I always forget what closures are, even if I continue to use them. The article makes a case for considering closures as stateful functions which is quite neat. Basically, closures allow a function to return another function (or functions); each returned function will also have access to any data that the original function had access to. In other words:

const add = function(x) {
    return function(y) {
        return x + y;
    }
}
const add10 = add(10);
add10(10); // 20

That will output 20 because the internal function has access to both variables. It's a bit freaky but it means you can define functions as reusable abstractions with preset variable values really easily. As the article points out, that can be a lot more secure and can help encapsulate sensitive data/functionality, like so:

const manageBankAccount = function(initialBalance) {
    let accountBalance = initialBalance;
    
    return {
        getBalance: function() { return accountBalance; },
        deposit: function(amount) { accountBalance += amount; },
        withdraw: function(amount) {
            if (amount > accountBalance) {
                return 'You cannot draw that much!';
            }

            accountBalance -= amount;
        }
    };
}

const accountManager = manageBankAccount(0);

accountManager.deposit(1000);
accountManager.withdraw(500);
accountManager.getBalance(); // 500