Begin forwarded message:
From: "David P. Reed" <dpreed@reed.com>
Date: June 25, 2009 8:50:30 AM EDT
To: dave@farber.net
Cc: ip <ip@v2.listbox.com>, Christopher Parsons <parsons@uvic.ca>, Lauren Weinstein <lauren@vortex.com>
Subject: Re: [IP] interesting notes -- Western DPI firms helping Iran government spy on Internet traffic
I also don't know what the Iranian ISPs are doing or not doing. But, in reading Parson's interesting blog post, I found it quite interesting that the core of his skepticism surrounds an extremely narrow definition of a "DPI appliance" as a device that
1. is capable of separating packets into distinct "streams" (which is not a precisely defined term, but I suspect he means "end-to-end virtual circuit" perhaps or a different concept used by some in the engineering community: "flow").
2. looks only at a few packets at the "beginning of a stream" and applies some "rules".
This would seem to omit boxes that, for example, record for later analysis elements of packet contents, shipping those to other points in the network. Such boxes, made by companies like Narus, exist. If Parsons' purpose is to create a distinction of "OK", "bad" and "evil" where DPI is limited to "OK" things, perhaps he is just "negotiating definitions".
We have seen this with "P2P" rhetoric - by those who deliberately equate "P2P" with copyright piracy, or perhaps with "using cable modems in ways that don't let cable providers "0wn your eyeballs".
WSJ may also be guilty of overstating the facts and capabilities. Though to be honest, the capabilities of nation-states to deploy large-scale tapping and analysis of data has been underestimated in the press, and kept secret by governments.
I'd call for engineering analysis, open discussion, etc.
On 06/25/2009 05:58 AM, David Farber wrote:
1. is capable of separating packets into distinct "streams" (which is not a precisely defined term, but I suspect he means "end-to-end virtual circuit" perhaps or a different concept used by some in the engineering community: "flow").
2. looks only at a few packets at the "beginning of a stream" and applies some "rules".
This would seem to omit boxes that, for example, record for later analysis elements of packet contents, shipping those to other points in the network. Such boxes, made by companies like Narus, exist. If Parsons' purpose is to create a distinction of "OK", "bad" and "evil" where DPI is limited to "OK" things, perhaps he is just "negotiating definitions".
We have seen this with "P2P" rhetoric - by those who deliberately equate "P2P" with copyright piracy, or perhaps with "using cable modems in ways that don't let cable providers "0wn your eyeballs".
WSJ may also be guilty of overstating the facts and capabilities. Though to be honest, the capabilities of nation-states to deploy large-scale tapping and analysis of data has been underestimated in the press, and kept secret by governments.
I'd call for engineering analysis, open discussion, etc.
On 06/25/2009 05:58 AM, David Farber wrote:
Begin forwarded message:
From: Lauren Weinstein <lauren@vortex.com>
Date: June 24, 2009 8:53:56 PM EDT
To: nnsquad@nnsquad.org
Subject: [ NNSquad ] Re: Western DPI firms helping Iran government spy on Internet traffic
[ Forwarded with permission -- Lauren ]
----- Forwarded message from Christopher Parsons <parsons@uvic.ca> -----
Date: Wed, 24 Jun 2009 14:29:13 -0700
From: Christopher Parsons <parsons@uvic.ca>
Subject: Re: [ NNSquad ] Western DPI firms helping Iran government spy on
Internet traffic
To: Lauren Weinstein <lauren@vortex.com>
Hi Lauren,
I have some serious doubts that the WSJ is accurate in their depiction of
DPI. I'm doing my doctoral studies on DPI as it relates to privacy, and
neither I nor the network engineers that I have communicated with (who are
using DPI appliances) are aware of ANY DPI appliance that is actually
capable of doing what the WSJ is claiming is going on. I've written about
this, informally, here:
http://www.christopher-parsons.com/blog/politics/iran-traffic-analysis-and-deep-packet-inspection/
Cheers,
Chris
******************************************
Christopher Parsons
Doctoral Student
Political Science, University of Victoria
http://www.christopher-parsons.com
*******************************************
On 21-Jun-09, at 9:41 PM, Lauren Weinstein wrote:
-- From the "Anything for a buck" department --
Western DPI firms helping Iran government spy on Internet traffic
http://online.wsj.com/article/SB124562668777335653.html
--Lauren--
NNSquad Moderator
----- End forwarded message -----
-------------------------------------------
Archives: http://www.listbox.com