1996-02-16 - Re: Computer unmasks Anonymous writer…

Header Data

From: sasha1@netcom.com (Alexander Chislenko)
To: pcw@access.digex.net
Message Hash: 4fe2d1d566ac06ec9d59fd1174c38e0614ed7ed664cc8066e85580a86734d923
Message ID: <199602161720.JAA05092@netcom.netcom.com>
Reply To: N/A
UTC Datetime: 1996-02-16 20:49:07 UTC
Raw Date: Sat, 17 Feb 1996 04:49:07 +0800

Raw message

From: sasha1@netcom.com (Alexander Chislenko)
Date: Sat, 17 Feb 1996 04:49:07 +0800
To: pcw@access.digex.net
Subject: Re:  Computer unmasks Anonymous writer...
Message-ID: <199602161720.JAA05092@netcom.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain


  I ran my essays through Word grammar checker a while ago,
and was surprised how stable the grammar statistics were.
Complexity of the text (grade level) was the same to the decimal point,
average length of sentences was consistent, etc.
People also use the same styles of smileys or *highlights*, make
consistent spelling errors, have their habits of indentation, etc.

My suggestion at the time was to have randomizing output filter that
would substitute synonyms, change spelling, modify paragraph formatting,
etc.  - Style anonymizer, I'd call it.  Also, if small random changes are
applied to every copy of the message you send out, and you keep track of
what recipient got which version, you will find it easier to identify
the leaks if you send private messages.  A dishonest recipient, to
protect himself, would have to further randomize the message, which
has apparent consequences for the ease of your identification and attributability of the message.

  Of course there are  more subtle consistencies of style one can't easily
mask with a filter - but they are not easy to detect either.

-----------------------------------------------------------
| Alexander Chislenko | sasha1@netcom.com | Cambridge, MA | 
| Home page:  http://www.lucifer.com/~sasha/home.html     |
-----------------------------------------------------------





Thread