X-Message-Number: 26711
Date: 28 Jul 2005 04:22:16 -0000
From: "Kevin Q. Brown" <>
Subject: Re: reputation too low

Charles Platt expressed surprise that the CryoNet software
found the reputation of "tru_numberman" to be too low for
inclusion of his/her message # 26699 in a CryoNet digest.
He also pointed out that documentation on the CryoNet rating
system was not easy to find.

To remedy the latter issue, I just added a new "MESSAGE RATINGS" 
section to the "CryoNet Information for New Subscribers" at:
  http://www.cryonet.org/cgi-bin/dsp.cgi?msg=0
which now contains links to messages that provide more details
about the rating system.

Next, I found why the CryoNet software considered the reputation
of "tru_numberman" too low for inclusion in a CryoNet digest.
From the main CryoNet page:
    http://www.cryonet.org/
I clicked on the link for "CryoNet Messages by Author"
to find the three messages (26676, 26685, and 26687)
posted by "tru_numberman" before # 26699.
Entering those numbers at:
  http://www.cryonet.org/cgi-bin/rate.cgi
revealed that 26676 and 26685 had no ratings, but 26687
had one rating of "Flamebait", giving "tru_numberman" an
overall rating of "Flamebait", which was too low for
inclusion in a CryoNet digest.

I expect that Charles would agree that the algorithm is too
sensitive to a single unflattering rating.  He proposed
requiring at least 20 ratings be given before a poster's
reputation could be considered too low.  That seems too
generous to me, but a threshold of one rating clearly is too low.
I'll change it to five.

    Kevin Q. Brown
     (please include "cryonics" in the subject line)

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=26711