Bitcoin is digital money that works without a bank. It was introduced in 2008 in a nine-page whitepaper by a person or group using the pseudonym Satoshi Nakamoto, and went live on January 3, 2009. To this day, no one knows who Satoshi really is.

What makes Bitcoin special: there is no central authority that maintains accounts, approves transfers, or issues new bitcoins. Instead, the entire system runs on thousands of computers worldwide, all keeping the same ledger. This shared ledger is called the blockchain.

Three properties define Bitcoin:

• Fixed supply: There will never be more than 21 million bitcoin. This cap is hard-coded and cannot be casually changed.
• Censorship-resistant: No one can ban, block, or reverse a Bitcoin transaction. Whoever holds the private key controls the coins.
• Pseudonymous: Bitcoin addresses are not directly tied to real names — though every transaction is publicly visible.

Bitcoin is often compared to gold: scarce, hard to counterfeit, storable — just digital.

Picture a huge ledger that contains every Bitcoin transaction since 2009. This ledger isn't stored in a bank's filing cabinet — it lives in identical form on tens of thousands of computers worldwide. Each of these computers (a node) holds a complete copy.

The blockchain is the technical form of that ledger. It consists of blocks — essentially "pages" — added roughly every 10 minutes. Each new block contains:

• the transactions of the last ~10 minutes,
• a reference to the previous block (a kind of fingerprint, the "hash"),
• and a mathematical proof that someone invested computational work (see Mining).

Because each block references the previous one, you get a chain — hence "blockchain." To tamper with an old block, an attacker would have to recompute every block after it as well — competing against the entire global compute power behind Bitcoin. Practically impossible.

The blockchain is public. Anyone can download it (~700 GB as of 2026) and trace every single transaction since day one.

A block is a bundle of transactions plus some metadata. Three numbers matter here:

Block time (~10 minutes): A new block is found roughly every 10 minutes on average. Sometimes it's 3 minutes, sometimes 25 — but the average always settles back to 10, because the difficulty (see below) adjusts. These 10 minutes were Satoshi's design choice: long enough for all nodes to agree on the new block, short enough that transactions confirm reasonably fast.

Block size (~1–4 MB): For a long time, the limit was 1 MB per block. With the SegWit upgrade (2017) and later Taproot (2021), a block can effectively hold up to 4 MB of data — though the math is complex (keyword: block weight, measured in "weight units"). The cap is deliberately small so that ordinary people can run a full node without renting a data center.

Block reward: The miner who finds a block gets two rewards: newly created bitcoin (the block subsidy) plus the transaction fees of all transactions in the block. The subsidy halves every 210,000 blocks (see Halving).

Block size is the most politically charged topic in Bitcoin's history. In 2017 the community split over it: one faction wanted bigger blocks and created Bitcoin Cash. Bitcoin itself kept the small blocks and solved the throughput problem with layer-2 solutions like the Lightning Network instead.

Mining is the process of producing new blocks — and at the same time the mechanism that secures Bitcoin.

Each miner collects transactions from the network and tries to assemble them into a valid block. "Valid" means: the block's hash (a kind of digital fingerprint, a number) must be smaller than a certain target. The network sets that target via the difficulty.

The catch: you can't "compute" the hash. You can only guess. You change a small number inside the block (the nonce), compute the hash, check if it's small enough — and if not, change the nonce and try again. Trillions of times per second, worldwide, on specialized machines (ASICs).

Whoever finds a matching hash first has "found" the block, broadcasts it to the network, and collects the block reward. The whole process is called proof of work, because the hash proves that real computational work — and therefore real electricity — was spent.

Why this seemingly absurd guessing game? Because it's tamper-proof. Anyone wanting to rewrite history would have to redo all that work — and outrun all honest miners combined. Practically impossible once the network is large enough.

Hashrate measures how many hash computations are performed per second across the entire Bitcoin network. It's the best metric for Bitcoin's security.

Current scale (2026): around 800 exahashes per second — i.e. 800,000,000,000,000,000,000 hashes per second. Eight hundred quintillion.

The difficulty is the number that defines how small a block hash must be. The higher the difficulty, the smaller the required hash, the rarer a valid one — and the more hashes are needed on average to find a block.

The elegant part: difficulty adjusts automatically. Every 2,016 blocks (about every two weeks), the network checks:

"How long did the last 2,016 blocks take? Target: 14 days (2,016 × 10 min)."

• Faster than 14 days → difficulty rises.
• Slower → difficulty falls.
• Maximum jump per adjustment: ±300%.

This keeps average block time at ~10 minutes whether there are 100 miners on the network or 100 million.

Example: In May 2021, miners had massive deployments online; difficulty was at all-time highs. Then China banned mining, and 50% of the hashrate dropped overnight. Blocks suddenly took ~20 minutes. Two weeks later: difficulty adjustment -28% — the largest drop in history. Block time back to 10 minutes.

So difficulty is a self-regulating mechanism. You can think of it as the thermostat of the Bitcoin machine.

The halving is arguably the most important event in the Bitcoin cycle. Every 210,000 blocks — roughly every 4 years — the block subsidy (i.e. the number of new bitcoin per block) is cut in half.

Mining is a brutal business. Four levers determine profitability:

1. Hardware efficiency (J/TH) — how much electricity a miner consumes per terahash. Modern ASICs (Antminer S21, WhatsMiner M50S+) sit around 18–22 J/TH; older units 30–40+.
2. Electricity price ($/kWh) — by far the biggest running cost. Industrial miners pay $0.03–$0.05/kWh; anything above $0.08/kWh is barely profitable today.
3. Block reward + fees — what comes in. Currently (post-2024 halving): 3.125 BTC subsidy plus an average of 0.1–0.3 BTC in fees per block.
4. Network hashrate — how many other miners compete for the same block.

Simplified math:

Expected daily revenue per miner ≈ (own hashrate / network hashrate) × 144 blocks × block reward × BTC price

Example: A miner with 100 TH/s in a network of 800 EH/s has a share of 0.0000125%. At 144 blocks/day × 3.125 BTC × $76,000/BTC, the expected daily revenue is about $4.30 — minus electricity.

Mining today is industrialized: Texas, Kazakhstan, Paraguay, Norway, the US sunbelt — anywhere electricity is cheap and the climate cool.

This is the technical core of Bitcoin. Three terms that often get conflated:

Private key — A randomly chosen number. Specifically: 256 bits — i.e. a number between 1 and 2²⁵⁶. That's an unimaginable number of possibilities — more than atoms in the Milky Way. Whoever knows this number owns the corresponding bitcoin. Period. No one needs to approve it, no account, no bank.

E9873D79C6D87DC0FB6A5778633389F4453213303DA61F20BD67FC233AA33262

Private Key  →  Public Key  →  Address
   (geheim)      (teilbar)      (öffentlich)

A wallet is — despite the name — not a container for your bitcoin. Bitcoin never sits in your wallet; it sits on the blockchain. A wallet is a keyring: it manages your private keys.

Seed phrase (12 or 24 words)

Instead of memorizing 256-bit numbers, every modern wallet uses a seed phrase — a sequence of 12 or 24 English words drawn from a fixed list of 2,048 words (the BIP39 standard).

witch collapse practice feed shame open despair creek road again ice least

Rule of thumb: anything you can't afford to lose belongs in a cold wallet. Like cash in your wallet vs. savings in a safe.

Bitcoin has no account model like your bank. There's no table that says "Wang has 0.5 BTC." Instead Bitcoin works with UTXOs — unspent transaction outputs.

Think of them as cash bills.

If you pay €12 at a shop with a €50 bill, you don't get the same bill back with "minus €12" written on it. You hand over the 50, the cashier keeps 12, and you get 38 in change.

UTXOs work exactly the same way:

• Every bitcoin amount someone receives is stored as a UTXO — like a bill of a particular denomination.
• Sending bitcoin means consuming one or more UTXOs in full.
• The recipient gets a new UTXO of the desired amount.
• The rest goes back to your own address as change.

Concrete example: you have a UTXO of 0.5 BTC and want to send 0.1 BTC. The transaction looks like this:

Input:  0.5 BTC (your UTXO)
        ↓
Output 1: 0.1 BTC → recipient
Output 2: 0.3998 BTC → your own change address
Fee:     0.0002 BTC → miner

When you send a Bitcoin transaction, it doesn't go straight into a block. It first lands in the mempool — a sort of "waiting room" on every node worldwide. Miners pick transactions from this pool for the next block.

How do miners choose? By fee — specifically by fee per byte (sat/vB). Whoever pays more gets in faster. It's an open market.

Rules of thumb:

• Mempool empty: 1–5 sat/vB is enough. Confirmation in 10–60 minutes.
• Mempool moderate: 10–30 sat/vB. Confirmation in 1–6 blocks.
• Mempool full (hype, Ordinals boom, etc.): 100+ sat/vB. Otherwise your transaction can sit for hours.

What if I underpaid the fee?

• RBF (Replace-By-Fee): resend the same transaction with a higher fee, replacing the old one. Most wallets support this.
• CPFP (Child-Pays-For-Parent): create a follow-up transaction with a high fee; miners are incentivized to include both together.
• Wait it out: after 14 days unconfirmed, the transaction typically gets dropped from mempools — the coins are back with you as if nothing happened.

Sat/vB? "Satoshi per virtual byte." One satoshi is 1/100,000,000 BTC. A typical transaction is 140–250 vBytes. So at 50 sat/vB that's 7,000–12,500 sats in fees — at BTC $76,000 that's roughly $5–$10.

Mempool as a signal: a full mempool means high demand for blockspace. That can be a market signal — e.g. Inscriptions/Ordinals hype, panic selling, or just rush hour.

A node is a computer that has downloaded the entire Bitcoin blockchain and verifies all the network's rules itself. It's the foundation of decentralization.

What does a node do?

• Stores the full blockchain (~700 GB).
• Validates every new transaction and block: is the signature valid? Is the hash correct? Is no bitcoin being double-spent?
• Forwards valid transactions and blocks to other nodes.
• Rejects invalid transactions/blocks — no matter who sends them.

Here's the crucial part: Bitcoin isn't secure because miners are honest. Bitcoin is secure because tens of thousands of nodes independently verify whether miners follow the rules. A miner producing an invalid block gets ignored by the rest of the network — regardless of hashrate.

Who runs nodes?

• Individuals (e.g. on a Raspberry Pi 4 with an SSD)
• Companies (exchanges, wallet providers)
• Developers, researchers, idealists
• Total: ~15,000–20,000 reachable nodes worldwide (plus an estimated 50,000+ private behind firewalls)

Why run a node yourself?

• Sovereignty: you trust no one. Your wallet queries your node, not a third party's server.
• Privacy: no one sees which addresses you're watching.
• Robustness: you strengthen the network.

Hardware requirements are modest: a €200 Raspberry Pi setup (e.g. Umbrel, Start9) is enough.

A fork is an update to Bitcoin's software rules. There are two kinds:

Soft fork — a backward-compatible tightening of rules. Old nodes still accept the new blocks. Examples: SegWit (2017) and Taproot (2021). Soft forks are the preferred upgrade mechanism for Bitcoin because they don't split the network.

Hard fork — a non-backward-compatible change. Old nodes reject the new blocks. If not everyone upgrades simultaneously, the network splits into two chains. Most famous example: Bitcoin Cash (BCH) in August 2017 — one faction wanted larger blocks, split off, and created their own coin. Bitcoin Cash ABC vs. Bitcoin SV and further splits followed later.

Activation mechanisms (for soft forks):

• Miner signaling (BIP9): miners signal approval in the blocks. At 95% support over a period, the upgrade activates.
• User Activated Soft Fork (UASF): nodes set a deadline — any miner producing old-style blocks after it gets ejected. Used in 2017 to activate SegWit and considered a turning point in the "Block Size War."

Bitcoin's philosophy: changes should be conservative. The network shouldn't shift based on slim majorities. Consensus across developers, miners, and nodes is required — and hard to reach. That's a feature, not a bug: it makes the money predictable.

Bitcoin can handle around 7 transactions per second on the base layer (Layer 1). Visa handles 65,000. So how is Bitcoin going to become a global means of payment?

Answer: the Lightning Network — a second layer (Layer 2) on top of Bitcoin that enables fast micropayments off the blockchain.

Simplified flow:

1. Two parties open a channel by jointly locking bitcoin into a multisig address (= an on-chain transaction).
2. Inside this channel they can send bitcoin back and forth as often as they like — completely off-chain, in milliseconds, with micro-fees (fractions of a cent).
3. When done, they close the channel with a second on-chain transaction. The final balance settles on-chain.

Routing: if Alice and Bob don't share a direct channel but both have one with Charlie, Alice can pay Bob via Charlie. Lightning is a network of such channels — hence the name.

Good for:

• Micropayments (tips, in-app purchases, pay-per-second streaming)
• Fast retail payments
• Cross-border transfers in seconds instead of days

Weaknesses:

• Channels need liquidity management — complex for beginners.
• Recipient must be online (bidirectional liquidity required).
• For large amounts, on-chain still makes more sense.

El Salvador has used Lightning as a national payment layer since 2021. Apps like Phoenix, Wallet of Satoshi, or Strike make Lightning simple for end users.

Bitcoin's security rests on three assumptions:

1. More than 50% of hashrate is honest. If an attacker controls more than half of the global compute, they can rewrite the last few blocks — a 51% attack.
2. Cryptographic primitives remain secure. SHA-256 (for hashing) and ECDSA/Schnorr (for signatures) aren't broken. A sufficiently powerful quantum computer could theoretically threaten the latter — but that's research today, not reality.
3. Enough nodes independently enforce the rules. Without decentralized validation, one party could unilaterally change the rules.

What can a 51% attacker actually do?

• Reverse their own transactions ("double spend").
• Block or censor other transactions.

What can they NOT do?

• Steal other people's coins (they don't have those private keys).
• Create more than 21M BTC.
• Rewrite history more than a few blocks deep (exponentially more expensive per block).

Cost of a 51% attack on Bitcoin (as of 2026): estimated at multiple billions of dollars in hardware plus ongoing power costs. For a short attack window. At the current network size, it's barely economically viable — and even then, the attack would likely crash the BTC price, hurting the attacker themselves.

What helps as a user?

• For large amounts, wait for 6 confirmations (~1 hour). After that a reorg becomes practically impossible.
• Run your own node.

There are thousands of cryptocurrencies. What sets Bitcoin apart?

1. Genuine decentralization. No founder premine, no VC with preferred shares, no "foundation" treasury. Satoshi vanished, their coins never moved. No other protocol comes close.

2. Longest track record. Bitcoin has run since January 2009 without meaningful downtime. Every night, in every war, every crisis, every bull and bear market — the blockchain keeps producing blocks.

3. Fixed monetary schedule. 21M cap. Halvings on a 4-year clock. No one can change this without overturning nearly the entire economy behind the network.

4. Conservative upgrade model. Bitcoin changes slowly and only by consensus. Other chains pivot regularly — great for innovation, bad for the property "predictable money."

Other categories:

• Ethereum (ETH): smart contracts, DeFi, NFTs. Different goal — less "digital gold," more "programmable platform." Switched from Proof of Work to Proof of Stake in 2022.
• Stablecoins (USDT, USDC): no inherent value claim, USD-pegged tokens. Practical for payments, not scarce.
• Memecoins (DOGE, SHIB, etc.): speculative assets without fundamental thesis. Not the same as Bitcoin.
• Privacy coins (Monero, Zcash): focus on full privacy. Bitcoin is only pseudonymous.

Rule of thumb: read the Bitcoin whitepaper, then read another whitepaper, and you'll usually see quickly that Bitcoin is a different animal. Other coins can be useful tools — they're not competing in the same discipline.

As of: April 2026. Bitcoin keeps evolving — this wiki gets updated accordingly.