From: Phil Karlton <karlton@netscape.com>
To: Nathaniel Borenstein <nsb@nsb.fv.com>
Message Hash: 463de07440e0f78c5f3dae8a38ff243360d336108c88fef54cf4c462cb2e719a
Message ID: <310F13E7.13AA@netscape.com>
Reply To: <c=US%a=_%p=msft%l=RED-66-MSG960129190324HH007C00@red-02-imc.itg.microsoft.com>
UTC Datetime: 1996-02-08 02:16:35 UTC
Raw Date: Thu, 8 Feb 1996 10:16:35 +0800
From: Phil Karlton <karlton@netscape.com>
Date: Thu, 8 Feb 1996 10:16:35 +0800
To: Nathaniel Borenstein <nsb@nsb.fv.com>
Subject: Re: FV's Borenstein discovers keystroke capture programs! (pictures at 11!)
In-Reply-To: <c=US%a=_%p=msft%l=RED-66-MSG960129190324HH007C00@red-02-imc.itg.microsoft.com>
Message-ID: <310F13E7.13AA@netscape.com>
MIME-Version: 1.0
Content-Type: text/plain
Nathaniel Borenstein wrote:
> We have a few pages of C code that scan everything you type on a
> keyboard, and selects only the credit card numbers. How easy is that to
> do with credit card numbers spoken over a telephone?
>
> The key is large-scale automated attacks, not one-time interceptions.
This fact that the filtering can be done on the client side is nearly
irrelevant. Most people do not hit enough keystrokes in a day to prevent sending
the entire keyboard stream back to the filtering agent.
Since most folks do not spend most of their time typing in nonsense phrase, you
could probably pick out the First Virtual account number also. With only a
little more cleverness you can get the file containing private keys. With a few
thousand tries through the stream you can decrypt that file using the user's
pass phrase.
If you have the ability to change the software on the user's machine to
something arbitrary, why bother stopping at something as "trivial" as a single
credit card number.
PK
--
Philip L. Karlton karlton@netscape.com
Principal Curmudgeon http://www.netscape.com/people/karlton
Netscape Communications Corporation
Return to February 1996
Return to “Tim Oerting <timo@microsoft.com>”