From Solana to CKB: The Mental Model Shift You Need in Week 1
Most people learning CKB are not learning blockchain from scratch. They are migrating from another chain, and that usually makes the first week harder, not easier. If Solana was your first serious mental model, you already carry a set of defaults about how blockchain systems work. You expect state to live in accounts, programs to load and mutate that state, and transactions to feel like requests for execution against a shared runtime. CKB does not break all of those instincts, but it rearranges them enough that familiar patterns can become misleading. That is the real purpose of Week 1. It is not to memorize new vocabulary. It is to replace one mental model with a better one. The core thesis for this note is simple:CKB makes more sense when you stop thinking about mutating accounts and start thinking about validating state replacement through cells and scripts.
Who this note is for
This note is for developers who already understand at least one blockchain beyond surface level. In practice, that usually means people coming from Solana, Ethereum, Polygon, Sui, or a similar ecosystem. That background helps, but it also creates drag. Once you understand one chain well, your mind starts carrying hidden assumptions about:- what state looks like
- how state changes
- where authorization lives
- where programmability lives
- what a transaction is really doing
0 -> 1 can sometimes feel easier than learning 1 -> 2. Starting from zero means you do not have to unlearn anything. Moving from Solana to CKB means you do.
Blockchain is a shared state machine
The cleanest foundation for Week 1 is not “CKB is like Bitcoin” or “CKB has cells.” It is this:A blockchain is a shared state machine.
That definition matters because it shifts your attention to the right questions. Different chains are not mainly distinguished by branding, programming language, or virtual machine design. They differ in more fundamental ways:
- what the unit of state is
- how state gets updated
- where authorization rules live
- where business logic lives
- how the system accounts for long-lived storage
The first bridge: rows, accounts, objects, cells
If you are coming from Solana, the easiest first bridge is to compare the unit of state across systems:- traditional databases use
rows - Solana uses
accounts - Sui uses
objects - CKB uses
cells
cell. Everything else in Week 1 becomes easier once that point lands.
The biggest shift: replacement instead of mutation
This is the first real mental reset. On Solana, the default picture looks something like this: a program loads accounts, checks constraints, and mutates account data in place. Even when the flow is complex, the state model still feels mutation-oriented. On CKB, the center of gravity is different. A transaction consumes existing cells, scripts verify whether that transition is valid, and new cells are created as the successors to the old ones. This side-by-side sketch is the fastest way to see the model change:- Solana:
programs mutating accounts - CKB:
scripts validating replacement of state objects
The cell is the unit of state
If the Solana developer’s basic object is the account, the CKB developer’s basic object is the cell. A cell is usually introduced through four parts:capacitydatalocktype
capacity tells you how much storage-backed value the cell carries. data is the payload. lock defines who is allowed to spend the cell. type can define the rules that govern what kind of state transition is valid for that cell.
This is a better place to begin than VM details because it gets you closer to the actual design of the chain. On CKB, storage budget, ownership conditions, and transition rules are closely tied to the state object itself.
Programmability lives in scripts
Solana teaches you to think in terms ofaccounts + programs. CKB is easier to reason about as cells + scripts.
That analogy is useful, but only if you keep the translation precise:
- a Solana account roughly maps to a CKB cell
- signer and spend authorization logic roughly maps to a
lock script - state-machine and business-rule logic roughly maps to a
type script
- the
lock scriptcontrols who can spend the cell - the
type scriptcontrols what state transitions are allowed
Ownership feels more asset-native
One phrase you will hear often is that users own cells, not contracts. The important part of that claim is not the slogan. It is the design implication behind it. On CKB, the state object itself carries the conditions for how it can be spent and, when relevant, how it can evolve. The object is not simply a slot in a large mutable contract-owned storage map. That makes ownership semantics feel more explicit and more asset-native. If you are used to contract-centric systems, this is another place where CKB may feel unfamiliar at first. The object is more self-describing, and the transition rules are closer to the object being protected.Storage is not treated as cheap
CKB forces you to treat state as scarce. More bytes mean morecapacity. More capacity means more long-lived claim on shared chain state. That is why CKBytes are not a side detail in the architecture. They are part of the architecture’s view of what persistent state should cost.
From a Solana perspective, the nearest intuition is account sizing and rent, but the comparison only goes so far. On CKB, storage accounting feels more native to the state object itself. The cost model is harder to ignore because it is built directly into the way cells are represented and carried forward.
Verification is the core operation
If you only keep one sentence from this note, keep this one:On CKB, computation serves verification.
That does not mean CKB cannot support rich logic. It means the cleanest way to read the chain is to ask:
- what cells are being consumed
- what cells are being created
- what scripts must approve this transition
Why the PoW claim belongs in the same conversation
CKB’s arguments for proof of work make more sense once you see CKB as a durable shared state layer. When CKB advocates say PoW matters for decentralization and censorship resistance, the important point is not abstract ideology. The stronger claim is about the state being protected. If the base layer responsible for validating and preserving long-lived state is more neutral and harder to capture, then the assets and data living on top of it inherit stronger guarantees. That is why consensus and storage belong in the same discussion on CKB. The chain’s security model is tied to its role as a durable store of shared state.How to use this in Week 1
Do not try to memorize every term in isolation. Use this mental model as a guide while you work through the foundation materials. Start with the local exercises inweek1/foundation:
01-cell-model-explorer.ts02-transaction-anatomy.ts03-capacity-calculator.ts04-first-ckb-transfer.ts
- the
cellis the base state object - transactions replace old state with new state
- scripts verify whether that replacement is valid
- storage has an explicit economic weight