Re: [COAS-List] SGML Thread on Report Types

[Wayne Wilson <wwilson@umich.edu>]

It would seem like the lesson here (as futilely noted by one
participant) is that the attempt to exhaustively list types has dubious
value.  This problem space has been investigated so many times with
similar results.  Attempts to standardize on a particular 'slice'
through an information space only lead to frustrations and ugly
implementations that have to find away around the arbitrary
classifications.  There will be as many classifications as there are
reasons for classifying.  Billing is a good reason for billing purposes
but maybe not for the in-person patient care, nor for the public health
investigation, und so on.