it is written into the published process.

Weight is not one number. Every addition is its own governance proposal carrying a named maximum, the approved set spans a wide range of caps, and recent ones are structured so weight is earned as milestones are verified rather than granted at approval. There is also a published process for reducing weight, which is the part almost nobody mentions.

Weight is bought with commitment. Underneath sits a framework with three tiers, where a Super Validator locks a share of its aggregate lifetime earnings and earns forward weight in proportion. Again a table rather than a number, and the top tier's requirement steps down over four years.

So the same set of institutions runs the infrastructure everyone settles through and writes the rules that govern it, with real capital committed against that role.

Whether you read that as alignment or as concentration is the open question, and I do not think it has a settled answer yet.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

A very large number lands with Canton's name attached, and it gets repeated all week. Usually nobody has said what it counts. Before you repeat it, ask the boring question first.

Most of the time, most of it is one company's own book. Its ledger, its customers, its business, run on software it controls. That total is real. It is not the network's.

The network's share is the part that crossed. What settled between parties who do not share a ledger. It counts less by construction, because it only counts what crossed. Neither figure is dishonest. They answer different questions, and only one of them tells you whether a shared settlement layer is being used as one.

This is a rule I had to learn by breaking it. I credited platform volume to the network once, wrote a standard so it would not happen again, then left the old claim sitting on my own homepage for weeks because the standard only governed what I wrote next. Both are in the public mistakes record.

So the test is two questions. What does this number meter, and who other than the party announcing it can check it? A figure with no answer to either is a headline, not a measurement.

Whether the smaller honest number grows into the big one is the open question, and it is the one worth tracking.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Your balance is not a number sitting in the network's ledger. It lives in a contract, and on Canton nobody can write a contract on your behalf without your authorisation. That single rule is why a withdrawal can leave an exchange and arrive nowhere.

A transfer is an offer, not a push. The sender proposes it. The receiving party has to accept before anything settles. On most networks the sender alone decides, and the coins land whether the recipient was ready or not.

The receiver arms a standing yes in advance. That standing yes is a transfer preapproval: a contract the receiving party creates so incoming transfers complete without accepting each one by hand. Most exchanges will not send to a party that has not got one.

Arming it is free inside a base window. CIP-0119, approved 15 June 2026, gives every preapproval a free base duration, ninety days by default. Creating or renewing one inside that window costs traffic and nothing more.

So the same design decision that lets an institution hold assets on a shared network without leaking its book to everyone else shows up at the retail edge as a withdrawal that quietly does nothing.

The failure is silent, and that is the open question. Nothing in what you see back tells you a missing preapproval is the reason. Whether that belongs to the exchanges to explain or to the standard to surface is unsettled.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

You never pay the Canton Network yourself. A validator hosts you, and the validator is charged. What it is paying for is called traffic, and every message submitted to the network consumes some.

The one that submits pays. The validators receiving a message are not charged for it. The cost rises with the size of the message, and a little more with every extra recipient it has to reach.

There is a free allowance first. Each validator holds an allowance that refills over time and is always spent before anything is bought. Only when it runs out does the validator buy more traffic, and buying means destroying CC.

So the bill follows whoever asked the network to do something, rather than whoever happened to be watching. That is a deliberate choice about who carries the cost, and it shapes who can afford to build here at all.

What happens when a single validator hosts thousands of parties and buys traffic for all of them is the open question. Traffic is accounted for per validator, so every party sitting behind one host shares a single balance.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Using Canton means buying traffic, and traffic is priced in US dollars. You pay for it by destroying Canton Coin. Those two facts sit underneath most of the confusion about what CC burn actually tells you.

The dollar price does not move when the coin does. A message costs what it costs in dollars. The network converts that to CC at the prevailing rate and burns it. So when CC is worth twice as much, the same message destroys half as many coins. When it is worth half as much, twice as many.

The same rule runs the other way. What an application can earn for a single action is capped in dollars too, so a higher coin price mints fewer coins for the same work. Both directions are denominated in something other than the coin.

Which means a chart of coins destroyed, on its own, is close to uninterpretable. It falls when the network is quieter. It falls when the coin is stronger. Nothing in the number itself separates those two.

Which of them is moving at any given moment is the open question, and it is not answerable from the burn figure alone. The dollar amount has to sit next to it.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Featured App status is the designation that lets an application earn Canton Coin rewards. Without it, an app can run on the network and earn nothing from doing so. It is not a favour the Foundation hands out. It has a price, and the price is paid in capital.

The price is a lock, not a fee. An ordinary application locks five million CC in its own identified party on the network. An application that issues assets locks twenty five million. Nothing is spent and nothing is paid to anyone. The coin is held, and only coin that is actively locked counts toward the requirement.

The lock has to hold. Fall below the threshold and the status is removed, and super validators are expected to be able to act on that quickly rather than at the next convenient moment. Unwinding a lock voluntarily takes sixty days, releasing a sixtieth of it each day. This is not a deposit that can be pulled the moment it becomes inconvenient.

The effect is to turn a permission into a position. To keep earning, an application has to hold a meaningful amount of the asset whose value its own activity helps create, and keep holding it.

Whether that selects for the most committed builders or simply the best funded ones is the open question. A well capitalised application that does very little can hold the status indefinitely. A busy small one may not be able to afford it at all.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Almost everyone files Canton under one of two headings: a public blockchain, or a private one owned by banks. It is neither, and the reason matters more than the label.

Permissioned does not mean secret. It means you cannot simply show up. Every participant is an identified company, checked before it can trade at all. Public is the other half: the network is shared infrastructure rather than one institution's private system.

The privacy sits at the level of the trade, not the network. A trade is visible to the two firms doing it and to nobody else. A regulator can be handed a window into that one trade without being given the whole book. And once a trade settles, it is settled; nobody quietly unwinds it afterwards.

So the difference from a chain like Ethereum is not who can look. It is who is let in. That gate is why serious institutions will touch this at all, and whether it also caps how fast Canton can grow is the argument worth having.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine.

Fees on Canton are set in dollars, paid in Canton Coin, and burned. New Canton Coin is minted to the apps and validators doing the work, up to 62 percent of issuance on a published schedule running to mid 2029.

Which means adoption reaches the coin through that gate and through nowhere else. Whether net issuance runs inflationary or deflationary is not a promise anybody can make. It is a live ratio between fee-driven burn and the minting schedule, and it is a number anyone can watch.

Watch the balance, not the headlines.

I hold $CC and $CNTN. Narration is synthetic. The script, the claims and the sign-off are mine. This is the archive cut: issue 01 shipped before the render lane existed and its narration script was not kept, so the words here are written from the lines the video shows on screen. The voice is the one it shipped with.