To: drums@cs.utk.edu
Subject: Re: 821bis outstanding issues list 
In-Reply-To: Your message of "28 May 1996 20:08:58 GMT."
             <19960528200858.12974.qmail@koobera.math.uic.edu> 
Date: Wed, 29 May 1996 06:32:46 +1000
Message-Id: <1964.833315566@munnari.OZ.AU>
From: Robert Elz <kre@munnari.oz.au>

    Date:        28 May 1996 20:08:58 -0000
    From:        djb@koobera.math.uic.edu (D. J. Bernstein)
    Message-ID:  <19960528200858.12974.qmail@koobera.math.uic.edu>

    The presence of a configurability requirement in a de jure 
    standard is not going to move users from apathy to action.

For most users this is certainly true - not for all though, there
are every now and then, just a few who are interested.

    The only reason a user asks that question is because he _wants_
    the feature.

Not true at all.   There are users who ask questions just because
they like asking questions.   Some simply present the boiler plate
"here is the standard, do you comply?"   Others because they know
there is something the manufacturer doesn't implement, and by using
the "no" answer they can apply some bargaining leverage.   And some
just because we're bastards and like to see the manufacturers (or
their agents) work to make a sale (it also gives us some points
later if the product should have some defect, if the manufacturer
failed to accurately answer one of the questions).

kre