Re: [Corpora-List] Constitution

From: John F. Sowa (sowa@bestweb.net)
Date: Sun May 15 2005 - 15:03:37 MET DST

  • Next message: Detmar Meurers: "Re: [Corpora-List] Constitution"

    Adam,

    These issues are typical of a large number of applications.
    The current discussion is about EU documents, but very
    similar issues arise in commercial applications.

    For example, a multinational company X, which has its roots
    in country A, will typically have most documents written in
    the language of A. Large numbers of new terms that refer to
    X's products and their novel features will be coined in terms
    of the A lexicon, and corresponding terms will have to be
    coined in each of the languages for each of the markets
    to which X sells their products. Furthermore, the product
    developers will constantly be extending the terminology,
    coining new terms, and modifying the meanings of old terms
    even while terminologists are trying to define equivalents.

    Replace the term "EU" with "market" and the same problems
    exist in commercial applications:

    > I'm not sure that corpus methods are very relevant.
    > For new and newish EU languages, the corpora don't exist
    > yet or where they do, are likely to be full of
    > inconsistencies and not dependable.

    The denotation of the term "corpus methods" changes with
    every new method that anyone invents. If the current methods
    can't handle problems of this sort, that's an important
    challenge for researchers to develop new ones.

    John Sowa



    This archive was generated by hypermail 2b29 : Sun May 15 2005 - 15:40:36 MET DST