Is Thanos’ Snap Truly Random? Exploring Randomness, True RNGs, and Java Generators

Randomness

Randomness refers to the lack of predictability in a process, where an indeterminate factor continuously generates outcomes. In computing, random numbers are widely used, especially in communication security and modern cryptography.

Random numbers are classified into true random numbers and pseudo‑random numbers. True random numbers are derived from physical phenomena (coin toss, dice, electronic noise, nuclear decay) and must satisfy randomness, unpredictability, and non‑reproducibility. Pseudo‑random numbers are generated by deterministic algorithms with a seed; they are not truly random but can approximate randomness.

True random numbers: obtained from physical experiments such as coin flips, dice rolls, electronic noise, nuclear fission, etc.

Pseudo‑random numbers: produced by algorithms and a seed; all software RNGs belong to this category.

Is Thanos’ Snap Fair?

In the Avengers movies, Thanos claims his “snap” eliminates half of all life with random, unbiased, and fair selection. By examining the three properties of randomness—randomness, unpredictability, and non‑reproducibility—we can assess the fairness of the snap.

Randomness : The list of survivors in Avengers: Infinity War shows many paired characters (e.g., Iron Man–Spider‑Man, Captain America–Winter Soldier) suggesting non‑random selection. Moreover, Thanos knows he will survive, contradicting true randomness.

Unpredictability : Doctor Strange uses the Time Stone to view 14,000,005 possible futures and selects one, indicating the outcome is not unpredictable.

Non‑reproducibility : The movies reveal only one winning outcome among millions of possibilities, and the final gesture in Endgame suggests that outcome is being reproduced, violating non‑reproducibility.

Therefore, the snap does not satisfy the criteria for true randomness.

The process can be modeled as a layered sampling algorithm:

Identify and cache DNA of individuals exempt from deletion.

Partition the remaining DNA pool into groups (e.g., Earthlings, Asgardians) and store each group in separate databases.

Further subdivide each group by attributes such as gender, age, or occupation.

Traverse each sub‑table and delete half of the entries (e.g., keep 512 of 1024 tables for Earthlings).

Re‑insert the cached exempt DNA back into the databases.

Restoring the “deleted” individuals would involve replaying the corresponding binlog entries.

True Random Number Generators

True random numbers are produced by hardware that exploits physical phenomena that are intrinsically unpredictable, such as thermal noise, photo‑electric effects, or quantum processes. Classical thermal‑noise generators are still deterministic in principle, but quantum‑based generators provide intrinsic randomness.

In 2018, NIST researchers reported a quantum‑mechanical method that generates provably random numbers, improving cryptographic security.

Random Number Generation in Java

Java offers several APIs for generating random numbers:

Using the system clock

final long l = System.currentTimeMillis();

To obtain a number in [0, 99]:

final long l = System.currentTimeMillis();
final int i = (int)(l % 100);

Math.random()

final double d = Math.random(); // 0.0 ≤ d < 1.0

Convert to an integer in [0, 100):

final double d = Math.random();
final int i = (int)(d * 100);

Random class

Random random = new Random();
int i2 = random.nextInt(100);

In multithreaded contexts, ThreadLocalRandom offers better performance and thread‑local state.

SecureRandom

The java.security.SecureRandom class provides a cryptographically strong RNG that draws entropy from the operating system. It is thread‑safe but may be slower and can block while gathering entropy.