I'm reading a book right now. (I know, this isn't news.) It's hard sci-fi, but the sort I like, which means the scientific details and speculation don't revolve around computers and spaceships. I'm enjoying it. It's fun. I may have stayed up too late last night to keep reading. However, there's one minor aspect of the story that bugs me whenever it shows up, and which pulls me out of the story as a result.
Let's say you've got an alien. It's humanoid, but because of the shape of its vocal tract, it can't produce one of the sounds found in English. You make a point of telling your readers this, with demonstrations. Yet a couple chapters further on, you have dialogue in which this alien is producing that sound, and you introduce an alien character whose name, in her own language (i.e., not English), has the sound as well. Um. I know linguistics isn't the author's speciality, but really, a little bit of thought on the part of the author or their editor should have caught that.
This is why accuracy and continuity (and research) are good things.* I'm willing to forgive a lot of things as a reader, and I'm willing to forgive this in the long run because it's a minor detail and I'm liking the book, but huge errors and glaring lacks of research are never fun. I don't like dropping out of the story like that. I expect authors to know at least as much as I do, especially since they research and all. I'd imagine I'm not alone in this—which is why writers should research and check for continuity even for small things. I would've caught the speech sound thing, so why couldn't they?**
I know facts get missed, research gets dropped due to time constraints, and authors think, "Nobody'll notice!" when in fact, someone always does. I am as guilty of that as the next writer, and am sure I'll be getting all sorts of … interesting fanmail when I'm finally published. If I want to be absolutely perfect on every front, I will never finish researching, let alone writing. At some point I'll have to start fudging things. (Let's not get into my angst over that, except to say yes, I'm angsting, and there's an interview I need to do that I'm terrified about.) But I still maintain that small details aren't hard to verify, and large details should be verified as much as possible—but who wants interesting fanmail?
* Not to say that the author in question didn't do research because man, did he!
** There are some who'd say I have high expectations. Me? High expectations? Never! Perish the thought.