[some at list: Questions about the illegal wiretaps]

Eugen Leitl eugen at leitl.org
Thu Feb 16 01:09:38 PST 2006


[from some at list]

More answers:

Andrew A. G**** wrote:
>They *can*, but their typical infrastructure is not set up to handle
>this.  A system like this isn't maintenance-free.  This list should know
>that better than most.  Besides, the intel has to output to a human
>eventually.  You can't just have a machine intercept a call, determine
>that it's a threat and send out a Predator drone to kill the threat
>without any human intervention.

No, but NSA's job is just the first part of that.  They don't target
drones, or anything like that.

>The FBI has said that they were swamped.  How many man-hours were
>involved in this in the FBI side?

yes, this is definitely a cost.  And the FBI finally said "enough"!  I
kinda wonder if this was due to miscommunication, and the FBI thought
they were getting finished analytical product.

>>> 3. According to the Washington Post, we had a backlog of
>>>    hundreds of thousands of hours of untranslated wiretaps. Do
>>>    we still have this backlog?
>>Yes.
>>
>>> 4. What's the point of conducting wiretaps when we can't
>>>    translate them?
>>None.
>
>That's pretty damning.

This is how the Intelligence Community works.  Collection is considered
the A#1 task, and they are inveterate packrats.  It's far easier to fill
hard drives with junk than to make active use of it.  Sad, but true.

>>> 8. Why didn't you ask for more translators, knowing that FBI
>>>    Director Mueller asked that all urgent translations be done
>>>    in 12 hours?
>>The unfortunate issue is not more translators, but more TS/SCI-cleared
>>translators with an active Counterterrorism- and lifestyle-poly who
>>are fluent in Arabic at level 4+ (nb:  this is a VERY small set of
>>individuals)
>
>This program's been running for something like three years.  In that
>time, couldn't that money have been used to encourage people to be
>trained for this, and then actually train them?

Well, realistically:  no, probably not.  The US is grieviously far
behind in this.  There is little or no chance in three years that
someone who can pass the security screen can achieve the necessary
language proficiency level in say, Dari or Uzbek, in 3 years -- meaning,
good enough to translate, with decent accuracy and speed, colloquial,
accented telephone chatter -- regardless of the amount of $ you throw at
the problem.

There are already a series of bonuses for military and civilian
government employees who can pass language proficiency tests, so it
isn't as though we aren't trying.

>So the answer to a fuck-up is to make it a clusterfuck?

I wasn't trying to justify it, just answer some of the questions you
posed.  I warned you they might not make sense when you received them :-)

----- End forwarded message -----
--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE

[demime 1.01d removed an attachment of type application/pgp-signature which had a name of signature.asc]





More information about the cypherpunks-legacy mailing list