It’s pretty old stuff, but still…

Here’s how you typically generate a random number in C++:

    std::random_device rd;
    std::mt19937 gen(rd());
    std::uniform_int_distribution<int> dis(1, 100);
    auto random_number = dis(gen);

Let’s dissect it a little. On the one hand, “random_device” is a rather nebulous concept: it may or may not be truly random, and it may or may not be cryptographically secure. Truth be told, it’s completely implementation dependent. On the other hand, “mt19937” is an awfully specific random number generator implementation. These two creatures from two different worlds are usually used in tandem: the above is a boiler-plate example for generating random numbers in C++.

Seasoned C++ programmers got used to a class named mt19937, but to the regular folks this resembles a Reddit user name rather than class name. If I saw a class named fb24498 in a pull request, I would reject it, and so would most of my colleagues. But it is somehow OK for the C++ standard. As a user, do I really need to remember the magic number 19937 and what “mt” stands for? Isn’t it, like “multi-threaded” or something? Don’t I have other important things to care about? Of course, I can always use ranlux24, but it’s hardly better.

Compare it to Python:

import random
random_number = random.randint(1,100)

The seeding is done for you automatically, no questions asked. Yes, it can be argued that this is kind of vague, but so is random_device. C++ version does not add much clarity, but totally kills the ease of use.

How about C# then?

var rng = new Random();
int num1 = rng.Next(1, 100);

Go?

rand.Seed(time.Now().UnixNano());
num := 1+rand.Intn(100);

Nobody but C++ seems to force users to remember magic numbers like 19937…