W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > January to April 1999

Re: terminology question

From: Roy T. Fielding <fielding@kiwi.ics.uci.edu>
Date: Sun, 11 Apr 1999 14:55:29 -0700
To: Koen Holtman <koen@win.tue.nl>
Cc: jschroeder@becomsys.de, http-wg@hplb.hpl.hp.com
Message-Id: <9904111455.aa24759@paris.ics.uci.edu>
>It is generally not a good idea to redefine terms, even if the
>original definition is somehow wrong.  Having two incompatible
>definitions of 'variant' around can only lead to confusion.  If you
>have some interesting concept you want to define a term for, there are
>plenty of other, uncontaminated, words you could pick.

I agree.  From the Merriam Webster online dictionary:

    variant

    noun
    Date: 1848

    : one of two or more persons or things exhibiting
    usually slight differences: as a : one that exhibits
    variation from a type or norm b : one of two or
    more different spellings (as labor and labour) or
    pronunciations (as of economics \ek-, Ek-\) of the
    same word c : one of two or more words (as
    geographic and geographical) or word elements
    (as mon- and mono-) of essentially the same
    meaning differing only in the presence or absence
    of an affix 

It makes absolutely no sense whasoever to use the term variant to
describe a singleton representation.  The HTTP spec does so because
somebody wanted a shortcut and didn't like the word "representation".
The HTTP spec only needs to be self-consistent, so this is not a
technical error -- it is just confusing.  I refuse to allow that confusion
to propagate into anything else I write about the Web, including e-mail.

....Roy
Received on Sunday, 11 April 1999 23:00:44 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:33:30 EDT