What do you do about wrong classifications?

I'm not talking about cases where a particular source has the "wrong" classification for a book -- I'm talking about cases where the classification itself is messed up.

I recently read a book about Unitarian Universalism, which is classified in BX, Christian Denominations. (It's what the in-book classification system is, and it's where Unitarianism is put in the official online LC system. So it's not "misclassified" in that sense. The problem is that UU is not a Christian denomination, though Unitarianism was when the classification was developed.

So do I get to count this? I mean, I've read a book in BX. I haven't read a book on Christian denominations (and I'm not likely to, if this doesn't count), which is what BX is supposed to cover.

I found a few weirdnesses in both systems. In LC, I think Color: A Natural History of the Palette was classified under "physics" when it really doesn't have much to do with science and would be better placed in one of the N subclassifications (the one for painting, which I don't remember off the top of my head).

Don't remember what I did about it, and can't look it up now as I just got pestered by my day job.

>2 AnnaClaire:
Well, the Library of Congress itself classified Color: A Natural History of the Palette as ND1488 .F56 2003. I wonder who put it in physics. Oh well.

See, this isn't really the same situation as 2/3; I've seen plenty of books where one library or another put it in a crazy place. This is in the "right" place according to the classification system, it's just that the classification system is wrong. It would be like finding that the LC system classified mushrooms as a subset of plants -- having a mushroom book under "plants" wouldn't be misclassified, just factually wrong, and it would feel weird to me to count a mushroom book as fulfilling the plant classification. That's what's going on here.

"Color" is one of my standard examples for explaining the difficulties of creating a universal non-faceted taxonomy. LC spreads books on color throughout the classification: Color as it relates to Art/Psychology/Philosophy/Marketing/Biology/Physics/Computer Graphics (just off the top of my head). That is a reasonable way for LC to divide the world up. Many libraries, however, have found that their audience is better served by pulling together a section on color, particularly if a high percentage of their books fall in one bucket. I would expect many Science libraries would consider placing the work in #2 into physics.

#1> While I don't think that, today, the UU's would claim to be a Christian denomination and I think that most Christian denominations would reject the UU's, the separation between them has not always been as distinct. The historical relationship is captured by the LC Classification.

Personally, I would count it. The challenge as I understand it is to read from each section and not necessarily an overview or representative sample from that classification. That is certainly how I've been counting.

Lorax, do you have a link to a list of what's topics are included under each category? E, F, and K and broken out very specifically on the wikipedia link, but I'm very curious now about what things are supposed to go in the other categories (other than the obvious).

In any event I agree with dcmdale that you should count the book. I think the category descriptions may not always be 100% accurate in part because the headings are supposed to be short and also because the system isn't that flexible.