A magnitude 7 is not slightly worse than a magnitude 6
The distance from 6 to 7 on a seismic scale is the same on the page as the distance from 1 to 2. It is not the same on the ground.
Each whole step of earthquake magnitude multiplies the energy released by about 31.6, which is 101.5, and it comes straight from how the scale is defined. Two steps is a factor of 1,000. From a magnitude 6 to a magnitude 9 is a factor of roughly 31,600.
A magnitude 9 is not fifty percent worse than a magnitude 6. It is thirty thousand times more energetic, and the two numbers sit three apart on a scale that the eye reads as a ruler.
Why the scale is built that way
Because the range is otherwise unusable. Earthquakes that instruments can detect span something like fifteen orders of magnitude in energy. A linear scale would need numbers from 1 to 1,000,000,000,000,000, and comparing two of them would mean counting digits.
Taking the logarithm turns multiplication into addition and squeezes that range into single digits. That is a real service, and it is why the same trick is used for sound (decibels), for acidity (pH), for stellar brightness and for the loudness knob on anything you own.
But the compression is the point and the danger. A logarithmic scale hands you a number whose spacing feels linear because every scale you learned before it was.
The same trap, in a glass
pH is defined as the negative logarithm of hydrogen ion concentration, so each unit is a factor of ten.
Water at pH 7 and rain at pH 5 are not "two apart" in any meaningful sense: the rain has a hundred times the hydrogen ion concentration. Lemon juice near pH 2 is a hundred thousand times more acidic than neutral water. When a report says a lake's pH has fallen from 6.5 to 5.5, it has not moved a little: the acidity has multiplied by ten, and that is usually the difference between fish spawning and not.
Decibels behave the same way and catch people out in the same direction: +10 dB is ten times the sound intensity, and +3 dB is roughly double. A road that is "only 6 dB louder" is delivering four times the acoustic energy.
What to do with a logarithmic number
Two habits, and they cost nothing.
Ask what one step multiplies by. The answer is not always ten. It is 10 for pH, 10 for decibels of intensity, and 31.6 for seismic energy, because that scale was built on a 1.5 exponent so that the numbers matched an older amplitude scale. There is no universal step size, so the number alone is not interpretable until you know the base.
Convert differences into ratios before comparing. "Magnitude 7.2 versus 6.8" is a gap of 0.4, which sounds like rounding error. As a ratio it is 100.6 ≈ 4, so one released four times the energy of the other. That is the sentence a reader can actually use.
The earthquake energy tool does the conversion both directions, which is the fastest way to build the reflex: type two magnitudes that look close together and watch what comes out.
And the scale in the headline is probably not the one in the paper
Almost every news report says "on the Richter scale". Almost none of them mean it.
Charles Richter's original 1935 local magnitude was calibrated to one specific instrument, and it saturates: above roughly magnitude 7 the response stops distinguishing larger events, so a genuinely enormous earthquake and a merely huge one return nearly the same number. Since Hanks and Kanamori in 1979 the figure quoted for large events is the moment magnitude, computed from the rigidity of the rock, the area of the fault and the slip across it — a physical measure of the work done, and one that does not saturate.
The two were deliberately aligned so the numbers agree through the middle of the range, which is why the substitution went almost unremarked. It matters at the top. The 1960 Chile earthquake is Mw 9.5, and no Richter-style local magnitude could have told you it was bigger than an 8.5.
The reporting problem
None of this is a criticism of logarithmic scales, which are the correct tool for quantities spanning many orders of magnitude. It is a criticism of quoting them unconverted to people who will read them as linear.
"A magnitude 7.1, up from the 6.4 last month" is accurate and communicates almost nothing. "About eleven times the energy of last month's" is the same fact, in units a reader already owns.