Quantcast

Jump to content

» «
Photo

FBI has access to all phone convos, E-Mails etc.

48 replies to this topic
DarrinPA
  • DarrinPA

    Orange Grove Member

  • Members
  • Joined: 31 Mar 2009
  • None

#31

Posted 06 May 2013 - 06:30 PM

QUOTE (sivispacem @ Monday, May 6 2013, 05:49)

QUOTE (DarrinPA @ Monday, May 6 2013, 01:41)
You don't need to. If another terriost attack should happen in America, they'll find out who was involved and then trace back their phone calls, emails, ect, to find who funded the group and to find anymore cells - should there be any.

You do, though. You've got the entire intelligence process backwards.

But sometimes it isn't caught in time, but you can try to stop an ongoing organization. In those cases you work your way backwards, or upwards, until you find the main guy, those who funded it, ect. We already know that American laws allow for surveillance in many different ways. I don't see why the collection of this data would be such a crazy conspiracy.

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#32

Posted 06 May 2013 - 07:12 PM

QUOTE (DarrinPA @ Monday, May 6 2013, 19:30)
QUOTE (sivispacem @ Monday, May 6 2013, 05:49)

QUOTE (DarrinPA @ Monday, May 6 2013, 01:41)
You don't need to. If another terriost attack should happen in America, they'll find out who was involved and then trace back their phone calls, emails, ect, to find who funded the group and to find anymore cells - should there be any.

You do, though. You've got the entire intelligence process backwards.

But sometimes it isn't caught in time, but you can try to stop an ongoing organization. In those cases you work your way backwards, or upwards, until you find the main guy, those who funded it, ect. We already know that American laws allow for surveillance in many different ways. I don't see why the collection of this data would be such a crazy conspiracy.

Well, for one it would be illegal under the current interpretation of US constitutional amendments, but that's rather beside the point. As I feel I explained in great detail, there is absolutely no reason why a competent intelligence agency would engage in data trawling on such a scale because it makes the whole intelligence analysis and dissemination process far harder. The fact is in your example you've already highlighted the surveillance target- friends and associates of the perpetrators. Hence the whole "no point in non-targeted intelligence analysis" comments.

DarrinPA
  • DarrinPA

    Orange Grove Member

  • Members
  • Joined: 31 Mar 2009
  • None

#33

Posted 06 May 2013 - 07:37 PM

QUOTE (sivispacem @ Monday, May 6 2013, 14:12)
QUOTE (DarrinPA @ Monday, May 6 2013, 19:30)
QUOTE (sivispacem @ Monday, May 6 2013, 05:49)

QUOTE (DarrinPA @ Monday, May 6 2013, 01:41)
You don't need to. If another terriost attack should happen in America, they'll find out who was involved and then trace back their phone calls, emails, ect, to find who funded the group and to find anymore cells - should there be any.

You do, though. You've got the entire intelligence process backwards.

But sometimes it isn't caught in time, but you can try to stop an ongoing organization. In those cases you work your way backwards, or upwards, until you find the main guy, those who funded it, ect. We already know that American laws allow for surveillance in many different ways. I don't see why the collection of this data would be such a crazy conspiracy.

The fact is in your example you've already highlighted the surveillance target- friends and associates of the perpetrators. Hence the whole "no point in non-targeted intelligence analysis" comments.

No, actually it would need to collect non targeted communications [ALL communications] to go back and analyse the data to work forwards. Because you wouldn't know which person to collect for if you don't know ahead of time who is guilty. You then use the technology to search for things that have become important to the case, wheither thats the words "flight school "pressure cooker" ect...

Listen, I'm not saying that they do or don't, I'm just saying it would work in their favor and isn't outside the realm of possiblilty. The American government has done many things without permission at the cost of americans privacy and even their health. These type of things are what usually take many years before they are exposed, although there are small leaks that do point to this being real or something along the lines of nationwide surveillance, we won't know for sure for many years. The ongoing lawsuits might bring somethings to light, but those can simply thrown out due to the state secrets privilege.

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#34

Posted 06 May 2013 - 09:15 PM

QUOTE (DarrinPA @ Monday, May 6 2013, 20:37)
No, actually it would need to collect non targeted communications [ALL communications] to go back and analyse the data to work forwards. Because you wouldn't know which person to collect for if you don't know ahead of time who is guilty. You then use the technology to search for things that have become important to the case, wheither thats the words "flight school "pressure cooker" ect...

It would based on what? Care to quantify this statement? The conduct of an attack gives enough knowledge about the perpetrator to effectively focus intelligence analysis efforts there is no such thing as an unknown threat. As I've said multiple times, taking more than what you absolutely need in intelligence gathering is fundamentally harmful to the practice. There is, as I have shown with a variety of external sources, absolutely no cost benefit in collecting unnecessary data. All of the points you are making have been extensively covered in my previous (huge) post. Please refer to that for answers to any of your questions before you post again.

DarrinPA
  • DarrinPA

    Orange Grove Member

  • Members
  • Joined: 31 Mar 2009
  • None

#35

Posted 06 May 2013 - 10:20 PM

QUOTE (sivispacem @ Monday, May 6 2013, 16:15)
QUOTE (DarrinPA @ Monday, May 6 2013, 20:37)
No, actually it would need to collect non targeted communications [ALL communications] to go back and analyse the data to work forwards. Because you wouldn't know which person to collect for if you don't know ahead of time who is guilty. You then use the technology to search for things that have become important to the case, wheither thats the words "flight school "pressure cooker" ect...

It would based on what? Care to quantify this statement?

Sure, it being the data collection system as a whole.

QUOTE
The conduct of an attack gives enough knowledge about the perpetrator to effectively focus intelligence analysis efforts there is no such thing as an unknown threat.

BS, Prove it in every case.

QUOTE
As I've said multiple times, taking more than what you absolutely need in intelligence gathering is fundamentally harmful to the practice. There is, as I have shown with a variety of external sources, absolutely no cost benefit in collecting unnecessary data.

You seem to think that the American Governement is some perfectly run system. "Fundamentally harmful" is not actually harmful. The government is known for wasting resources, why would this situation/techinque be any different? And "cost benefit" is not their concern. We spend money like no other, why stop now?

QUOTE
All of the points you are making have been extensively covered in my previous (huge) post. Please refer to that for answers to any of your questions before you post again.

I forgot that your first post is the end-all of every debate on GTAForums. sarcasm.gif

SagaciousKJB
  • SagaciousKJB

    Captain tl;dr

  • The Connection
  • Joined: 21 Jun 2003

#36

Posted 07 May 2013 - 03:45 AM

QUOTE (sivispacem @ Monday, May 6 2013, 03:49)
QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
The second I saw this thread the first thing I could think of was "inb4 sivis corrects the 'politically incorrect information'"... and what do ya know lol.

And the moment I posted in this thread, I though "I'm sure Grandmaster Smith will be along to blind everyone with his utter ignorance". And what do you know...

QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
They're called Fusion Centers,

...Certainly not what fusion centres actually are. Fusion centres are intelligence handling centres which allow internal discussion between federal organisations and local-level government and law enforcement. They are hardly unique to the US. They are not specifically involved in the creation, collation, analysis and dissemination of SIGINT, which is what this topic is about, but about general intelligence sharing between federal organisations and local-level law enforcement.

QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
“A lot of [the reporting] was predominantly useless information,” one former official told committee investigators after working for the "Reporting Branch" of Intelligence and Analysis unit of the DHS from 2006 to 2010. “You had a lot of data clogging the system with no value."

Welcome to the world of intelligence. There's always a great deal of chaff with the wheat. It's referred to as a wilderness of mirrors for a good reason.

QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
The amount of Fusion Centers in US that collect and go through all digital data-

Oh, now that's a dramatic logical leap. Where in your post has any of the various people you've quoted claimed that fusion centres engage in the collection of intelligence? None of them have. They've mentioned that they process large quantities of raw information. Well that's a given, that's what the basic analysis phase of the intelligence cycle is about. Again, as you so frequently, do, you have extrapolated an argument that you think is implied by the sources you use to support it, but due to your limited knowledge on the subject you've actually ended up telling everyone what they already know about intelligence, bringing nothing new or sensational to the table, and then claiming that normality is part of a wider conspiracy.

QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
now can someone tell me how many democracies have turned into a dictartorship?

Not actually very many, without external influence or the overthrow of legitimate leaderships by violent non-state actors. In fact, I can't think of a single democracy to dictatorship transition after 1945 which came about because of anything other than the involvement of a foreign power in an internal power struggle.

QUOTE (DarrinPA @ Monday, May 6 2013, 01:41)
You don't need to. If another terriost attack should happen in America, they'll find out who was involved and then trace back their phone calls, emails, ect, to find who funded the group and to find anymore cells - should there be any.

You do, though. You've got the entire intelligence process backwards. You don't use intelligence to find out who is involved in a plot after the fact, you use it to break up the plot before it formulates. And if you want to destroy the operating ability of terrorist organisations, militias, criminal cells et cetera, you need to analyse, collate and disseminate the intelligence to law-makers and practitioners rapidly and accurately. Hence why this "scattergun" theory of intelligence is so absurd to anyone trained or experienced in the intelligence world. Why make your already time-dependent job even harder by collecting vast quantities of completely useless information when there are specialist analysts whose sole role it is to tell the people responsible for intelligence collection what to collect?

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
How does that not directly contradict the article? Besides you either forgot what you said earlier or you're trying to distract from it intentionally for whatever reason:

One, it's not an article. It's a comment piece. It's a comment piece that's actually much less clear cut than you imply:

QUOTE (CNN @ 1st May 2013)
CLEMENTE: "No, there is a way. We certainly have ways in national security investigations to find out exactly what was said in that conversation. It's not necessarily something that the FBI is going to want to present in court, but it may help lead the investigation and/or lead to questioning of her. We certainly can find that out.

Which, followed by his second comment, isn't actually tantamount to "we record absolutely everything, ever". That's certainly how our good friend Glenn Greenwald at The Guardian interprets it, but that's not actually what's being said in the interview. There's certainly the capability to recover an awful lot of valuable data from phone calls, but nowhere does anyone with any experiences of SIGINT categorically say that every phone call, electronic communication and web browsing session recorded.

A word about Mr Greenwald, if you will. He's no an intelligence expert. He's not even that versed on foreign policy. He's a constitutional lawyer with a string of rather provocative books to his name, bearing titles like With Liberty and Justice for Some: How the Law Is Used to Destroy Equality and Protect the Powerful. A noted columnist whose works are certainly interesting, yes, but are his opinions- this being present in the "Comment Is Free" section of the website- the gospel truth? Lets look a little further into it in the following post.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
There you claimed that it was not "technically impossible" as you do now but you said it was questionable; you also said that simply because of one reason: you claim nobody could afford that much space for data. I already linked the Utah data center thing and explained that I think this isn't the only facility of its kind so I don't know what more there is to discuss here, you claim one thing and they (NSA, FBI etc.) prove another.

No, I still maintain that it isn't possible. Lets run through how such a system world work in theory to discredit it's existence in reality, shall we, using internet traffic alone as an example?

First of all, you'd need access to every LAN and WAN through which data would travel. This is no mean feat in and of itself. The company I work for does real-time monitoring of network security in large organisations for counter-espionage purposes. We do that by fitting network taps at critical points inside infrastructure, and then drawing down data through our hardware where it is stored for analysis. This is complex and expensive- quite feasibly for a medium-sized company to spend millions of pounds on the hardware alone if they are operating across two or three sites. Therefore you are talking, say, £1.5m in hardware costs alone for ~8-10,000 machines monitored, with a data flow of about 100 million sessions per 24-hours across numerous protocols. Expand this across the millions of small-scale networks that exist in the US and your talking a huge financial cost.

The question comes about, where would you tap into a country wide network? The first port of call would be the ISPs, who already have much of the requisite hardware. But there are a number of serious issues with this. The first of which are transport-layer security protocols, encryption standards and custom protocols. Lets take the standard, Secure Socket Layer/TLS 1.2 protocol as an example. That utilises asymmetrical encrypted handshakes for key generation and symmetrical encryption of transmission data. You use this protocol every time you log into online banking, for instance, or when you make VOIP phone calls. It's protected by the SHA-256 cryptographic hash function, which the brute-forcing of is currently computationally infeasible. Effectively, SHA-256 encrypted communications can only be intercepted by acquiring the cryptographic key through collision analysis (theoretical in this case, having never been definitively proven workable) or by fitting SSL decryptors inside the network, like the boxes Sourcefire provide. These are excruciatingly expensive and very reliant on what is basically a modified man-in-the-middle attack to obtain the cryptographic keys for TLS communications. Which is all well and good, but you need to have one fitted on every network through which SSL traffic could pass internally. Which means every junction box, every independent network, every home broadband setup would require an SSL decryptor in order to capture encrypted data- once you start doing pee-to-peer non-external-facing SSL connections, you are no longer communicating through the defined, clear area of a wide area network and therefore they cryptographic keys can't be acquired. That's why secure peer-to-peer protocols and onion routing are so effective at encrypting data- because you can apply an additional layer of encryption to each transitional stage and therefore you need all the available private keys in order to decrypt the traffic.

To put that into perspective, the NSA couldn't break the OTR-messaging encryption standard used for Apple iChat, let alone the AES encryption of the devices themselves. So, given that most terrorist organisations already use encrypted communication channels, what purpose would there be in hoovering up the 30% or so of all internet traffic which is encrypted seen as it can't be decoded anyway? Wouldn't that just produce a huge amount of unnecessary chaff?

The second is economies of scale. We're at a point of seeing approximately 200 petabytes of conventional internet traffic every day. A large percentage of this goes through US servers and service providers. Now, ISPs distribute their data handling capability- tens of thousands of terrabit-per-second junctions scattered across the country, but in order to hoover up all that data and to draw it off into a single repository you'd need an absurd collection infrastructure. We're talking about potentially hundreds of thousands, or even millions of individual taps. The US digital infrastructure just doesn't have the capability to effectively double already stretched capacity into transmitting captured data of this nature to a central repository.

Then you come onto the issue of identification. Your argument is that all network traffic is being captured; my response is why? You can't decode anything that's encrypted with a reasonable cypher, so why bother capturing it at all? That's 30% of all network traffic rendered effectively pointless. Another 20% is machine-based protocols that don't carry any human information, and can effectively be discarded. Why do you want to listen in on RPC and DHCP activity inside a network; it doesn't tell you much other than about the structure of the network. Why, also, would you want to listen in on the huge quantities of business-related data traffic which would have little to no intelligence purpose? But the question then becomes how do you determine what is what? ISPs provide for both businesses and individuals; MAC and IP addresses are just numerical representations of devices which don't actually possess any real identifying characteristics; the most you can determine from them in isolation would be the device manufacturer. Most externally facing IP addresses are dynamic, so the geography-specific data for them is only applicable for the time at which activity occurs from a single machine with a single dynamic address. ISPs keep a quantity of this data but they also hold vast subscriber records- how would any system of filing such data be capable of discerning what data comes from where and in what form without some kind of metadata tagging system, and the only effective way of providing that metadata that assigns a numerical address to a particular machine, never mind individual, would be by obtaining it from the ISPs. And that's quite aside from the fact that deep packet inspection capability- which the NSA already possesses- renders the requirement to store vast quantities of completely worthless data entirely null and void. Are you starting to see the flaws in this argument?

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Next cheap tactic, claiming I wasn't being accurate to the facts in the source without offering any proof of that whatsoever. Try harder man.

Actually, as I've shown above, you weren't. I've quoted, verbatim, the comments made by the subject matter expert in your initial post, and they don't actually say what you claim they do. I can only presume that you've handily chosen to ignore this and instead accuse me of engaging in "cheap tactics" because on second reading you know that your initial hypothesis doesn't actually hold water and therefore you want to continue the debate whilst not actually referring back to the sources. Cast your eyes back to your comments on the Utah Data Centre, and then consider two issues. One, that the data centre's primary purpose isn't to be a repository for all electronic communications in the US, but merely all electronic communications with an intelligence value. The NSA does much of their SIGINT work abroad, and the biggest target for data collection in the current economic and strategic climate is China. Also, whilst the trillion-terrabyte-storage capability sounds terrifying, it handily ignores the primary purpose behind having such computing power and storage capability- the breaking of complex cryptographic algorithms and cyphers. The Wired Threat Levels article discussing the data centre goes into the value of this for intelligence gathering to a great degree.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
More of the same, trying to ridicule me without any content or basis whatsoever, all you said is "you're stupid", just in more words. I'm used to reading through lots of bullsh*t so this won't work either, sorry Joe.

Not really. Modern intelligence analysis is based on a target-centric approach. Target-centric intelligence analysis, even when utilising a great wealth of open-source intelligence data as outlined in the June 21st, 2005 congressional hearing on the subject requires target acquisition capabilities that pre-exist the intelligence collection capability. Part of the fundamental failures in the US intelligence community in the lead-up to the September 11th, 2001 attacks was information saturation- too much unspecific, uncultured and largely ineffectual data. The idea of "grab everything and then sort it" is a very Cold-war mentality which works pretty well with electronic intelligence gathered specifically from targeted individuals and nation states but is fundamentally flawed when it comes to accurate and timely assessment and dissemination of intelligence collected from the open-source macrocosm that is wider society. My issue with your comments is that you appear to have formulated an argument which doesn't correlate logically with taught or operational intelligence practice. I questioned your understanding of the intelligence cycle and modern intelligence analysis techniques because your argument doesn't appear to acknowledge how the intelligence cycle works, nor does it accurately portray the most effective ways of gathering targeted, specific intelligence on subnational violent actors- who are the primary security threat to the US mainland. One of the strongest arguments against your thesis is that it doesn't actually make sense unless you suspend the accepted knowledge of how intelligence analysis works- instead it required a Hollywood cultured understanding of intelligence which doesn't actually reflect reality.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Surprise surprise, suddenly when confronted with direct proof for the points made by those retired agents, ex AT&T employees and journalists Mr. self proclaimed intelligence analyst doesn't know what I'm talking about and tries to redirect the conversation.

Direct proof of what? I only said I had no familiarity with the specifics of the Boston bombing intelligence response, or to the specific allegation that recorded conversations were used as evidence and that this is somehow proof of an existing capability to store all communications data of all kinds. You appear to be seriously misunderstanding my responses; whether intentionally or otherwise. You also seem to seriously over-estimate the quality of your own sources and the extent to which they actually support your thesis. You have an ex-federal agent who says that the US has a great deal of capability in determining content and data from phone communications- big surprise. An AT&T operator saying that intelligence agencies are drawing down huge quantities of data to support investigative capabilities- hardly earth-shattering news. And a noted constitutional lawyer with a history of speaking out against all kinds of perceived violation of civil liberties speculating about the capabilities of the US intelligence apparatus- how astonishing. None of these people have actually produced the same argument you have with the possible exception of the latter- but he talks more in terms of capability than actual active operational usage. You've made the fundamental mistake of extrapolating comments and extending them to what you see as a logical conclusion. As evidenced here, handily:

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
QUOTE
Meanwhile, investigators in the Boston bombing case want to find out what Tamerlan Tsarnaev and his wife discussed when he phoned her a few hours after the FBI released photos of him and his brother as suspects in the deadly attack, a separate law enforcement official said Thursday.

From http://www.usatoday....ombing/2129289/
And that's precisely the point, how can they do it if they hadn't saved it? They said themselves, I have to repeat this because you ignored it so conveniently, that they had NOT been looking into the Tsarnaevs despite the repeated warnings they had gotten from foreign agencies.

So then enlighten me now, how the f*ck can they access the content of the phone conversations if they haven't saved them specifically? The only way is to save them all indiscriminately. No idea for how long, I assume there must be a point at which data that is deemed useless gets deleted. Still...

There is nothing in this comment which says that they are trawling through electronic records of the call in order to find this data. That is merely your interpretation. I provided you with a list of other ways in which it would be perfectly possible to obtain the same data content without recording the entire conversation. How can they access this information? Well, the first port of call would probably be interrogating Tsarnaev's wife. There are clearly numerous other ways in which this information could be obtained- external third parties, for instance. The actual phrasing of the comment itself seems to indicate investigators calling on members of the public and existing individuals with known relationships to Tsarnaev in order to obtain this information. If they were interrogating a huge federal database of every conversation ever, why would they "want to find out" about it? Surely they'd already know?

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
No I did not. Echelon was not the first time anyone thought of "keywords" or filtering data using them, even though you appear to think so (or actually I doubt you do, you'd just like to get others to believe it so it seems like you were pointing out actual errors in what I wrote to try to weaken my arguments). But I'm used to you claiming I said things I never did by now, at least you manage to keep a straight face doing it, respect where due.

I never said it was- that was your inference from my comments. I said it was the first time that it became part of a wide-ranging conspiracy theory that entailed the drawing down of all communications data. But nice job of not actually addressing the point I've made.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Well, they're not conspiracy theories anymore once they're proven no?

Are they? I'm pretty sure it hasn't been proven that the government are slurping up all communications data inside the US borders.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Like Echelon... as for "keyword theory", for some reason I can't find anything on that via Google, only related to marketing and such. I seriously doubt Echelon was really the first time in human history anyone came up with the idea to look for some specific keywords in a text. Actually every child does that when "reading" a book for school, they just look at what seems interesting and make up the rest as they go along to save time. The motives are different but the idea is the same.

Keyword analysis theory is pretty well known. Advertising is based almost solely on it. And yes, it is used in intelligence analysis. But not how you seem to think it is. Remember, conventional SIGINT (that is, communication data as opposed to FISINT and subordinates like TELINT and MASINT) is intrinsically tied with cryptography and cryptanalysis. The people for whom interception may be worthwhile tend to use code and go to great lengths to obfuscate their activities, because they tend to have a relatively strong awareness of intelligence gathering capability and activity combined with the desire to not get caught. Keyword analysis has been utilised when codes used by organisations are known- one of the most famous ones in relation to violent Islamism is the discussion of weddings as a coded precursor to planning and perpetrating suicide bombings in particular- but this requires a pre-existing knowledge of the organisation which you don't possess by vacuuming up all the data available and running it through a slightly more intelligent version of Wolfram Alpha.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Blah blah blah dude,
More blah.

How riveting and engrossing. One would almost think you'd run out of argument to make?

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
If you have the space just gather everything, that makes no extra effort once it's in place.

See, this is fundamentally incorrect. Let me explain:

Intelligence analysis is an art. It is scientific, but is not purely empirical. It is as much about cultural and human terrain analysis as it is anything else. Analysts work with an operational and strategic goal in mind- the strategic goal usually relating to the target on which intelligence is gathered, and the operational goal being the specifics of what ever macro-level analysis they are engaged in. You cannot just simply give a team of analysts a huge quantity of data and say "find bad stuff". That's not how it works. Hence my repeated reference to the requirements of pre-existing knowledge. The "find bad stuff" fallacy is kind of what happened to the US intelligence community at the end of the Cold War, once their primary strategic adversary ceased to exist. Analysts have different areas of expertise. There is no real "catch-all" analyst. As we've already established, you can't really electronically sift communications information stored in bulk without human analysis. In order for that human analysis to take place, you need a strategic goal, and an operational goal for the analysts working on the project to aspire to- that's how you measure success. It is fundamentally flawed to think that intelligence analysis is the art of finding a needle in a haystack- which is what you seem to be implying it consist of. It's actually the art of finding a particular needle in a shoebox full of needles, and then explaining why that needle is important, and making predictions and assessments about what it has stitched in the past and may sew in the future. Analysis operates inside defined borders and boundaries- it's when you start crossing borders and spreading too much data too thinly amongst an analytical team that you wind up having problems. As I've said numerous times, large quantities of data actually inhibit the ability to analyse them.

An example, if you will. Say you are a theoretical physicist who wishes to do an assessment on the validity of an experiment, it's potential implications and what findings other related experiments might have to it. You've got your data set- the experiment itself; your operational goal- to assess validity; and your additional corroborative sources- the other related papers and journals. You might have to hunt around through numerous journals in order to find valid experiments and you might have to analyse and assess their worth in the context of your strategic goal, but you have a point of focus. This seems totally logical, does it not, and that's largely how intelligence analysis works. Now, try doing the same if your data set and related experiments are still present, but are stored on something like JSTOR for which you have full access but no search function. You've got to manually sift through all the articles on anthropology, Greek history and media management theory in order to find your data set and sources. Seems a bit silly, doesn't it? That's how you seem to imagine intelligence analysis works.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
I don't know what you're referring to by "shotguns", I assume "cases" or something?!

My mistake, typing on a phone without proof-reading.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
But this I'd like you to explain. How exactly can you collect intelligence by using keywords? That just makes no sense, unless you mean "there's a folder on the table, it says BOMB on top, maybe I should read that". But I doubt that happens very often.

Simple- by already possessing a decent understanding of the organisation which you are targeting, and knowing what keywords they are most likely to use. Intelligence analysis feeds strategic understanding, and strategic understanding focuses intelligence analysis. Hence the importance of things like theology, politics, human terrain analysis, sociology and anthropology in the world of intelligence.

QUOTE (baguvix_wanrltw @ Monday, May 6 2013, 02:04)
Again, ask Schneier for example.

Schneier, as in the cryptographer and security researcher Bruce Schneier? He's been pretty vocal in his statements that human-based macro-sphere applied intelligence analysis is vastly superior in all ways to automated intelligence trawling. In fact, in his essay Homeland Insecurity, he said categorically that increasing the amount of intelligence gathered inhibits the ability to accurately and effectively analyse it. Which is what I've been saying all along. And what accepted intelligence analysis theory says. And what all intelligence agencies that are vaguely competent in their role do.

This post was fun to read as a cyrptogeek. I need some popcorn or something.

Hey sivi, you ever read "Cryptonomicon" by Neal Stephenson? You might find it entertaining, at least for a good laugh. Pretty much all about the history of cryptology and the intelligence sector of WW2, tied up in a nice fictitious treasure-hunters tail. I think it kind of echoes a lot of what you've been saying, pretty interesting book.

Also...

"...SIGINT....SIGINT...SIGINT"

Just SIGKILL already

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#37

Posted 07 May 2013 - 08:48 AM

QUOTE (DarrinPA @ Monday, May 6 2013, 23:20)
Sure, it being the data collection system as a whole.

But as I've explained several times, it's totally superfluous. Go have a read of the Bruce Schneier essay link I posted in my huge response, seen as you really seem to be struggling with the concept that more raw data makes intelligence analysis harder. He explains it extremely well. I'll explain it again, quite simply. There is absolutely no purpose to drawing down all communications data, quite aside from the fact that it's physically impossible given the state of the US' current digital infrastructure. Intelligence analysis has, and probably always will be, a targeted art and isn't frequently used for forensic purposes, which is what you seem to be implying. That's primarily because there's no such thing as an unknown threat, just one that the probability of hasn't been accurately assessed. Do you want to see if you can find me a single strategic or operational-level security threat throughout history which has been entirely unknown to the US intelligence apparatus?

QUOTE (DarrinPA @ Monday, May 6 2013, 23:20)
BS, Prove it in every case.

What, prove that there isn't an unknown threat? Or prove that the requisite intelligence collected after an attack usually allows an intelligence agency to rapidly and accurately form a picture of the perpetrator organisation? They aren't really quantifiable things, given that they mostly happen behind closed doors, out of the public eye, and not being a US citizen my level access to protectively marked US intelligence is pretty low, don't you think? Tell you what, you seem to be the one who is arguing against accepted, legally enshrined intelligence practices. Why don't you bring something to the table that supports your hypothesis that everything that's known and studied about the art and science of intelligence analysis is wrong in practice? Also, I had to laugh at your "BS" claim. On what prior experience are you issuing that claim? What grounding in the subject do you have to assess the worth of other people's statements?

QUOTE (DarrinPA @ Monday, May 6 2013, 23:20)
You seem to think that the American Governement is some perfectly run system. "Fundamentally harmful" is not actually harmful. The government is known for wasting resources, why would this situation/techinque be any different? And "cost benefit" is not their concern. We spend money like no other, why stop now?

Not at all. But you really don't seem to understand the point I'm making. The cost implications are pretty unimportant, but the benefit implications are not. If there is no benefit to hoovering up vast quantities of completely worthless data- which according to every source I've provided on the issue, and according to accepted intelligence doctrine, there isn't- then why do it? The justifications that it's just an exercise of "I want" and that the US government is an imperfect entity don't really hold water if you understand that by doing so they actually make their own lives more difficult, more complex, and actually make their nation less secure by doing it. States are rational actors- why would the US government act so irrationally around the issue?

QUOTE (DarrinPA @ Monday, May 6 2013, 23:20)
I forgot that your first post is the end-all of every debate on GTAForums.  sarcasm.gif

It's very poor form to wade into a debate without reading all the supporting material and the arguments made by both sides..

DarrinPA
  • DarrinPA

    Orange Grove Member

  • Members
  • Joined: 31 Mar 2009
  • None

#38

Posted 07 May 2013 - 08:07 PM

QUOTE (sivispacem @ Tuesday, May 7 2013, 03:48)
QUOTE (DarrinPA @ Monday, May 6 2013, 23:20)
BS, Prove it in every case.

What, prove that there isn't an unknown threat?

Yes. You said "The conduct of an attack gives enough knowledge about the perpetrator to effectively focus intelligence analysis efforts there is no such thing as an unknown threat."

baguvix_wanrltw
  • baguvix_wanrltw

    Cynical, yeah. Bitter, probably.

  • Members
  • Joined: 05 Apr 2012

#39

Posted 07 May 2013 - 08:43 PM

We have to continue this sometime soon but first I'll need to clear some time, reading and replying to all of that... man. I think we might have overdone it on the walls of text lol.gif

Just 1 quick remark because that stuck out to me, I think you (sivi) mentioned on more than 1 occasion that not all data is accessible because some is encrypted or otherwise inaccessible, I saw you mention SSL, SHA, OTR etc. so I'm sure I'll have fun reading all that. But generally, without having read the details of what you wrote, I think we must admit that for example the CA concept hasn't really worked out well which means, as we've often seen, that governments (or black hats for that matter) with far fewer resources than the US can easily eavesdrop on "encrypted" comms.

Even if they can't only a small portion of traffic/data is encrypted, at least in a fashion that is truly hard or "impossible" to crack so this doesn't apply to most of the people (or data) in the real world that *might* be stored wherever.

And all of those assumptions regarding cryptography are based on the belief that the NSA can not crack more than we think. Some believe they can crack AES, others say that's a crackpot theory, I personally don't work there so I don't know.

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#40

Posted 07 May 2013 - 10:09 PM

QUOTE (baguvix_wanrltw @ Tuesday, May 7 2013, 21:43)
Just 1 quick remark because that stuck out to me, I think you (sivi) mentioned on more than 1 occasion that not all data is accessible because some is encrypted or otherwise inaccessible, I saw you mention SSL, SHA, OTR etc. so I'm sure I'll have fun reading all that. But generally, without having read the details of what you wrote, I think we must admit that for example the CA concept hasn't really worked out well which means, as we've often seen, that governments (or black hats for that matter) with far fewer resources than the US can easily eavesdrop on "encrypted" comms.

In these cases, it is seldom people actually breaking the encryption standard. Aside from SHA-1, MD5 and the Blowfish precursor I can't remember the name of, I don't think a single widely implemented encryption algorithm has ever been broken. Plaintext capture of encrypted communications is almost solely done using man-in-the-middle attacks where the adversary obtains both keys in the handshake by effectively tricking both respective parties into thinking that they are the other. This is how electronic surveillance of terror suspects and the like is possible- you basically conduct a man-in-the-middle attack against them (or failing that, you lure them into a honeypot as so many Jihadist and neo-Nazi online forums have become these days). The issue again boils down to targeted versus non-targeted intelligence gathering. If you are engaging in specific targeting, you can direct resources like SSL decryptors fitted to external infrastructure, or by recording data at a pre-encrypted stage, from the source (that's how most banking trojans work, for instance- they don't need to break the encryption, just log the plain-text keystrokes). It's very hard to apply on a wide scale because of the cost and throughput penalties- even the most powerful SSL decryptors can only handle about 4GBit/s traffic, so multiple ones would be needed to decrypt traffic at a single one of the tens of thousands of TBit/s junction boxes around the US.

QUOTE (baguvix_wanrltw @ Tuesday, May 7 2013, 21:43)
Even if they can't only a small portion of traffic/data is encrypted

Around 30% according to most estimates. Also, the traffic they actually want to listen to is statistically far more likely to be encrypted.

QUOTE (baguvix_wanrltw @ Tuesday, May 7 2013, 21:43)
And all of those assumptions regarding cryptography are based on the belief that the NSA can not crack more than we think. Some believe they can crack AES, others say that's a crackpot theory, I personally don't work there so I don't know.

Indeed, but given that the NSA use a variety of these encryption standards and cyphers (SHA-256 and 512, AES and to a lesser extent Threefish) in their own encryption of data, and given that the cracking of these algorithms is deemed computationally infeasible based on the known computing capability of the entire world (I think AES would take something like 56 million years to crack if every known desktop, workstation, server and supercomputer was used solely for that purpose), it's a fair assumption. For instance, the computational complexity of an attack on AES-256 is 2^254. Which is a number so large that it makes the number of particles in the universe look like a rounding error.

QUOTE (DarrinPA @ Tuesday, May 7 2013, 21:07)
Yes. You said "The conduct of an attack gives enough knowledge about the perpetrator to effectively focus intelligence analysis efforts there is no such thing as an unknown threat."

I can't think of a single modern strategic security threat which wasn't foreseen to a greater or lesser extent by intelligence analysts. The rise of the Soviet Union? Yes. Communism in Central and Southern America? Yes. Proxy conflict in sub-Saharan Africa? Yes. The rise of militant Islam? yes. Conflict in the fifth dimension? Yes.

Tell you what, seen as you appear to be so confident that I'm wrong in saying that there is effectively no such thing as an unknown threat, how about you furnish me with an example of what you consider to be an "unknown" strategic threat to the United States? You obviously have one in mind, as you can't honestly want me to list every conflict that the US has been directly and indirectly involved in and provide you a rough date at which they were estimated to be potential future security threats.

Jake
  • Jake

    Vagina Mine

  • $outh $ide Hoodz
  • Joined: 02 Dec 2003

#41

Posted 08 May 2013 - 02:46 AM

As far as I read, they said they simply store it all for later use if need be. Back tracking type deal. And being able to intercept phone calls. I doubt they would anaylyze on the fly. Way too much sifting.

Doesn't really bother me if true.

beavis
  • beavis

    Lone Wolf

  • $outh $ide Hoodz
  • Joined: 28 Apr 2002

#42

Posted 08 May 2013 - 02:56 AM Edited by beavis, 08 May 2013 - 02:59 AM.

Not news to me.

http://en.wikipedia.org/wiki/Room_641A

QUOTE
Room 641A is located in the SBC Communications building at 611 Folsom Street, San Francisco, three floors of which were occupied by AT&T before SBC purchased AT&T.[1] The room was referred to in internal AT&T documents as the SG3 [Study Group 3] Secure Room. It is fed by fiber optic lines from beam splitters installed in fiber optic trunks carrying Internet backbone traffic[3] and, as analyzed by J. Scott Marcus, a former CTO for GTE and a former adviser to the FCC, has access to all Internet traffic that passes through the building, and therefore "the capability to enable surveillance and analysis of internet content on a massive scale, including both overseas and purely domestic traffic."[4] Former director of the NSA’s World Geopolitical and Military Analysis Reporting Group, William Binney, has estimated that 10 to 20 such facilities have been installed throughout the nation.[2]
The room measures about 24 by 48 feet (7.3 by 15 m) and contains several racks of equipment, including a Narus STA 6400, a device designed to intercept and analyze Internet communications at very high speeds.[1]
The very existence of the room was revealed by a former AT&T technician, Mark Klein, and was the subject of a 2006 class action lawsuit by the Electronic Frontier Foundation against AT&T.[5] Klein claims he was told that similar black rooms are operated at other facilities around the country.


If they can do this with internet traffic it must be easy dealing with phone conversations as well.

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#43

Posted 08 May 2013 - 07:04 AM Edited by sivispacem, 08 May 2013 - 08:50 AM.

QUOTE (beavis @ Wednesday, May 8 2013, 03:56)
Not news to me.

http://en.wikipedia.org/wiki/Room_641A

QUOTE
Room 641A is located in the SBC Communications building at 611 Folsom Street, San Francisco, three floors of which were occupied by AT&T before SBC purchased AT&T.[1] The room was referred to in internal AT&T documents as the SG3 [Study Group 3] Secure Room. It is fed by fiber optic lines from beam splitters installed in fiber optic trunks carrying Internet backbone traffic[3] and, as analyzed by J. Scott Marcus, a former CTO for GTE and a former adviser to the FCC, has access to all Internet traffic that passes through the building, and therefore "the capability to enable surveillance and analysis of internet content on a massive scale, including both overseas and purely domestic traffic."[4] Former director of the NSA’s World Geopolitical and Military Analysis Reporting Group, William Binney, has estimated that 10 to 20 such facilities have been installed throughout the nation.[2]
The room measures about 24 by 48 feet (7.3 by 15 m) and contains several racks of equipment, including a Narus STA 6400, a device designed to intercept and analyze Internet communications at very high speeds.[1]
The very existence of the room was revealed by a former AT&T technician, Mark Klein, and was the subject of a 2006 class action lawsuit by the Electronic Frontier Foundation against AT&T.[5] Klein claims he was told that similar black rooms are operated at other facilities around the country.


If they can do this with internet traffic it must be easy dealing with phone conversations as well.

Interesting, but the mere presence of hardware which allows the analysis of communication data is hardly evidence of a plot to slurp up all of it, which is what the initial post regards. There are a whole raft of perfectly legitimate uses for such hardware. The mention of the Narus STA 6400 is particularly interesting in this context- these devices are designed for semantic, network and user relationship analysis. I can't find much information on the STA 6400, but I've found plenty of info on their latest piece of analytical hardware- the nSystem. They have the capability to categorise a large throughput of data, analyse machine-level protocols and produce semantic information like relationships, with a functionality not dissimilar to Maltego or IBM Analyst's Notebook. But that if anything acts as evidence against the idea that this data is recorded en masse- because there would be no purpose in doing so if you could already extract communication metadata from it and develop relational intelligence and semantic information without the need for slower, more intricate full deep packet inspection.

EDIT

In my large post I mentioned Bruce Schneier. He's actually recently made a statement pertaining to the original subject of this topic-

QUOTE (Bruce Schneier)
I'm very sceptical about Clemente's comments. He left the FBI shortly after 9/11, and he didn't have any special security clearances. My guess is that he is speaking more about what the NSA and FBI could potentially do, and not about what they are doing right now. And I don't believe that the NSA could save every domestic phone call, not at this time. Possibly after the Utah data center is finished, but not now. They could be saving the all the metadata now, but I'm skeptical about that too.


More info

baguvix_wanrltw
  • baguvix_wanrltw

    Cynical, yeah. Bitter, probably.

  • Members
  • Joined: 05 Apr 2012

#44

Posted 09 May 2013 - 09:09 PM

Ok so here we go smile.gif

First let me extend the peace pipe again - I'm sorry for having been agitated. While on that topic, I'm also sorry it took so long until I had a chance to reply - trust me, I wanted to. I just couldn't get to it, lots of other stuff that unfortunately had higher priorities.

I see you (sivi) linked Schneier's post already, yeah, that was indeed the Schneier I was talking about. But if I might add to your quote, he later added this to his post:

QUOTE
EDITED TO ADD (5/7): Interesting comments. I think it's worth going through the math. There are two possible ways to do this. The first is to collect, compress, transport, and store. The second is to collect, convert to text, transport, and store. So, what data rates, processing requirements, and storage sizes are we talking about?

So even Schneier isn't denying that it might be possible, unlike what one might think without reading this update smile.gif I'm sure it just wasn't there yet when you posted but still I think it's worth mentioning.

Also Schneier failed to acknowledge the various other sources pointing to such activities, many of which are linked in the comments, among others the AT&T and NSA guys I mentioned before plus a couple of others. Jacob Appelbaum is linked there too, btw, on that topic might I recommend http://assange.rt.com/, especially the two full length Cypherpunks episodes with Mrs. Appelbaum, Assange, Müller-Maguhn and Zimmermann.

Oh and not to be boastful or anything but I'd like to point out that Schneier apparently failed to take into consideration the voice-to-text aspect before as well but has now and sees the whole thing as less of an impossibility... which btw is surprising, no idea how he could have missed that ("can't store all data...only metadata... oh well, it's worth doing the math").

On that subject IIRC people mentioned different languages, accents etc; that was addressed in the Schneier blog post comments as well, you can easily determine a language and accent first and filter by content later, discard of the most obvious trash etc., then there's data compression that can be done for archival which works beautifully on written (in files) text, etc etc.

I'd really recommend people read the whole blog post, including the comments; some people tried to get the math straight with more or less (seemingly) accurate results; but what it *appears* to come down to is: it is possible, very possible indeed; the cost for storage of all comms are surprisingly low if you take into consideration cloud services and the like, which also makes it obvious that the NSA and similar agencies with similar resources might not really have a hard time financing something like this.

sivi would probably like to stop me right there and say that's not the problem, the aqusition or analysis is; but first off scratch the analysis because that comes later, we're not even that far yet. The whole thread is about the steps before the analysis.
And as for the costs of the aquistion, some interesting devices were already mentioned that are capable of acquiring the required data for surprisingly low prices in relation to the massive surveillance they enable. Again read the Schneier post and especially the comments and the sites linked there, I might pick a few links to repost here later but still, really interesting.
Damn I see "Clive Robinson" (who is a regular poster but I've never felt a need to try and check out who he is) also just mentioned CarrierIQ and similar doing the NSA's job for them there. Oh snap lol.gif

Lucky for us euroc*nts we seem to not be quite where the US of A are in this respect yet biggrin.gif

Oh and btw here's (apparently) the full transcript of the interview with Clemente on CNN.

Now to the "big post", god help me...


QUOTE (sivi)
QUOTE (GrandMaster Smith @ Sunday, May 5 2013, 23:51)
now can someone tell me how many democracies have turned into a dictartorship?

Not actually very many, without external influence or the overthrow of legitimate leaderships by violent non-state actors. In fact, I can't think of a single democracy to dictatorship transition after 1945 which came about because of anything other than the involvement of a foreign power in an internal power struggle.

Just one comment on this, come on, you know precisely what he meant by "turning into a dictatorship". I'm not defending what he said otherwise but getting bogged down in semantics won't help.

QUOTE (sivi)
QUOTE (DarrinPA @ Monday, May 6 2013, 01:41)
You don't need to. If another terriost attack should happen in America, they'll find out who was involved and then trace back their phone calls, emails, ect, to find who funded the group and to find anymore cells - should there be any.

You do, though. You've got the entire intelligence process backwards. You don't use intelligence to find out who is involved in a plot after the fact, you use it to break up the plot before it formulates. And if you want to destroy the operating ability of terrorist organisations, militias, criminal cells et cetera, you need to analyse, collate and disseminate the intelligence to law-makers and practitioners rapidly and accurately. Hence why this "scattergun" theory of intelligence is so absurd to anyone trained or experienced in the intelligence world. Why make your already time-dependent job even harder by collecting vast quantities of completely useless information when there are specialist analysts whose sole role it is to tell the people responsible for intelligence collection what to collect?

Well, I'm not sure about the example Darrin gave but there certainly are cases in intelligence where the process works backwards as you refer to it. Like Boston appears to become now. Or the case where the sheik was killed in his hotel by what, 15 or so Mossad agents dressed as tourists?

Of course maybe the definition of "intelligence" in this context just escapes me as a non-native speaker but from what I gather it's nothing but another word for gathering (and processing, analyzing etc.) information in an effort to find out something that can help against an adversary of some sort. It doesn't necessarily mean that the adversary can't already have acted, right? If it does my apologies.

But really I think me and Darrin weren't debating the precise order of events in usual intelligence cases but more that the capabilities for such new kinds of massive surveillance might exist, or actually that we find it probable that they do.

QUOTE
One, it's not an article. It's a comment piece.

My bad.

QUOTE
It's a comment piece that's actually much less clear cut than you imply:

QUOTE (CNN @  1st May 2013)
CLEMENTE: "No, there is a way. We certainly have ways in national security investigations to find out exactly what was said in that conversation. It's not necessarily something that the FBI is going to want to present in court, but it may help lead the investigation and/or lead to questioning of her. We certainly can find that out.

Which, followed by his second comment, isn't actually tantamount to "we record absolutely everything, ever". That's certainly how our good friend Glenn Greenwald at The Guardian interprets it, but that's not actually what's being said in the interview. There's certainly the capability to recover an awful lot of valuable data from phone calls, but nowhere does anyone with any experiences of SIGINT categorically say that every phone call, electronic communication and web browsing session recorded.

To be entirely fair, let's get the entire quote from the guardian in here, important parts bolded to ruthlessly further my arguments:
QUOTE
BURNETT: Tim, is there any way, obviously, there is a voice mail they can try to get the phone companies to give that up at this point. It's not a voice mail. It's just a conversation. There's no way they actually can find out what happened, right, unless she tells them?

CLEMENTE: "No, there is a way. We certainly have ways in national security investigations to find out exactly what was said in that conversation. It's not necessarily something that the FBI is going to want to present in court, but it may help lead the investigation and/or lead to questioning of her. We certainly can find that out.

BURNETT: "So they can actually get that? People are saying, look, that is incredible.

CLEMENTE: "No, welcome to America. All of that stuff is being captured as we speak whether we know it or like it or not."

So to pick out the most important points:
- "a conversation", aka voice;
- "find out what happened", which obviously means the content of said conversation
- "have ways to find out exactly what was said in that conversation" - any more questions? He is NOT talking about metadata. He is not talking about trying to reconstruct something to get a rudimentary understanding of the content. He is talking about finding out exactly what was said, that means word for word, that means it has to be saved somewhere, whether temporarily or forever is anyone's guess. The same thing goes for the format, OPUS, ASCII or whatever.
- "No, welcome to America" clearly implies that this is standard procedure, not something that is only targeted at very specific people
- "All of that stuff is being captured as we speak whether we know it or like it or not" is also not really open for interpretation now is it?!

So tell me then how you would reasonably interpret what he said? I mean if anything I'd have expected you to tell me the guy wasn't credible for whatever reasons, despite even the US govt not having said anything about the matter as far as I'm aware, again despite the waves the topic seems to start creating around certain places.

But attacking Greenwald and saying his interpretation was somehow weird... I just don't see where. I'm guilty of attacking you as well so I can hardly blame you but I think generally we can now agree to keep it more civil, the same IMHO goes for the people we're talking about.

QUOTE
A word about Mr Greenwald, if you will. He's no an intelligence expert. He's not even that versed on foreign policy. He's a constitutional lawyer with a string of rather provocative books to his name, bearing titles like With Liberty and Justice for Some: How the Law Is Used to Destroy Equality and Protect the Powerful. A noted columnist whose works are certainly interesting, yes, but are his opinions- this being present in the "Comment Is Free" section of the website- the gospel truth? Lets look a little further into it in the following post.

Hm, afaik you didn't get back to him yet, not that I need or want you to. Tbh I'm not interested in whatever your opinion is on Greenwald because his opinion did not influence me when I started making up my mind about this (and I'm not finished obviously because most of the important data just isn't there). So let's skip the "don't believe this or that guy" spiel about opinions, I try to form my own anyway. The facts and statements in that piece were interesting, that's why I linked it, not because of the author.

QUOTE
First of all, you'd need access to every LAN and WAN through which data would travel. This is no mean feat in and of itself. The company I work for does real-time monitoring of network security in large organisations for counter-espionage purposes. We do that by fitting network taps at critical points inside infrastructure, and then drawing down data through our hardware where it is stored for analysis. This is complex and expensive- quite feasibly for a medium-sized company to spend millions of pounds on the hardware alone if they are operating across two or three sites. Therefore you are talking, say, £1.5m in hardware costs alone for ~8-10,000 machines monitored, with a data flow of about 100 million sessions per 24-hours across numerous protocols. Expand this across the millions of small-scale networks that exist in the US and your talking a huge financial cost.

Well but you're now comparing networks or large organizations (corporations?) to the entire internet, or the "entire US internet". (Which is a problem like all "local" internets [=ones that can basically be switched off or disconnected from the oustide "internets" remotely; UPDATE:just like Syria showed once again today] btw, not to mention root DNS servers being pretty centralized blah... a little off topic at this moment)

I can't say much about the numbers you offered here, I don't know the capacities involved etc. so I can't say much else right here except remind of Schneier's post and the comments where they did the math as well, and linked a popular post that wants everyone to make sure to not refer to it as "news".

Generally you're right of course, it's not easy to tap into everything; but if you have the cooperation of the Telcos, like with Room 641A as mentioned later, it's very very possible and has actually been done before. And by whom? AT&T & the NSA of course, the usual suspects. Those guys that suddenly had ex-employees going whistleblower on this sh*t after they had realized it was against the best interest of the American people and againts the constitution, the same people I mentioned in my earlier posts multiple times.

QUOTE
Lets take the standard, Secure Socket Layer/TLS 1.2 protocol as an example...

I've addressed this a little already; people like to act like SSL was secure but Türktrust, the case of the forged MS certificates, Abobe Certificates, Realtek certificates, etc. etc. etc. have again and again shown one thing: the CA system is fundamentally broken and is open to any adversary who truly wants "in". Oh another funny thing in that case: HTP took over the control over the root DNS for .edu domains, I can only recommend reading their zine on their exploits (the latter pretty much literally; sh*t site is down atm, try later). They released the logins for most. If you get that level of control there isn't much you can't do, especially if you have some sort of country specific DNS like many of the arab countries did that were later critizied for shutting them off, blocking certain sites or redirecting them; oh and I remember one country actually even blocked all SSL traffic at one point.
The HTP case is also a fine example against something else you said: in some cases it's just beautiful to have tons of info to look through tounge.gif In the context of our topic however it's true, you don't want all of the garbage but I'm sure the NSA has pretty good filtering techniques; actually another much older Schneier post on that topic- from 1999 (!), take note of what he mentions as possible for such agencies almost one and a half decades ago - gets into that more deeply; talking about the "ECHELON Technology" part: https://www.schneier...-gram-9912.html

A little quote:
QUOTE (Schneier)
One of the major technological barriers to implementing ECHELON is automatic searching tools for voice communications. Computers need to "think" like humans when analyzing the often imperfect computer transcriptions of voice conversations.
The patent claims that the NSA has solved this problem. First, a computer automatically assigns a label, or topic description, to raw data. This system is far more sophisticated than previous systems because it labels data based on meaning not on keywords.
Second, the patent includes an optional pre-processing step which cleans up text, much of which the agency appears to expect will come from human conversations. This pre-processing will remove what the patent calls "stutter phrases." These phrases "frequently occurs [sic] in text based on speech." The pre-processing step will also remove "obvious stop words" such as the article "the."
The invention is designed to sift through foreign language documents, either in text, or "where the text may be derived from speech and where the text may be in any language," in the words of the patent.


QUOTE
SSL decryptors inside the network, like the boxes Sourcefire provide.

Cool, I didn't know of these before.

QUOTE
once you start doing pee-to-peer non-external-facing SSL connections, you are no longer communicating through the defined, clear area of a wide area network and therefore they cryptographic keys can't be acquired.

Obviously, but you can't read people's thoughts either (yet, at least not officially - well, actually I remember reading they were officially working on it a few months ago but you know what I mean).

But if they go out you can listen to them. Same here. No problemo for usual intelligence gathering imho, it's always been like that.

QUOTE
That's why secure peer-to-peer protocols and onion routing are so effective at encrypting data- because you can apply an additional layer of encryption to each transitional stage and therefore you need all the available private keys in order to decrypt the traffic.

Hm, my impression was more that such networks relied on the idea that you couldn't tell which peer really requested what data but that may well be assured by such encryption.

QUOTE
To put that into perspective, the NSA couldn't break the OTR-messaging encryption standard used for Apple iChat, let alone the AES encryption of the devices themselves. So, given that most terrorist organisations already use encrypted communication channels, what purpose would there be in hoovering up the 30% or so of all internet traffic which is encrypted seen as it can't be decoded anyway? Wouldn't that just produce a huge amount of unnecessary chaff?

I know what the (as usual) attention grabbing Gawker headline is screaming but there's nothing even remotely like that in there, lol. Read it yourself. They just say "check it out, Apple uses AES! So buy APPLE NAO BECAUSE APPPPPPPPLEEEEEEEEEEEEE!!!!!111111 Oh and guess who says it's totally secure? APPPPPLE!!!!! f*ck YEAH!!!" and they throw in a mention of the NSA approving AES for "top secret" info. First off after Wikileaks I think it became clear that "top secret" is by far not the most secretive designation there is, second again from Wikileaks and the things Manning has said that the "top secret" networks he had access to were wide open to hundreds of thousands of soldiers and people all over the world. On machines where they played Angry Birds, watched porn and heard Lady Gaga on. Security only helps until layer 8 comes into play.

QUOTE
The second is economies of scale. We're at a point of seeing approximately 200 petabytes of conventional internet traffic every day.

According to the Wikipedia numbers for the US internet backbone(s?!) you are probably correct smile.gif

QUOTE
A large percentage of this goes through US servers and service providers. Now, ISPs distribute their data handling capability- tens of thousands of terrabit-per-second junctions scattered across the country, but in order to hoover up all that data and to draw it off into a single repository you'd need an absurd collection infrastructure. We're talking about potentially hundreds of thousands, or even millions of individual taps. The US digital infrastructure just doesn't have the capability to effectively double already stretched capacity into transmitting captured data of this nature to a central repository.

All this requires is the cooperation of ISPs (which have no choice in the USA) and money. IIRC the NSA is not exactly an agency with very limited funding...

QUOTE
Another 20% is machine-based protocols that don't carry any human information, and can effectively be discarded. Why do you want to listen in on RPC and DHCP activity inside a network; it doesn't tell you much other than about the structure of the network.

20% of all (American) internet traffic? Hard to believe for me but can't claim/prove it was impossible, do you have a link by chance?

QUOTE
Why, also, would you want to listen in on the huge quantities of business-related data traffic which would have little to no intelligence purpose? But the question then becomes how do you determine what is what? ISPs provide for both businesses and individuals; MAC and IP addresses are just numerical representations of devices which don't actually possess any real identifying characteristics; the most you can determine from them in isolation would be the device manufacturer.

Well, if you truly have capacity (of the lines, sniffing equipment or storage) issues you could just eliminate that from the get go by using the ISP's billing database to separate between business and private. Then if you have no businesses you want to keep an eye on just don't watch; but we've seen massive surveillance systems are often (ab-)used to spy on companies as well.

QUOTE
Most externally facing IP addresses are dynamic, so the geography-specific data for them is only applicable for the time at which activity occurs from a single machine with a single dynamic address.

Usually not even that; when I connect from my PC through my router to my ISP my outside IP resolves to the location of my ISP who owns the entire IP range along with several others, so there's no way of accurately resolving my location from my IP alone. Unless you're my ISP or have their DBs of course, then you can find out which customer logged in from which line under what IP (which addresses your next point after the above quote, it's very possible and all ISPs do it all the time, unless you mean something different).
And that's the point here; there's, as usual, one point where all the data is just readily available: the ISPs. And with POTS going the way of the dodo and almost all communication even today being done via TCP/IP that just means all the more data to collect, easily available for anyone who asks (and can produce a badge or letter to show).

QUOTE
And that's quite aside from the fact that deep packet inspection capability- which the NSA already possesses- renders the requirement to store vast quantities of completely worthless data entirely null and void.

DPI costs a lot of resources. Storing sh*t costs nothing once you have the space (yes, electricity, cooling etc, yeah; but nothing impossible). Besides you can still do DPI later on, filter stuff like I said. But first you just save it away, ask warez leechers, they usually follow the same principle; it's much much harder to check everything 100% reliably before it comes in than just letting everything come in and deleting what you definitely don't need later, all you need is the capability to capture and attribute (=cooperation of telcos/ISPs + specialized hardware), store (data centers like the one in Utah, cloud etc.) and analyze (DPI, speech recognition etc.). They have all they need so it seems naive to assume they won't do it, or aren't doing it.

Yet (like someone on the Schneier blog and I think here as well) one could make the argument that it's in the NSA's (and other similar Agencies') best interest for us to overestimate them but I strongly disagree, I very much doubt such people, who are no doubt intelligent if reckless and with no regard for the law or "what is right", would be so short sighted.

An adversary overestimating you is only a good thing as long as it doesn't come to a fight; then, you're f*cked, and the other guy will laugh his ass off. The other guy being the "bad guys" here of course, terrorists, pedophiles, Nazis, muslims, whatever the scapegoat du jour is that they're using to justify what they're doing against their own people.

QUOTE
Cast your eyes back to your comments on the Utah Data Centre, and then consider two issues. One, that the data centre's primary purpose isn't to be a repository for all electronic communications in the US, but merely all electronic communications with an intelligence value.

And who gets to define what has an "intelligence value"? Besides, how do you know that? And is "intelligence value" something static or can it change when the next nut kills people with a pressure cooker?

Seriously man, this is all very very unstable... let's not act like we know all the details here, neither of us do (no offense to you or your job but I assume the innards or the NSA are foreign to you as well).

QUOTE
The NSA does much of their SIGINT work abroad, and the biggest target for data collection in the current economic and strategic climate is China. Also, whilst the trillion-terrabyte-storage capability sounds terrifying, it handily ignores the primary purpose behind having such computing power and storage capability- the breaking of complex cryptographic algorithms and cyphers.

Dude pick one tounge.gif First you said "the data centre's primary purpose isn't to be a repository for all electronic communications in the US, but merely all electronic communications with an intelligence value", then suddenly its "primary purpose behind having such computing power and storage capability [is] the breaking of complex cryptographic algorithms and cyphers".

I'll pretend there was no contradiction here and just answer the 2nd "primary purpose" you mentioned: are you actually proposing that the Utah data center will be used to STORE RAINBOW TABLES? lol.gif ² Sorry but that's classic.

I remember back when the NSA was looking for someone to program their quantum computers; it said that the people looking to apply should just assume that such a system existed. Why do I mention this? That was at the very least 5 years ago. And I'm still far from convinced that quantum computers, in the hands of some of the best code breakers in the world, can't really speed up the process of cracking stuff.

So I doubt that's Bluffdale's purpose. That seems to be storage, and perhaps analysis of the data stored there, whatever it might be. And like I said I doubt they're planning to store Yottabytes worth of rainbow tables.

QUOTE
The Wired Threat Levels article discussing the data centre goes into the value of this for intelligence gathering to a great degree.

Ah ok but I still had to make the joke before tounge.gif And you make it seem like the wired piece said the Utah data center would not be used to store comms which is not the case at all:
QUOTE
Under construction by contractors with top-secret clearances, the blandly named Utah Data Center is being built for the National Security Agency. A project of immense secrecy, it is the final piece in a complex puzzle assembled over the past decade. Its purpose: to intercept, decipher, analyze, and store vast swaths of the world’s communications as they zap down from satellites and zip through the underground and undersea cables of international, foreign, and domestic networks. The heavily fortified $2 billion center should be up and running in September 2013. Flowing through its servers and routers and stored in near-bottomless databases will be all forms of communication, including the complete contents of private emails, cell phone calls, and Google searches, as well as all sorts of personal data trails—parking receipts, travel itineraries, bookstore purchases, and other digital “pocket litter.” It is, in some measure, the realization of the “total information awareness” program created during the first term of the Bush administration—an effort that was killed by Congress in 2003 after it caused an outcry over its potential for invading Americans’ privacy.


------ this part was written yesterday; not that it makes a difference, just fyi; had to stop here.

QUOTE
The idea of "grab everything and then sort it" is a very Cold-war mentality which works pretty well with electronic intelligence gathered specifically from targeted individuals and nation states but is fundamentally flawed when it comes to accurate and timely assessment and dissemination of intelligence collected from the open-source macrocosm that is wider society. My issue with your comments is that you appear to have formulated an argument which doesn't correlate logically with taught or operational intelligence practice. I questioned your understanding of the intelligence cycle and modern intelligence analysis techniques because your argument doesn't appear to acknowledge how the intelligence cycle works, nor does it accurately portray the most effective ways of gathering targeted, specific intelligence on subnational violent actors- who are the primary security threat to the US mainland. One of the strongest arguments against your thesis is that it doesn't actually make sense unless you suspend the accepted knowledge of how intelligence analysis works- instead it required a Hollywood cultured understanding of intelligence which doesn't actually reflect reality.

Heh I read Schneier's movie plot threat contest too so don't lecture me tounge.gif Seriously though, I understand where you're coming from but I think your scope is too narrow.

You're only applying current standards, completely dismissing any possibility of things being done that are beyond the standards. And that I simply find naive, no offense, in the context of agencies such as the NSA. Yes as I believe I said before we might well think too much of them sometimes but tbh I'd rather be "safe" than right.

QUOTE
You have an ex-federal agent who says that the US has a great deal of capability in determining content and data from phone communications- big surprise.

Well, I addressed this earlier. Other than that, saying "big surprise" doesn't make it any better imho.

QUOTE
An AT&T operator saying that intelligence agencies are drawing down huge quantities of data to support investigative capabilities- hardly earth-shattering news.

Not to you and me maybe, but I doubt it hurts to get the "general public" informed as well.

QUOTE
And a noted constitutional lawyer with a history of speaking out against all kinds of perceived violation of civil liberties speculating about the capabilities of the US intelligence apparatus- how astonishing. None of these people have actually produced the same argument you have with the possible exception of the latter- but he talks more in terms of capability than actual active operational usage. You've made the fundamental mistake of extrapolating comments and extending them to what you see as a logical conclusion.

Again I didn't mean to imply anything was a surprise; it's just nice to have evidence to back up things others might like to refer to as "conspiracy theories".

QUOTE
I provided you with a list of other ways in which it would be perfectly possible to obtain the same data content without recording the entire conversation. How can they access this information? Well, the first port of call would probably be interrogating Tsarnaev's wife.

Oh come on. They (and Clemente) were talking about finding out precisely what was said; that's something entirely different than questioning his wife. If they meant that they could have said it, instead they said they'd try to find out exactly what was said - which is not open to interpretation.

The only way to find out exactly what was said is to have it on record in some way, whether it's just snippets (as you seem to imply by specifically saying "the entire conversation"), a text log or the conversation saved directly.

QUOTE
There are clearly numerous other ways in which this information could be obtained- external third parties, for instance. The actual phrasing of the comment itself seems to indicate investigators calling on members of the public and existing individuals with known relationships to Tsarnaev in order to obtain this information.

Can you point me to anything at all that makes it sound like that?

QUOTE
If they were interrogating a huge federal database of every conversation ever, why would they "want to find out" about it? Surely they'd already know?

I assume "keeping" or "maintaining" instead of interrogating? Because that's my point. You always assume that they absolutely have to (for whatever reason...?) check everything right away. I say they don't - they save indiscriminately. And analyze later. That way they know they have the records, but they simply haven't looked into them yet. And suddenly that "finding out" thing makes sense again.

QUOTE
I never said it was- that was your inference from my comments. I said it was the first time that it became part of a wide-ranging conspiracy theory that entailed the drawing down of all communications data.

Hold it right there, Echelon is no conspiracy theory, it's well proven. And it did entail going through a whole lot of data.

QUOTE
Are they? I'm pretty sure it hasn't been proven that the government are slurping up all communications data inside the US borders.

Not conclusively yet, no; many things seem to point to it though. However you know as well as I do that many so called "conspiracy theories" are proven true sooner or later, I think I don't need to mention the likes of Wikileaks etc. here.

QUOTE
Keyword analysis theory is pretty well known. Advertising is based almost solely on it. And yes, it is used in intelligence analysis.

Well like I said "keyword theory" as you called it before only led to marketing related sites so this makes a lot of sense smile.gif

QUOTE
MASINT

f*ck, for a second I thought they had masturbation intelligence now.

QUOTE
The people for whom interception may be worthwhile tend to use code and go to great lengths to obfuscate their activities, because they tend to have a relatively strong awareness of intelligence gathering capability and activity combined with the desire to not get caught.

True. That doesn't stop crazy "security" fanatics from wanting to know everything about everyone. Facebook does it, why should governments be different?

QUOTE
Keyword analysis has been utilised when codes used by organisations are known- one of the most famous ones in relation to violent Islamism is the discussion of weddings as a coded precursor to planning and perpetrating suicide bombings in particular- but this requires a pre-existing knowledge of the organisation which you don't possess by vacuuming up all the data available and running it through a slightly more intelligent version of Wolfram Alpha.

<< implying by "sucking up everything" suddenly all other kinds of intelligence gathering had to be stopped because... well, makes for a nice point.

I never claimed anything like that. And if you have prior knowledge of who to look for an increased amount of data stored makes it no harder to look for; don't act like they don't keep databases, querying which easily leads to the precise HDD the selected conversation is stored on.

Plus you got lots of (temporarily at least) useless data that you can look through later if you have to, as seems to be the case with the Tsarnaevs.

QUOTE
Intelligence analysis is an art.

Nope, I'm not gonna say it tounge.gif

QUOTE
It is scientific, but is not purely empirical. It is as much about cultural and human terrain analysis as it is anything else. Analysts work with an operational and strategic goal in mind- the strategic goal usually relating to the target on which intelligence is gathered, and the operational goal being the specifics of what ever macro-level analysis they are engaged in.

No offense but I don't think I've ever heard more marketing bullsh*t speak in so few words. Are you a politician on the side? tounge.gif I get your point but seriously.

QUOTE
You cannot just simply give a team of analysts a huge quantity of data and say "find bad stuff". That's not how it works.

Good thing I never claimed anything even close to that.

QUOTE
It is fundamentally flawed to think that intelligence analysis is the art of finding a needle in a haystack- which is what you seem to be implying it consist of.

Clear misunderstanding here; I never intended to claim that suddenly all US intelligence was only focused on getting everything. Nope, of course not. But I think it's becoming a part of the game, if it hasn't been for a while.

QUOTE
As I've said numerous times, large quantities of data actually inhibit the ability to analyse them.

Of course they do (in most cases, I have contrary examples, especially in console hacking, but that's a different topic so yeah). But I never claimed that they analyzed everything, only that they stored it; and for how long we can't even really guess of course.

QUOTE
You might have to hunt around through numerous journals in order to find valid experiments and you might have to analyse and assess their worth in the context of your strategic goal, but you have a point of focus. This seems totally logical, does it not, and that's largely how intelligence analysis works. Now, try doing the same if your data set and related experiments are still present, but are stored on something like JSTOR for which you have full access but no search function. You've got to manually sift through all the articles on anthropology, Greek history and media management theory in order to find your data set and sources. Seems a bit silly, doesn't it? That's how you seem to imagine intelligence analysis works.

As addressed right above, as well as before.

QUOTE
Simple- by already possessing a decent understanding of the organisation which you are targeting, and knowing what keywords they are most likely to use.

But my question was how one can gather intelligence that way, not judge/analyze it. And that you still can't.

What you're describing is some sort of blind spear fishing; you know who to target and the phrases they use but you can't see them, so you just hope you pick up something with the keywords you're looking for. Is that really how intelligence operates? I don't mean to doubt it but then it sounds less interesting than tactics employed by chinese corp hackers tbh.

QUOTE
Schneier, as in the cryptographer and security researcher Bruce Schneier? He's been pretty vocal in his statements that human-based macro-sphere applied intelligence analysis is vastly superior in all ways to automated intelligence trawling.

Indeed but you seem to have gotten the impression I claimed the US suddenly switched tactics; but I don't think that, I just think they keep adding to their repertoire.

GunWrath
  • GunWrath

    Needs More Beer

  • Members
  • Joined: 15 Jun 2012
  • United-States
  • Most Helpful [Expression] 2013

#45

Posted 11 May 2013 - 08:31 PM

Sweet mother of god, I don't normally say it, but that's definitely tl;dr.

Sgt. Foley
  • Sgt. Foley

    Big Homie

  • BUSTED!
  • Joined: 23 Oct 2011

#46

Posted 11 May 2013 - 08:42 PM

Hot damn, I agree with you GW.

OT:
I honestly don't care if the FBI has access to all of my emails, and phone convos. I barely call people on my phone, and I never email... unless it's for a problem with something.

sivispacem
  • sivispacem

    Absolute Dunkel:Heit

  • Moderator
  • Joined: 14 Feb 2011
  • United-Kingdom
  • Contribution Award [D&D]
    Contribution Award [General Chat]
    Most Knowledgeable [Vehicles] 2013
    Best Debater 2013, 2012, 2011

#47

Posted 11 May 2013 - 09:58 PM

Bag_wan

I've only had an opportunity to skim-read your response so far, and am currently rather too drunk to formulate anything even beginning to resemble a cohesive response. I will so, as and when I can, and given current circumstance I don't know when that will be, but may I just come out and say that it's extremely invigorating to have such a complex and in-depth discussion on such complex issues. Bravo.

baguvix_wanrltw
  • baguvix_wanrltw

    Cynical, yeah. Bitter, probably.

  • Members
  • Joined: 05 Apr 2012

#48

Posted 11 May 2013 - 10:41 PM Edited by baguvix_wanrltw, 12 May 2013 - 01:43 PM.

Thank you kindly sir smile.gif Again it'd have been nice if I could have kept my head - like you have - but still, nice discussion otherwise.

And take your time with the response, I'm in no hurry. And I definitely agree, this is a nice change of pace sometimes icon14.gif

Oh and sorry for the TL;DR walls of text... I don't know, maybe after sivi's response I'll try to update the OP with some of the points that were made?


EDIT: Just read through Schneier's comments section again and found another guy (Gulfie) who did some of the math involved, even though he doesn't address data compression, speech to text or anything else that could save massive amounts of space - still, same conclusion...
QUOTE
Recap :
How much does an average person talk on the phone a day? Assume 1 hour, 3600 seconds.
8KB/sec for an hour is ~30 MB
300 Million americans * ~30 MB = 9000 million million bytes ~ 9 Petabytes / day uncompressed.
9 Petabyte/ day
Given only slightly custom hardware : http://blog.backblaze.com/2013/02/20/... , $60k / PB .
9 PB/day * $60k / PB = $540,000 / day (purchase cost)
Half a million dollars seems like a lot of money, but it is less than a 4 hour flight in a B-2a St elth bomber and the DOD does that kinda of thing every day.. ( http://swampland.time.com/2013/04/02/... )
As for choosing and picking what to listen to, as mentioned before there are already databases that record all the calls... they are called billing databases. Duplicating the feeds and upsizing t he storage was done a while ago. ( http://www2.research.att.com/~daytona/ )
The tape version is even simpler, IBM sells it. Just get one of these IBM TS3500 Tape Libraries. Max system capacity at 3:1 compression, 2700 PB, or about 10 years worth of storage (http://www-03.ibm.com/systems/storage/tape/automation/ ) . ( I have not requested a quote ).
9 PB / day is not that much anymore. 9 PB /Day averages out to ~100 GB/sec ( 1TBit ). Even undersea cables are able to do 1 Tbit. ( http://en.wikipedia.org/wiki/...
Aggregation and capture of data? I believe that AT&T room 641A answers that question. Yes, and yes. ( http://en.wikipedia.org/wiki/Room_641A ) Legal indemnification of Telcos keeps cropping up. ( http://www.democrati...nd.com/discuss/... ) If they didn't need it, they wouldn't be getting it.
Result: Totally doable, almost off the shelf at this point. Trapping the traffic… totally possible, legally muddy. There is nothing to contradict the guy's story.

Ah damn and while we're at it of course Clive Robinson makes nice points as well... I'd paraphrase but he put it very well imho.
QUOTE
@ Buck,
For argument's sake let's say we do have 100% accuracy for speech to text. How does one ascribe metadata for code speak? Inside jokes? Sarcasm? Good luck!
You don't need to.
Firstly have a look at how Google does language convertion, it does not use a dictionary as you or I would do. It uses the statistics of the data it has with regards the vast tract of various languages. In effect it translates sentances within the context of the document and other documents it has. The result is it does understand slang, jokes and various other nuances of day to day usage of a language.
The problem with people tryying to convay other information is that it forms styalised speach at variance to the norm. This changes it's statistics just like any other form of steneography and there are known ways of detecting it as such.
As for speach to text not being 100% reliable it realy does not need to be better than around 50% to get usefull results when you are analysing not as words but sentances and paragraphs within a general context.
The reason being you need to turn the problem on it's head, they are not looking to find a needle in the hay, but how to identify hay so it can be thrown away. And this can be done by successive filters. Thus the filters can be very crude rejecting only one or two percent at each stage but still producing high quality results (to see why go and look up how a uranium enrichment process works where each centrefuge only provides a fractional percentage of enrichment).

Triple Vacuum Seal
  • Triple Vacuum Seal

    Ghetto Star

  • Members
  • Joined: 02 Dec 2011
  • United-States

#49

Posted 12 May 2013 - 08:35 AM Edited by canttakemyid, 12 May 2013 - 08:37 AM.

Hypothetically speaking, the average Joe wouldn't even have another human being assigned to him in some intelligence agency. It would all just be retrievable information on a need to know basis at best. It comes with the territory of having a state of the art military and intelligence community; which I ultimately appreciate. In the worst case scenario where some NSA creep is rummaging through my personal activities and history, then...


I can say with a straight face that I don't care what some computer whiz's in Washington whom I will never meet in my entire life knows about me and my digital footprint. Jokes on him/her. I'm only interesting in person. lol.gif




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users