SueDNym messages and our friend

Arsen Ray Arachelian rarachel at prism.poly.edu
Wed Aug 3 21:54:19 PDT 1994


This is the output of the program Medusa's Tentacles.  This isn't
complete.  The 1st "level" is buggy or I'm doing something wrong in the
data gathering.  I will however fix this up later.  This is basically a
sneak previous at what Medusa produced:

SOURCE TEXT: det30
FUZZYBASE  : det
1: 6136990.06793633 deg of closeness  [TOKEN TOUPLES]
2: 0.00025409 deg of difference [TOKEN FREQ   ]


SOURCE TEXT: suednym
FUZZYBASE  : det
1: 425581.03653014 deg of closeness  [TOKEN TOUPLES]
2: 0.00021086 deg of difference [TOKEN FREQ   ]


SOURCE TEXT: tcmay.txt
FUZZYBASE  : det
1: 264235.00574214 deg of closeness  [TOKEN TOUPLES]
2: 0.00016356 deg of difference [TOKEN FREQ   ]

SOURCE TEXT: c:\autoexec.bat
FUZZYBASE  : det
1: 14771672.69300460 deg of closeness  [TOKEN TOUPLES]
2: 0.00035529 deg of difference [TOKEN FREQ   ]


SOURCE TEXT: c:\config.sys
FUZZYBASE  : det
1: 10857800.74279867 deg of closeness  [TOKEN TOUPLES]
2: 0.00035317 deg of difference [TOKEN FREQ   ]


Again, please ignore the 1: stats for now as I have to further investigate
their validity.

Fuzzybase "det" is the collected statistics based on the detweiler
archives collected by Rishab Aiyer Ghosh.

det30 is a portion of that collection, so I'm comparing it with the
collection to see how different a small portion will be when comparing it
to the whole.  This gives a difference level of 0.00025409

suednym is one of the messages Sue sent.  NOTICE that the level of
difference is 0.00021086.  These two "differences" are very very close
when compared to the others.

Autoexec and config.sys are thrown in as sanity checks.  If I got low
numbers on these I'd expect something was terribly wrong.

Weirdly enough TC May's post about the bleak future which includes tons of
paranoid like words matches most of Detweiler's posts more closely than
one of his own messages.   However lets not jump to conclusions here
because Detweiler's own message showed a differences of 0.002... and Sue
matched it.

With this sort of data it's hard to separate the subjects from the authors
from the language.  Language is easy to detect, but two people who
constantly post the same kind of material based on the same subject might
show up with the same difference levels.

Again this is only ONE sample so please no lynching Sue just yet (although
this was already a moot point.)

Remember, Medusa is still under construction.  This is only a preview.

If you'd like to preview the sources/executables to medusa please email me
and I'll somehow get them to you.

===============================================================================
| + ^ + || ' . . .   .   . .   .             Ray  (Arsen)  Arachelian        ||
|  \|/  ||   . . .  ' . ' .  : . .           rarachel at photon.poly.edu        ||
|<--+-->||.   . . |' '| .' .  .  ...    ___  sunder at intercom.com             ||
|  /|\  ||   .  . \___/ .  . . : .... __[R]                                  ||
| + v + || . oOOo /o.O\ oOOo :. : ..    |A| "And bugs to kill before I sleep"||
=========/---vvvv-------VVVV------------|I|----------------------------------/
        /      .    :   . ' : '         |D|     This signature pannel is    /
       /  The Next Bug to kill(tm)      ---     now open.                  / 
      /___________________________________________________________________/








More information about the cypherpunks-legacy mailing list