first commit

This commit is contained in:
libroot 2026-11-19 04:39:11 +02:00
commit 0d90363116
Signed by: libroot
GPG key ID: 2AC68C029EBE9ACE
6510 changed files with 679146 additions and 0 deletions

View file

@ -0,0 +1,85 @@
DYNAMIC PAGE -- HIGHEST
POSSIBLE CLASSIFICATION IS
TOP SECRET // SI / TK // REL TO USA
AUS CAN GBR NZL
Welcome! Saturday, 10 Nov 2012
* Web search
* Agency-all Emails
* SID-all Emails
* NSA Rolodex
* SCQAWK: The SID Mailbag
* SIDtoday Blog
* SIDtoday Series
* SIGINT Worldwide VTC
* SIDtoday Article
* Letter to the Editor
* SIGINT-y Social Media Page
(U) The SIGINT Philosopher: Descartes Would Have Been a Lousy SIGINT Reporter
FROM: (U//FOUO) [REDACTED]
Run Date: 06/15/2012
(U) I am about to go back on my word, and in only my second entry as the SIGINT Philosopher. I said I wasn't
generally interested in Philosophy as an academic pursuit -- I was a goals-oriented philosopher. But here I'm about to get all abstract and epistemological.*
(U//FOUO) Perhaps this can't be helped. Most NSA employees -- not just in SID are intelligent, thoughtful, somewhat introspective people. After we've been here for a while, we start to ask fundamental questions about the meaning of our work. For those of us in SID who are in the business of interpreting SIGINT, those questions often revolve around how best to share the information that SIGINT gives us with others in the Intelligence Community.
(S//SI//REL) NSA hasn't always reported SIGINT the way it does today. Currently, we task, collect, scan, translate or decrypt as necessary, and then try to interpret the raw
SERIES:
The SIGINT Philosopher
1. The SIGINT Philosopher Is Back -- with a New Face!
2. The SIGINT Philosopher: Descartes Would Have Been a Lousy SIGINT Reporter
3. The SIGINT Philosopher: When Brevity Is Just the Soul of "Huh?"
4. The SIGINT Philosopher: In Praise of Not Knowing
5. The SIGINT Philosopher: Unlike All My Terrible Teammates, I Am a Wonderful Teammate
6. The SIGINT Philosopher: Lessons for Civil Servants from the American Civil War (That Don't Concern Killing Vampires)
decrypt as necessary, and then try to interpret the raw information by writing reports that present the facts as a finished SIGINT report. But in the past -- a past that increasingly few here remember -- we used to simply pass
along raw collections and translations to the customers. When I worked with [REDACTED] version of NSA, I saw
that their reporting was the same thing -- basically just transcriptions of intercepts with short context statements tacked on.
(C//REL) Like many in S2, I both translate SIGINT and report it. I also have a hand in the collections process through tasking selectors. This gives me a reasonably wide vantage point of the intelligence through a number of stages. That vantage point often leaves me feeling that there is a lot lacking in the final product that goes to customers. SIGINT is usually partial information. Two communicants share information on a subject for which they have context that the analyst lacks. These communicants may or may not fully understand what they are talking about. So information is missing from the beginning. Add to this the fact that errors -- or at least difficult renderings -- can occur in translation. The information is already several steps removed from fact (or dare I say from "truth"?) by the time a transcript is finished and I begin to write a report on it.
(U//FOUO) SIGINT analysts who report do an amazing job of puzzling out what their collection means. They also are careful to separate what they know and what they think they know by using words like "probably," "possibly," and "I've only been working here for three weeks." But however brilliantly we may report the information, I can't help thinking that we are essentially another node in the "telephone game":
real information communicants talking about real
information collection translation reporting
(U//FOUO) That's a lot of steps before it ever gets to one of our customers. To make matters epistemologically worse, those customers typically use our reports to then write ANOTHER report for their customers. Usually, when I see an assessment from somewhere in the IC that has something I disagree with in it, the origin can be traced to this long supply chain of information.
(C//REL) Of course, good translation, good reporting, and good collaboration with customers can alleviate these issues.
And as one reporting legend in my office who remembers the old days tells me, almost nobody -- our customers included -- wants to go back to just dumping raw information downstream. As many are pointing out now as we discuss these very issues as part of analytic modernization, the logistical hurdles we would have to endure (USSID 18 training for the entire IC, anyone?) to move our customers more upstream in the information chain are possibly more troublesome than the knowledge gaps that occur through our current system. So this may all be a wistful, empty complaint without a proffered solution. How like a philosopher of me.
(U//FOUO) But the philosopher in my head doesn't really care about practical issues like these. He just knows that this is all unsatisfying somehow. He doesn't like presenting something to others as knowledge when it is really just my interpretation of knowledge. It is my painting of a scene that somebody else has described to me.
(U//FOUO) But enough from that gadfly philosopher. I have reports to write. I know -- I'll tell the philosophy professor in my head that they're cutting his department's funding to fund a new locker room for the football team. He'll be fuming so much, I won't hear from him for months.
Notes:
* Epistemology, "the study of knowledge" is a branch of philosophy that answers annoying questions like "What does it mean to know something?" or "How do you know that you know something?" Its chief purpose, according to the possibly apocryphal memoirs of Leucippus, is to allow philosophy professors to feel superior to us by deconstructing what we all feel to be common-sense answers to these questions.
(U//FOUO) The SIGINT Philosopher is interested to know what you think. Is the idea that customers should be able to hear what I hear and read my best translation word for word doctrinaire mumbo jumbo? Or is it something long overdue in the age of need to share? Is it a good idea that is just too impractical to work? Post your comments on the Tapioca Pebble.
Comments/Suggestions about this article?
"(U//FOUO) SIDtoday articles may not be republished or
reposted outside NSANet without the consent of S0121
(DL sid_comms)."
Information Owner: [REDACTED]
Page Publisher: [REDACTED]
Last Modified: 11/10/2012 / Last Reviewed: 11/10/2012
DYNAMIC PAGE -- HIGHEST POSSIBLE
CLASSIFICATION IS
TOP SECRET // SI / TK // REL TO USA AUS CAN GBR
NZL
DERIVED FROM: NSA/CSSM 1-52, DATED 08 JAN 2007
DECLASSIFY ON: 20320108

View file

@ -0,0 +1,57 @@
{
"title": "The SIGINT Philosopher: Descartes Would Have Been a Lousy SIGINT Reporter",
"type": "SIDTODAY",
"description": "A 2012 SIDtoday SIGINT Philosopher column using Cartesian epistemology to examine the fundamental reliability problem in SIGINT reporting — arguing that by the time raw intercepts pass through collection, translation, interpretation, and reporting, the finished intelligence product is several steps removed from ground truth and risks becoming little more than an analyst's painting of a scene described to them secondhand.",
"long_description": "This SIDtoday column, published June 15, 2012 as the second installment of the new SIGINT Philosopher series, takes an epistemological turn to examine a structural problem in how NSA produces and disseminates finished intelligence. The author — who both translates and reports SIGINT while also participating in the tasking process — uses his unusually broad vantage point across multiple stages of the intelligence cycle to identify a compounding chain of information degradation: intercepts begin as partial communications between parties who share context the analyst lacks, may themselves not fully understand the subject, and are then subject to translation errors before an analyst attempts to interpret and report them. By the time a finished report reaches a customer, the author argues, the product is epistemologically several generations removed from the original facts.\n\nThe column traces a historical contrast between NSA's current practice of producing finished analytical reports and an earlier model of simply passing raw collection and transcriptions to customers — a model the author encountered in a foreign partner version of NSA. While acknowledging that almost no one, customers included, wants to return to raw data dumps, the author questions whether the current system's convenience comes at the cost of systematically obscuring uncertainty. He notes the irony that IC customers then use NSA reports as source material for their own assessments, further extending the telephone-game chain. The piece engages directly with the analytic modernization debate occurring within SID at the time, acknowledging the practical obstacles — including USSID 18 training requirements — that would accompany any move to bring customers further upstream in the information chain, while refusing to treat practicality as a philosophical resolution to the underlying epistemic problem.",
"tags": [
"epistemology",
"SIGINT reporting",
"analytic tradecraft",
"intelligence reporting",
"translation",
"raw SIGINT",
"finished intelligence",
"analytic modernization",
"analyst culture",
"philosophy",
"SIGINT Philosopher"
],
"document": {
"creator": "NSA",
"date_created": "2012-06-15",
"classification": [
"CONFIDENTIAL"
],
"dissemination": [
"REL"
],
"banner": "DYNAMIC PAGE -- HIGHEST POSSIBLE CLASSIFICATION IS TOP SECRET // SI / TK // REL TO USA AUS CAN GBR NZL",
"genre": "newsletter",
"series": {
"name": "The SIGINT Philosopher",
"part": 2
},
"programs": [
"SCQAWK"
],
"countries": [],
"people": []
},
"publication": [
{
"publisher": "The Intercept",
"date_published": "2015-08-11",
"title": "What Happens When a Failed Writer Becomes a Loyal Spy?",
"link": "https://theintercept.com/2015/08/11/surveillance-philosopher-nsa/",
"reporters": [
"Maass, Peter"
],
"contributors": [
"McNeill, Sheelagh"
],
"doc_url": "https://s3.documentcloud.org/documents/2259079/descartes-would-have-been-a-lousy-sigint-reporter.pdf"
}
],
"pdf_sha256": "5110c0a6bbacd73bb44359beb11fe1448feb66a2118215f9bd3909b2859a8596",
"pages": 4
}