Happens-Before and Lamport Clocks
There is no global "now"; causality is the only ordering you get. The foundational conceptual shift of the whole part.
There is no global "now"; causality is the only ordering you get. The foundational conceptual shift of the whole part.
"NP-hard" hides that some hard problems are easy when one parameter is small. The most practically useful refinement of Karp's 21 problems.
The positive side: 2-approximate vertex cover, PTAS, and where the approximability cliff sits.
Every proof can be checked by reading three bits, and therefore some problems are hard even to approximate.
Exact distinct-count needs linear space; approximate needs almost none. HyperLogLog and Count-Min as the constructive half.
What a hash function actually guarantees, as opposed to what it seems to do.
Simplicity, formalized as compression; the bridge from information theory to model selection in PAC learning and VC dimension.
The length of the shortest program producing a string. Uncomputable, and the correct definition of "random" and "simple".
Why P vs NP has resisted for fifty years, stated as two theorems about which proof techniques cannot work.
Thousands of unrelated-looking problems are secretly the same problem.
SAT is NP-complete: circuits encode computation. The technical heart of the whole field.
Verifying vs finding. State it properly, including what a proof either way would actually change.
How many errors you can fix given how much you're willing to pay in redundancy.
You can communicate at capacity with arbitrarily low error. Shannon's most surprising result: reliability is free below the limit.
And the proof is an exchange argument you can follow. Opens with Kraft (The Kraft inequality and prefix codes): which length assignments exist at all, then which...
Random placement gives max load ~log n / log log n; picking the emptier of two random bins collapses it to log log n.
"Solve A by solving B" is how undecidability and hardness both spread. The mechanical skill behind Parts II and IV.
Four conditions, all necessary for deadlock; break any one. The cleanest example of a characterization theorem doing engineering work.
Any sufficiently strong consistent system has true statements it can't prove, and can't prove its own consistency.
Any program can obtain its own source. Quines aren't a party trick; self-reference is guaranteed by the model.