Day 2 — What is a blockchain, really?
You have probably already heard the shared spreadsheet analogy. A blockchain is a shared spreadsheet that nobody owns and nobody can secretly edit. As far as analogies go, it's a good one. It gets you about 80% of the way there. The other 20% is where this thing actually becomes interesting, because the part the analogy hides is the part nobody could figure out for forty years.
The problem looks deceptively simple. Imagine you and a thousand strangers want to share a single database. Anyone can write to it. Anyone can read it. There is no central authority deciding whose writes count. Nobody trusts anyone else. The strangers can be on different continents, using different software, joining and leaving the network whenever they want. Half of them might be trying to cheat. You want all of you to end up agreeing on what the database says.
Computer scientists called this the Byzantine Generals problem. It was first formalized in 1982. It described a group of generals trying to coordinate an attack on a city by sending messengers. Some of the generals were traitors. Some of the messengers were unreliable. The math kept proving that under certain conditions, the loyal generals could never reliably agree on a plan. The problem of getting strangers to agree on a shared state, without trusting any of them, was considered effectively unsolvable for general-purpose computing.
Then Satoshi proposed a solution nobody had quite tried. Two pieces fit together to make it work.
The first piece is the chain. Every entry in the database (a block) contains a cryptographic fingerprint (a hash) of the previous entry. If anyone tries to alter an earlier block, the fingerprint changes, and every block built on top of it breaks. The chain is what makes the ledger tamper-evident. You cannot rewrite history without leaving a trail.
The second piece is the incentive. To add a new block, you have to do real computational work, a lot of it, and you only get rewarded with new bitcoin if your block is accepted by the rest of the network. This is proof-of-work, and it does something subtle. It makes honesty economically rational. If you spent all that compute trying to push a fraudulent block, the rest of the network would reject it, and you would have burned the electricity for nothing. The cheapest way to get rewarded is to be honest.
Stack those two pieces together and you have a system where everyone sees the same record, nobody can alter the record without being detected, new entries get added in a way that makes cheating more expensive than cooperating, and the whole thing keeps running even if a meaningful percentage of the participants are bad actors.
That is the blockchain. The spreadsheet analogy is right that it is a shared record. What the analogy misses is that the design solves a forty-year-old computer science problem in a way that turns trust into math.
When people talk about Bitcoin's "trustlessness," this is what they mean. You are not trusting any specific person to maintain the record. You are trusting the incentive structure to make it irrational for participants to lie. That is a very different kind of trust. It is the kind of trust you have in a vending machine versus the kind of trust you have in a salesperson. The vending machine cannot decide to keep your money. The math is the math.
A few more properties worth knowing.
The blockchain is append-only. You can add new blocks. You cannot edit or delete old ones. This is why people say a blockchain is immutable. It is not literally impossible to change, but the cost of changing it grows with every block added on top, and after a few thousand blocks of additional history, the cost is effectively infinite.
The blockchain is distributed. There is no master server. Every full participant in the network (a node) has a complete copy of the entire history. If half the participants disappeared overnight, the other half could keep going. This is why you cannot "shut down" Bitcoin the way you can shut down a company.
The blockchain is public, at least for Bitcoin and most major chains. Anyone in the world can download the entire history. Every transaction since January 2009 is sitting on hard drives all over the planet, freely inspectable. You can look any transaction up yourself using a tool called a block explorer (we link one below).
Tomorrow we use this foundation to look at Bitcoin specifically and answer the question almost nobody asks out loud: why does a token created by an anonymous person on the internet actually have value, and how do you reason about it as more than a number on a screen.
Glossary
| Term | Definition |
|---|---|
| Block | A bundle of transactions added to the blockchain at one time. New blocks are produced on a regular cadence (about every 10 minutes for Bitcoin, about every 12 seconds for Ethereum). |
| Hash | A cryptographic fingerprint of a piece of data. If the data changes at all, the hash changes completely. Used to chain blocks together and detect tampering. |
| Proof-of-work | The mechanism by which Bitcoin participants compete to add the next block. Requires real computational effort, which makes attacks expensive. |
| Append-only | A property of the blockchain. New entries can be added; previous entries cannot be deleted or edited. |
| Immutable | Describes a record that cannot practically be changed once it is written. The blockchain is immutable in this practical sense. |
| Node | A participant in the blockchain network. Full nodes hold a complete copy of the chain's history and help validate new blocks. |
| Distributed | A system where no single party hosts the entire record. Every node holds its own copy. |
| Byzantine Generals problem | A 1982 computer-science problem describing the difficulty of getting independent parties to agree on a shared state when some of them may be dishonest. Bitcoin's blockchain solves a real-world version of this problem. |
| Block explorer | A website that lets you read the blockchain. Etherscan, mempool.space, and similar tools show every transaction, address, and block in human-readable form. |
Reality check
A blockchain is "trustless." But you are still trusting something. What, specifically, are you trusting when you use one?
If your answer was "I'm not trusting anything," go back. There is always something you are trusting. The shift is in what that thing is, not whether it exists. Get specific about what replaces the bank or the government in the trust diagram. If you can name it, you have the lesson.
Read deeper
1. A brief history of Bitcoin by The Block
A timeline of how the first blockchain went from nine-page whitepaper to global asset.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. Most "history of Bitcoin" pieces read like a Wikipedia article. This one is tighter. Read it for the rhythm of how slowly the early years moved and how fast everything compounded after about 2017. There is an important lesson hidden in that pacing: every legitimate new technology spends a long time looking like a toy before it starts looking like infrastructure. Twitter looked like a toy. The internet looked like a toy. Bitcoin spent ten years looking like a toy and is now sitting on most institutional balance sheets that matter. Use that pattern as a lens for everything else you encounter in this course.
2. What are block explorers? by The Block
The window into the chain. If you want to actually look at a blockchain, this is your tool.
Read on IMPCT (curated commentary) | Read original (theblock.co)
Deven's take. This is the moment in the course where you stop reading about crypto and start looking at it. Spend ten minutes on mempool.space or Etherscan and pick a transaction at random. Click through it. Look at the addresses. Look at the fees. The point is not to understand every field. The point is to internalize that this is open. Every transaction, every wallet, every flow of capital is sitting there for anyone to inspect. Hold that in your head every time someone tries to tell you crypto is opaque or shady. The chain itself is the most transparent financial record in human history. What's opaque is usually the centralized layer sitting on top of it.
3. The Byzantine Generals Problem by Lamport, Shostak, and Pease (1982)
The original computer science paper that defined the problem Bitcoin would later solve.
Read on IMPCT (brief curated summary) | Read original (microsoft.com/research)
Deven's take. You do not need to read the math. You do need to know this paper exists, because it is the formal definition of the problem Bitcoin solved. For most of the 26 years between this paper and the Bitcoin whitepaper, getting strangers to agree on a shared state without a trusted third party was considered a known limitation of distributed computing. Bitcoin did not invent the question. It produced the first practical answer. Skim the abstract and the introduction, then move on.
4. Bitcoin and the Byzantine Generals (short blog explainer)
An accessible bridge between the 1982 paper and what Bitcoin actually did about it.
Read on IMPCT (curated commentary)
Deven's take. If section 3 felt dense, read this one. It is the bridge. Pay attention to the part about how the cost of attack grows with the size of the network. That is the load-bearing insight. Bitcoin's security is not magical, it is economic. The bigger the network, the more it costs to attack, the less rational it becomes to try. Most things in life that scale destroy something on the way up. This is one of the few systems that gets stronger as it grows.
Tomorrow
We use everything you learned today to answer the question almost nobody can answer clearly: why does Bitcoin actually have value? Anyone can copy the code. Anyone could launch a competing chain. What makes the original worth what it's worth, and how do you reason about it without leaning on either "digital gold" hype or "internet money" dismissal?
See you in the morning.
Preview reader
You are reading a private preview of IMPCT Institute. If something landed, didn't land, or felt confusing on this lesson, tell us. Short notes are useful. Long notes are useful. No notes are also fine.