Subject: Re: Knowledge classification systems
From: Erik Naggum <erik@naggum.no>
Date: 17 Sep 2002 21:56:30 +0000
Newsgroups: comp.lang.lisp
Message-ID: <3241288590332494@naggum.no>

* Don Geddis
| Does this mean that a given Dewey classification encodes multiple attributes?

  Yes.  Geography and audience are particularly well-utilized examples.  E.g.,
  a book about Indiana high-school women's basketball would be classified
  under sports, basketball, pre-college, in Indiana, for women.  A book about
  cats for children would be classified under cats, intended for children.

| Please correct me if I'm wrong, but I assume that the Dewey system (and the
| others you've mentioned) are designed for physical objects like books, where
| each item belongs in exactly one place.

  Yes, the books would be in one place, but you would obviously have multiple
  index cards for them if they were cross-classified.  It is useful to find
  related items grouped in the bookshelves, but this is a consequence of the
  nature of bookshelves, not of the classification.

| With online systems, it would seem much better to label each data item with
| a whole set of applicable features, and then basically navigate through the
| hierarchy by doing searches and building virtual indexes.  In such a scheme,
| a given object might appear at numerous "leafs" in the classification tree,
| rather than only one.

  This is already the case.  I have several books that are cross-classified.

| Do Dewey systems already allow for this somehow?  Or do you think the idea
| that an item might not have a unique classification is misguided?

  I think you have assumed that a classified item has only one classification.
  This is false.  It must have one, and the better it is, the more useful that
  one is, but if a book covers some more than subject, there is nothing to
  stop it from getting several classifications.  (Unless, of course, it is a
  "general collection", where the classifiers basically give up on it with the
  assumption that people would not look for it under any specific area.

-- 
Erik Naggum, Oslo, Norway

Act from reason, and failure makes you rethink and study harder.
Act from faith, and failure makes you blame someone and push harder.