Re: HIDDEN-USER

Mike StJohns <stjohns@umd5.umd.edu> Thu, 13 August 1992 22:53 UTC

Received: from ietf.nri.reston.va.us by IETF.NRI.Reston.VA.US id aa01483; 13 Aug 92 18:53 EDT
Received: from NRI.NRI.Reston.Va.US by IETF.NRI.Reston.VA.US id aa01479; 13 Aug 92 18:53 EDT
Received: from ietf.NRI.Reston.Va.US by NRI.Reston.VA.US id aa20969; 13 Aug 92 18:54 EDT
Received: from ietf.nri.reston.va.us by IETF.NRI.Reston.VA.US id aa01474; 13 Aug 92 18:53 EDT
Received: from NRI.NRI.Reston.Va.US by IETF.NRI.Reston.VA.US id aa01470; 13 Aug 92 18:53 EDT
Received: from umd5.umd.edu by NRI.Reston.VA.US id aa20964; 13 Aug 92 18:54 EDT
Received: by umd5.umd.edu id <AA24643@umd5.umd.edu>; Thu, 13 Aug 92 18:54:28 -0400
Message-Id: <9208132254.AA24643@umd5.umd.edu>
To: dupuy@cs.columbia.edu
Cc: ident@NRI.Reston.VA.US
Subject: Re: HIDDEN-USER
In-Reply-To: Your message of Thu, 13 Aug 92 18:11:43 -0400. <9208132211.AA12009@tiemann.cs.columbia.edu>
Date: Thu, 13 Aug 1992 18:54:25 -0400
From: Mike StJohns <stjohns@umd5.umd.edu>
X-Mts: smtp

I was hoping not to have to get into the middle of this. 

An implementor can choose not to provide HIDDEN-USER style
functionality in his implementation, or he can provide an #ifdef or
command line option to disable this functionality.  In the end, its
the choice of the system administrator whether or not he enables said
functionality.  

There is no requirement in the ident document to implement a
HIDDEN-USER capability, but some implementors (and some system
administrators) want such a capability.  I see no merit in removing
HIDDEN-USER from the draft especially as an implementor could return
"XHIDDEN-USER" and still be in compliance.  There is no benefit to
removing it and some benefit (one standard error code) to keeping it,
so it stays.

WRT to privacy/anonymity issues,  implementors may want to have the
server log the user information as well as the port identification
information for later use in auditing if they refuse to transmit that
information due to a user's HIDDEN-USER request.


Mike