I thought of another example, but, (like my previous example), outside of the context of storylines, but *is* related to semantics and the use of language.
In cellular biology, there is a term "copy number variation." At my last count, I tallied about 9 different phenomenon that use this term, and each mean something very unique, and can be detected using a specific technique. For example, when cancer cells proliferate, they are rapidly dividing. There is a technology to measure the amount of DNA in each cell. Right before the cell divides, its DNA is double what it was. So if you measure the amount of DNA in hundreds or thousands of cells, and analyze the results, you'll get two groups, one with a normal amount of DNA, and a second with double that amount, because the DNA has been copied, but the cell has not yet divided.
Then there is a second type of copy number variation. Cancer cells can lose their ability to repair their chromosomes. During normal cell division, the daughter cells end up with two of each of the chromosomes. But during cancer, a chromsome might break, or you might end up with three chromsomes in one cell, and only one in the other. These can be detected by looking at the cells under a microscope. Completely different technology than the "copy number" technology used to see if a cell is actively dividing.
As DNA technology has proliferated, so has the use of this term. But what happens is that researchers will analyze cells using a technology for one of these "copy numbers" or "CNVs", and then report their findings as if they've measured a completely different "CNV". Because the two have the same name (CNV), the scientists are completely oblivious that their work includes a semantical error.
Another example of the problem of semantics in science is the term "energy" which can be used to describe just about anything. In using xrays to determine the three dimensional structure of a protein, the output data is a list of the atoms of the molecule, with the coordinates of that atom (much like a 3D map). Some proteins will have an arm that has two possible conformations. Some proteins might have that branch sitting to the left, others to the right. Because of the technology used, which measures thousands of atoms and records the average location, the "signal" for these arms is weaker than for the atoms that have a constant location. To indicate this uncertainty in the output file, each atom is provided a numerical value to indicate how strong the signal is. The name of of this value is called a "temperature factor", with the idea that it's an appropriate term, since at high temperatures, molecules move around, meaning both the factor and "temperature" are a measure of a change of position. But what happens next, is you have PhD scientists who take these "temperature factors" literally, and they argue that different regions of the protein have different physical temperatures and thus different energies. And while it is true that energies differ everwhere... in the universe even, the argument they're using to equate the two is completely semantic. They'll even report "temperatures" (instead of "temperature FACTORS") above 100, and ignore the implication they're suggesting, that molecule is boiling or being fried. If you point this out, you are met with hostility.
Decades ago, as a student being confronted with all of this, I used to have nightmares about coming into highschool English class, and not being prepared for an exam, or not having prepared an assignment that was due. Looking back, I wonder if those dreams were trying to tell me the problems I was facing with the use of language.