[Sigia-l] RE: Core set of classification terms

[Christopher G. Fox]

One "simple" alternate approach to the problem below would be to
introduce the concept of "scope" into a faceted classification, so that
a query for "blue" things would return both "blue" and "black" if the
scope of the query were Viking or Irish cultural contexts or,
conversely, a query for "black" things also included those things with a
value of "blue" when "blue" is scoped to Viking or African.  

The difference between this approach and the notion of core terms would
be that some values are more likely to be universal (as well as more
pragmatic) than others. In principle, everyone at an airport needs to
distinguish between the Arrival and Departure halls. A fundamental
"pidgin" set of classifications could certainly emerge from the global
practice of information architecture (taken broadly). 

The slippery slope is that eventually you end up positing a universal
mental model of human thought. Not likely to work as well on concepts
with any significant degree of cultural valence. But, anything beyond a
simple set of universals would need either the faceting I describe
above, or the equivalent. <understatement> Which sounds to me like a lot
of work </understatement>.

I suppose it turns back to the question "how much classification is
enough?" with a provisional answer being " as much as you need to
support the finding/browsing/querying/etc. retrieval requirements of
your users (some of which may be not be human, e.g., agents that need to
federate query results across systems.

By the way, I think as a former lurker I'm supposed to introduce myself.
Thanks to you all for the ongoing and interesting conversations.

Christopher G. Fox
Director, Experience Design and Branding
Logical Design Solutions, Inc.

>Date: Mon, 03 Feb 2003 16:47:17 +0000
>To: sigia-l at asis.org
>From: Gerry McGovern <gerry at gerrymcgovern.com>
>Subject: [Sigia-l] Core set of classification terms

Jon Hanna wrote:

> > A new standard for wireless is called "Bluetooth" after a Danish
Viking
> > king. He was called Bluetooth because he had a black tooth. If you
don't
> > believe me, I can ad that the Vikings called the Africans
> > "Bluemen"! So they
> > actually did not distinguish between blue and black. Why not?
Perhaps
> > because blue was a very rare color before artificial colors were
> > invented by
> > the French and the Germans, so you did not have much need for this
> > distinction.
>
>I was going to reply to Gerry's post saying that the problem isn't so
much
>with classification not scaling as with classification not translating.
This
>gives me a perfect example though.
>
>In Irish "blue" is also used for people with black skin. Only it isn't
>really, "gorm" is used and that isn't an exact translation of "blue".
>(Incidentally in Irish people would be called "red", "fair" or "black"
on
>the basis of their hair colour, hence "dubh" which is a closer
translation
>of "black" signifies a white person with black hair).

>The above is an interesting point. There can definitely be problems
with 
>classifications translating. However, I feel that the Internet is a
strong 
>driver in the internationalization of language (the Englishization, to
be 
>more precise.)

 

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.449 / Virus Database: 251 - Release Date: 1/27/2003