Loading ...
Sorry, an error occurred while loading the content.

4280Re: [TaxoCoP] Re: Taxonomy Executive Summary for Executives / Senior Managers

Expand Messages
  • John O'Gorman
    Feb 6, 2012
    • 0 Attachment
      Thanks everyone for the very informative responses to my questions.
      Claude, I especially like your answer to the Enterprise Taxonomy question.
      John O'
      -----Original Message-----
      From: Claude [mailto:cbaudoin@...]
      Sent: Sunday, February 5, 2012 11:14 AM
      To: TaxoCoP@yahoogroups.com
      Subject: [TaxoCoP] Re: Taxonomy Executive Summary for Executives / Senior Managers


      > On Feb 2, 2012, at 1:08 AM, John O'Gorman wrote:
      > 1. The presentations I looked at inferred that a controlled
      > vocabulary is a requisite part of a taxonomy... is that true?

      I don't claim to be nearly as knowledgeable as Patrick, but I have found it useful to follow the definitions in Z39.19, in which a "controlled vocabulary" is the more generic term that can designate anyone of four successively more sophisticated structures:
      - a (flat) list of terms
      - a "synonym ring" (weird term, but basically it just adds the
      notion of synonyms to a list)
      - a taxonomy (same as previous, PLUS hierarchy)
      - a thesaurus (same as previous, PLUS other semantic relationships)

      > 2. Is it possible to have more than one 'Enterprise' taxonomy?

      Contrary to other respondents, I tend to think that there should be only one enterprise taxonomy, but it contains multiple facets. Any particular application or system that consumes the taxonomy may expose to the user all facets, or just one facet, or even just a branch of one facet, and moreover may present just one level of decomposition of a term (e.g., for a pulldown selection menu) or multiple levels (e.g., for a "marching menu").

      This may be an oversimplification, but when I say this to various stakeholders (CMS managers, community leaders, MarCom people, etc.) they seem to understand this, and can then contribute valuable information in return about what their domain of discourse is.

      Claude Baudoin

    • Show all 15 messages in this topic