[Sigia-l] Re: first principals (bias in cataloging & classification)

[Peter VanDijck]

Karl Fast wrote:

> >> 7. There is no such thing as an unbiased IA. But we have to try
> > No - often (if not always) the value lies in the bias.
>   Can we be unbiased in classification and resource description?

Nope. No way you can classify without bias.

> * No bias is often possible:
> A horse is a horse is a horse, of course. And a horse is not a
>   cow. What I mean is that sometimes the way we describe something
>   is clear and unambiguous (but not always).

Nope. I know there are levels of ambiguity and such, but even a horse is
not always a horse. How about a mule? Or a new, genetically engineered
cross between a horse and a cow? How about a tribe somewhere that doesn't
really distinguish between horses and cows. It's possible they exist. How
about a dog: would they distinguish between horses and cows? We classify
things because of who we are.

Especially in the domain of classifying animals and plants, there is a lot
of bias and fuzzyness. Historically, things were classified mostly by
appearance. Then analyzing the genetic lines of animals became possible,
and lo and behold, the existing classification schemes turned out to be
incompatible with a genetic scheme.

I guess the point is this: there are levels of bias but it is pretty much
correct to say there is *always* some bias in classification. For example,
something seemingly basic as colour classification is biased. Different
cultures have different colours. All have black and white. But for any
other colour it depends: some cultures recognize blue, others don't.
Lakoff has a good description of this in one of his books (I forget which
one)

I do feel I need to clear up my thinking about this a bit, so keep the
discussion coming.

Cheers,
Peter