Try measuring the thickness of a piece of paper with a ruler. The smallest tick marks on the ruler I have are one mm apart, so the smallest length I can measure is 1 mm, much larger than a sheet of paper. Now, if I were just a non-thinking machine, I would report the thickness as "1 mm with an error of 1 mm". Someone dishonest then states that this method of measuring thickness reported a measurement of 1mm (without mentioning the error), and then claims that all thickness measurements are flawed and cannot be trusted. This is similar to the dating of the Mt. St. Helen's rocks by a method that cannot be accurate for very rocks.
This particular Mt. St. Helens example is a bit more complex; I will repeat JonF's
link to better discussion for this.
™ Version 4.2