The blockchain trilemma: why "Ethereum killers" always sacrifice something

Every new L1 promises "fast, cheap, and decentralized". The blockchain trilemma is the empirical rule that of the three properties (decentralization, security, scalability) only two are achievable at once. The third is sacrificed, and the entire difference between networks is in what exactly was given up.

How to read a network's design

• Ethereum sacrificed base-layer scalability: expensive and slow, but thousands of independent validators and the maximum cost of attack. Scale was moved to the L2s.
• Solana sacrificed the validator entry threshold: the hardware costs serious money, there are fewer and larger nodes - in exchange for speed and penny fees on the L1.
• The small "killers" most often sacrifice everything at once for a pitch-deck TPS number: a few dozen validators controlled by the foundation is a distributed database with a token.

Questions that expose the sacrifice

How much does validator hardware cost? How many independent operators hold a third of the stake? Has the network halted, and who restarted it? Who can roll back transactions? The answers are public, and they say more about a network than any TPS counter.

Reviews of specific L1s - in the section.