Strong consistency: After a write completes, all subsequent reads return that value. Simple to reason about, expensive to implement.
Eventual consistency: After a write, reads might return stale data for a while. Eventually, all reads will return the new value.
Causal consistency: If event A causes event B, everyone sees A before B. But unrelated events can appear in different orders.
Most large-scale systems use eventual consistency with a convergence window of milliseconds to seconds. Your Twitter feed doesn't need to be perfectly synchronized worldwide.