From a comment on a Jonathan McCalmont blog:
On the subject of science fiction it seems that most reviewers do not consider the science worth paying attention to. Either they cannot recognize when it is bad or do not care. But readers have this problem also.
I have seen two commentators getting the artificial gravity wrong in Heinlein’s Orphans of the Sky even though they claim to like the story. There was zero G at the center of the ship and pseudo-gravity at the outer edge. 7th and 8th graders should know that.
In Leviathan Wakes the asteroid dodges the starship to avoid impact but there is no explanation of how the zombie mud knew it was coming. Ian Banks never explicitly says the Air Spheres have very low gravity but describes a scene that can only happen in low-G. Some things deserve bad reviews.
The commenter’s position seems to be that the science needs to be paid attention to in genre novels. Fair enough! How far down do you go with that. I am reminded of the Heinlein “butcher paper” story, where he figured out a number for one of his stories after 3 days of equations and calculations. Was that too much? Is there too much?
A writer can spend years down a rabbit hole of fractal detail, though–so when does the writer stop?
To give a basic example: There are four bases of DNA: Cytosine, Thiamine, Guanine and Adenine. Uracil for Thiamine only in RNA, right?
Well, there are viruses where there IS Uracil instead of Thiamine in DNA. Does it bear mentioning in a story with a basic touch on DNA? Do you break a reader if you don’t mention Uracil in connection with certain DNA if its tangentially relevant? If its deeply relevant?
This also applies to any discipline, really, but science fiction, in the name, suggests the science has to be dead on perfect. But given how fractally complex science is, where do you stop in that recursive knowledge or looking for it?