I’m going to push back on you a little bit here, because at least within your examples I don’t think you’re being consistent. You ask “what’s the point of” the inaccuracies in Fault Line when they could have been so easily fixed? And I agree, they could have. But so too, could Cory Doctorow have given his manuscript to any San Franciscan and gotten at least a few of his errors immediately fixed. I agree with you that they don’t affect the plot in any serious way, but it doesn’t sound as it Tashjian’s errors affect her plot either (I haven’t read Fault Line). Why mistrust Tashjian but not Doctorow? Her errors don’t have anything to do with the subject of abuse. It sounds to me as if you have a pre-existing trust of Doctorow (perhaps because you’ve read his other novels, or at least know about him from Boing Boing, etc.) but don’t have that for Tashjian. While that might work for you personally, I think we should be looking for something a little more objective–something that works for a novice to Doctorow and Tashjian both.
A big problem I’m having with getting my head around this discussion is how reader-specific it seems to get. You’ve lived in San Francisco and Berkeley and have seen What’s Up Doc? and American Graffiti, so you noticed the mistakes you mentioned. For someone who hasn’t done those things, I can’t see how the mistakes could possibly affect their enjoyment of the books and movies at issue, or how one can even expect them to realize the mistakes exist. We were discussing this topic on the phone yesterday and you mentioned an example in which you were reading a historical book and noticed a word that you thought might be an anachronism, so you looked it up and found out that it was. But even in that case, you knew enough about the word to know that it could be anachronistic, and you (as a librarian) knew where to go to find out the truth. Many readers would just fly by the word and never think twice about it.
None of this is to deny that the mistakes are real, or that because its fiction, anything goes, because I don’t really believe that either, but I do think we need to keep some perspective. I tend to get very self-satisfied when I notice an error in a book (fiction or nonfiction), never really considering how many hundreds (thousands?) of errors I’ve missed over the years. And then there are cases when I think I’ve spotted a mistake only to look it up and find out, nope – the author had it right. So I think we need to fight against that self-satisfaction. Books go through multiple print runs and editions and mistakes get corrected (just see the comments to your post Backmatter for an example), and the vast majority of readers aren’t going to notice or be affected by most of these mistakes in any case.
So do they never matter? Well, no I wouldn’t say that either. I think you were onto something when you said that Doctorow’s geographical errors weren’t deal breakers because they didn’t affect “the main points of the book.” And therefore I actually think that the One Crazy Summer error gives us a perfect example of an error that really does affect the main points of the book, because the geographical error was so ingrained in the character development that it could not be fixed in a later printing. Now, that particular error is somewhat small – a hill where there isn’t one. But I think it is the kind of inaccuracy that we should be on the lookout for. Tashjian could have used different movies (or the reader could even supply her own); Doctorow could delete the word “the” and change 45 minutes to 145 minutes; but Williams-Garcia would have to create an entirely new metaphor for her characters, and that seems like its important. Thoughts?