Subject: Urgent: Heating Repair

Hi [Landlord’s Name],

I’m following up about the heating, which has been out of service for three weeks. I understand repairs can take time, but the issue is still unresolved, and I need the heating restored as soon as possible.

Please let me know by [date] when a qualified repair person will attend and when you expect the system to be working. I’d appreciate a prompt update and a clear plan for completing the repair.

I’d like to resolve this cooperatively, but the continued lack of heating is becoming difficult to manage. Thank you for treating this as urgent.

Best, [Your Name]

import { useEffect, useState } from "react";

export function useDebouncedValue<T>(value: T, delayMs: number): T {
  const [debouncedValue, setDebouncedValue] = useState<T>(value);

  useEffect(() => {
    const timeoutId = window.setTimeout(() => {
      setDebouncedValue(value);
    }, delayMs);

    // Clears the pending update when the value changes or the component unmounts.
    return () => window.clearTimeout(timeoutId);
  }, [value, delayMs]);

  return debouncedValue;
}

The hook returns the initial value immediately, then updates it after delayMs without further changes. Cleanup cancels any pending timeout on unmount.

Assuming “99% accurate” means the test has 99% sensitivity (it catches 99% of users) and 99% specificity (it correctly clears 99% of nonusers), the probability is about 33%.

Out of 10,000 people:

• About 50 use the drug; the test catches roughly 49.

• About 9,950 don’t; 1% of them—roughly 100—test positive by mistake.

So only about 49 of the roughly 149 positive results are true positives: (49/149 \approx 33\%).

It surprises people because they focus on the test’s high accuracy and overlook the low base rate: there are so many more nonusers that even a small false-positive rate produces many false positives. Also, “99% accurate” by itself is ambiguous; the calculation depends on the test’s sensitivity and specificity.