Subject: Urgent: Heating Repair Needed – [Your Name], Unit [Your Unit Number]

Dear [Landlord or Property Manager Name],

I hope you are doing well.

I am writing to follow up on the heating issue in my apartment. Unfortunately, the heat has not been working for three weeks now, and the unit is becoming uncomfortably cold.

I understand that repairs can sometimes take time to coordinate, but this issue is now affecting the livability of my home. I would like to resolve this amicably and get the heating fixed as soon as possible.

Could you please let me know when a technician will be scheduled to come out? I am available [mention your general availability, e.g., weekday afternoons or weekends] to provide access.

Thank you for your prompt attention to this matter. I look forward to hearing from you.

Best regards,

[Your Name] [Your Phone Number]

Here is a clean, TypeScript implementation. It tracks the previous value so it only resets the timer when the value actually changes, and it cleans up the timeout on unmount (or when the value/delay changes).

import { useEffect, useRef } from 'react';

/**
 * Debounces a value. Returns the value after `delay` ms of inactivity.
 * The timeout is cancelled automatically when the component unmounts,
 * when the delay changes, or when the value changes.
 */
export function useDebounce<T>(value: T, delay: number): T {
  const debouncedValueRef = useRef<T>(value);
  const timeoutRef = useRef<ReturnType<typeof setTimeout> | null>(null);

  // Update immediately when the delay changes so the new delay applies right away
  useEffect(() => {
    debouncedValueRef.current = value;
  }, [value, delay]);

  useEffect(() => {
    // If the value hasn't actually changed, keep the existing debounced value
    // and do not restart the timer.
    if (debouncedValueRef.current === value) {
      return undefined;
    }

    // Cancel any pending timeout
    if (timeoutRef.current !== null) {
      clearTimeout(timeoutRef.current);
    }

    // Set a new timeout to commit the debounced value
    timeoutRef.current = setTimeout(() => {
      debouncedValueRef.current = value;
    }, delay);

    // Cleanup: cancel on unmount or before re-running this effect
    return () => {
      if (timeoutRef.current !== null) {
        clearTimeout(timeoutRef.current);
      }
    };
  }, [value, delay]);

  return debouncedValueRef.current;
}

Example usage

import { useState } from 'react';
import { useDebounce } from './useDebounce';

export default function Search() {
  const [query, setQuery] = useState('');
  const debouncedQuery = useDebounce(query, 500);

  // Only fetch after the user stops typing for 500ms
  // useEffect(() => { ... fetchResults(debouncedQuery) }, [debouncedQuery]);

  return (
    <input
      value={query}
      onChange={(e) => setQuery(e.target.value)}
      placeholder="Search..."
    />
  );
}

Key points

If you need to execute a callback (rather than just a value) after a debounce, consider a useDebouncedCallback hook instead.

Assuming “99% accurate” means the test correctly identifies drug users 99% of the time (true positive rate = 99%) and correctly identifies non-users 99% of the time (true negative rate = 99%):

Use Bayes’ theorem

[ P(\text{user}\mid\text{positive}) = \frac{P(\text{positive}\mid\text{user}) \cdot P(\text{user})}{P(\text{positive})} ]

[ = \frac{0.99 \times 0.005}{(0.99 \times 0.005) + (0.01 \times 0.995)} ]

[ = \frac{0.00495}{0.00495 + 0.00995} = \frac{0.00495}{0.0149} \approx \boxed{0.332} ]

So the probability is about 33.2%, or roughly 1 in 3.

Why this surprises people

Intuitively, many hear “99% accurate” and assume a positive result means there is a 99% chance the person uses the drug. That ignores the base rate: only 0.5% of the population are users, so they are vastly outnumbered by non-users.

Even though the false-positive rate is just 1%, it is applied to the 99.5% of people who don’t use the drug. That produces a large number of false positives (about 9,950 per million people), which nearly doubles the number of true positives (about 4,950 per million). As a result, a positive test is only about twice as likely to come from a user as from a non-user, pulling the posterior probability down to roughly one-third.

This is the classic base-rate fallacy: people overweight the test’s accuracy and underweight how rare the condition is in the population.