From: Harris Cohen <hlc2105@columbia.edu>
  To  : <cpc@emoglen.law.columbia.edu>
  Date: Thu, 9 Mar 2006 13:21:26 -0500

Re: Data mining, security and terrorism

------=_Part_5999_17969637.1141928486225
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Schneier's point is what I was had on my mind last week when I asked what
the NSA was comparing this vast store of data against in the hopes of
uncovering terrorist plots.

Instinctively, it seems like it would be nearly impossible for investigator=
s
to build a reliable model of "A Terrorist's Data Stream" based on the
activity leading up to past attacks, since most talking heads I've seen
agree that terrorists are exceedingly unlikely to use the same tactics
twice, and there is very little formal structure to their operations.

What seems a great deal more likely is that the NSA will use proxies for
"Terrorist" that might be easier to pick out of a data mine, but are
constitutionally, ethically, and practically problematic, like
"Fundamentalist Muslim," "Protest Organizer," "Recent Immigrant," etc. Not
only are these likely to be inappropriate bases for the application of law
enforcement scrutiny, but ---since they're based on stereotypes and not rea=
l
intelligence---they seem likely to overlook real threats and waste agents'
time.

On 3/9/06, Alexander van der Wolk <av2139@columbia.edu> wrote:
>
>
> Following up on class discussions and in light of Congress' renewal
> of the Patriot Act:
>
> "Intense congressional and public scrutiny has not produced a single
> substantiated claim that the Patriot Act has been misused to violate
> Americans' civil liberties," said House Judiciary Committee Chairman
> James Sensenbrenner, R-Wisconsin. "Opponents of the legislation have
> relied upon exaggeration and hyperbole to distort a demonstrated
> record of accomplishment and success."
> You gotta love politics...
> http://www.wired.com/news/wireservice/0,70362-0.html?tw=3Drss.politics
>
> An interesting op-ed in Wired today about why data mining will not
> stop terrorism:
>
> "We're not trading privacy for security; we're giving up privacy and
> getting no security in return."
>
> http://www.wired.com/news/columns/0,70357-0.html?tw=3Drss.politics
>
>
>
> Alex
>
> -----------------------------------------------------------------
> Computers, Privacy, and the Constitution mailing list
>
>

------=_Part_5999_17969637.1141928486225
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Schneier's point is what I was had on my mind last week when I asked what t=
he NSA was comparing this vast store of data against in the hopes of uncove=
ring terrorist plots.<br><br>Instinctively, it seems like it would be nearl=
y impossible for investigators to build a reliable model of "A Terrori=
st's Data Stream" based on the activity leading up to past attacks, si=
nce most talking heads I've seen agree that terrorists are exceedingly unli=
kely to use the same tactics twice, and there is very little formal structu=
re to their operations.
<br><br>What seems a great deal more likely is that the NSA will use proxie=
s for "Terrorist" that might be easier to pick out of a data mine=
, but are constitutionally, ethically, and practically problematic, like &q=
uot;Fundamentalist Muslim," "Protest Organizer," "Recen=
t Immigrant," etc. Not only are these likely to be inappropriate bases=
 for the application of law enforcement scrutiny, but ---since they're base=
d on stereotypes and not real intelligence---they seem likely to overlook r=
eal threats and waste agents' time.
<br><div><span class=3D"gmail_quote"><br>On 3/9/06, <b class=3D"gmail_sende=
rname">Alexander van der Wolk</b> <<a href=3D"mailto:av2139@columbia.edu=
">av2139@columbia.edu</a>> wrote:</span><blockquote class=3D"gmail_quote=
" style=3D"border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0=
.8ex; padding-left: 1ex;">
<br>Following up on class discussions and in light of Congress' renewal<br>=
of the Patriot Act:<br><br>"Intense congressional and public scrutiny =
has not produced a single<br>substantiated claim that the Patriot Act has b=
een misused to violate
<br>Americans' civil liberties," said House Judiciary Committee Chairm=
an<br>James Sensenbrenner, R-Wisconsin. "Opponents of the legislation =
have<br>relied upon exaggeration and hyperbole to distort a demonstrated
<br>record of accomplishment and success."<br>You gotta love politics.=
..<br><a href=3D"http://www.wired.com/news/wireservice/0,70362-0.html?tw=3D=
rss.politics">http://www.wired.com/news/wireservice/0,70362-0.html?tw=3Drss=
.politics
</a><br><br>An interesting op-ed in Wired today about why data mining will =
not<br>stop terrorism:<br><br>"We're not trading privacy for security;=
 we're giving up privacy and<br>getting no security in return."<br>
<br><a href=3D"http://www.wired.com/news/columns/0,70357-0.html?tw=3Drss.po=
litics">http://www.wired.com/news/columns/0,70357-0.html?tw=3Drss.politics<=
/a><br><br><br><br>Alex<br><br>--------------------------------------------=
---------------------
<br>Computers, Privacy, and the Constitution mailing list<br><br></blockquo=
te></div><br>

------=_Part_5999_17969637.1141928486225--

-----------------------------------------------------------------
Computers, Privacy, and the Constitution mailing list