The Business-Technology Weave

Apr 23 2013   2:55PM GMT

Big Data and the FBI

David Scott David Scott Profile: David Scott

Big Data - FBI[Note:  Nothing in the following post is meant to be a criticism of the FBI’s agents – those people are supreme, and I thank all of them for the job they do.  But as to the executive management, and government bureaucrats and politicians, it’s a fair question to ask:  What year are these people living in?]

The events in Boston were horrific and, as trite as it sounds, I express my condolences to the victims and families of those killed and injured.

I was further impacted by Krystle Campbell’s mother, Patty, and her statement.  I won’t belabor nor detail her heart-rending appearance on camera, not long after the bombing, but she said something that I… “took away” for lack of better words.  Actually, she said many things in that regard, but one thing in particular made me think:

She mentioned Krystle as having “worked so hard…” – she was achieving, she was living, she was motivated.  It made no sense to her mother that someone like Krystle can be gone due to such senseless circumstances.

And again, it sounds trite, but I thought:  How often have I wished that I worked a little harder?  Whether to stick to my exercise regime, to eat right, to deliver a little faster to a client, to get along with family members, to get my damn lawn mowed, or just giving someone a hand – whatever it is I’ve got going.

Whatever it is you’ve got going:  I’m lousy at preaching, and I’m often lousy at living, but why don’t we all work a little harder?  In remembrance of Krystle, we can strive for a bit more excellence.  Let’s do better.  I’m not exactly the sensitive type, but knowledge of Krystle and her life came to me through terrible circumstances, but perhaps she lives on if someone out there does something good by virtue of Krystle’s ethic and hard work.  And somebody please –  maybe even me, maybe even everyone reading this – please do it on a consistent basis – and if that happens, can Krystle really be gone?  That sounds trite, a little stupid, and perhaps naïve.  But it’s all I can offer at the moment – and it’s weak.

Well, in a crazy way, that brings me to the FBI and Big Data – ‘cause I sure am getting tired of terror.

Major news organizations, like this one, have asked:  Why did it take four days to find the Boston bombers?  They had the elder brother, Tamerian Tsarnaev, on file – having been warned by the Russian government (no less) that he had radical ties, and was a potential terror risk.

As Peter Foster, US Editor of the U.K.’s The Telegraph so succinctly puts it:

“I’m not sure what the FBI equivalent of Google is, but surely “Boston + terrorist + jihad watch list + Muslim Community” should have thrown up the name “Tamerlan Tsarnaev.”

Within hours of the bombs, the FBI should have had ready access to records of people in the Boston area that had anything terror-related in their backgrounds – which of course this guy did.  Here, the FBI didn’t even have to do a “known associates” scrounge – they’d have seen that the younger brother was a student – and would have gone to the campus, and found the guy in the gym!  Or at that campus party he went to – in the days after the bombing.

Foster again:  “Even more staggering is that when they did eventually isolate Tamerlan and his younger brother Dzhokhar from the hours of CCTV and smartphone footage taken at the race – a process that took three precious days – even then the FBI didn’t make the connection.”

In my book, I.T. Wars, I actually discussed back in 2007 the FBI’s failures with their VCF (Virtual Case File) system – a project that was so screwed up it was abandoned even though it was mounted specifically to meet the challenges of the post-9/11 world.  It was superseded by a system called Sentinel, which has its own problems.  Do you know, in more than 25 years in high-level IT, I’ve never abandoned a project?  I spec and mount them too well.  And, I’ve written Project Management Frameworks for Fortune100 enterprises, as well as the Pentagon – so it’s not like I haven’t had challenges during my career.

I’m sure many others here too harbor similar records.  People in IT take pride in being empirical, and in fitting technical solutions to whatever the business at-hand is.  Can we get simple Big Data concepts implemented at the FBI, and other government agencies that work in harmony, to thwart terror?  –  and, when bad outcomes do happen, can we please make more efficient responses?

There’s also reportage that the FBI missed the elder brother’s trip to Russia due to a misspelling on a manifest of some sort.  Even Google coughs up ‘near-enough’ suggestions when I do searches, and make typos.

To repeat:  This is not a criticism of the FBI’s agents – those people are supreme.  But as to the executive management, and government bureaucrats and politicians, it’s a fair question to ask:  What year are we living in when Big Data is not making already known data serve us best?

To our government and key agencies I say:  Please swing Big Data for all it’s worth.  And, just to note, this is not a “Big Brother”-type nightmare.  This is the effective leveraging of data the government already has.

Make it work.

7  Comments on this Post

There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when other members comment.
  • davidcarr
    Great Post. I agree that our law enforcement agencies must be aware of and use the technologies that are available today and leverage them to be better at their jobs. This is not about spying or control. In this day and age, it is common sense.
    265 pointsBadges:
  • TomLiotta
    Criticism of the handling is fine, but it's not clear what any suggestions are for what should have been done differently. The quip by Foster on a Google-like search is bizarre at best. What was the evidence up until the capture that there was any reason to think specifically of "jihad watch list" or "Muslim Community"? Is there any reason to think that that exact search was not done? Was there reason to think that many similar searches were not also done? (Maybe it was pure domestic terrorism. Maybe a disgruntled racer did it. Maybe it was...?) Was there reason to think there weren't many, many suspects and that Tamerian Tsarnaev wasn't on any search results along with many, many others? Then, is there certainty that the connection wasn't quickly made after analyzing video? Does anybody have a good idea of how much video had to be analyzed? How much showed nothing useful and merely wasted time? (Too much info can be as bad as too little.) Do we know how long between the time useful video was actually received and the time it was recognized? (I wouldn't expect it to be the very first video watched.) How many video processing systems were instantly available? Do we know how many potentially dangerous faces actually were recognized? Do we know how much video was good enough quality? This isn't exactly "big data"; it's a huge amount of generally low or medium quality video that might have something useful in it. And nobody knew what (who) was being looked for ahead of time. Personally, I'm impressed enough that it was resolved as fast as it was. It was very likely a big confusing mess. -- Tom
    125,585 pointsBadges:
  • David Scott
    Hi Tom: Excellent points you make here, enjoyed your analysis. My thoughts (read the "run up" articles about Big Data, likely you have). Post-bombing, the FBI should have, as the UK editor suggested, merely searched their own trove of info and had an identification made on late Day One - we know they had the brothers ID'd as, minimally, persons of threat by Russia - the FBI didn't leverage their own data. When you consider that the FBI must have had a data-record, in existence, hopefully available through a search engine of theirs, that had the following fields: Name, location (Boston), threat-info-origination (Russia), Purported Terror-links/Connection (expressions regarding/ interest in Jihad), Known Associates... I could go on. I agree the video review was an enormous task. I believe elements of prior prevention weren't exercised effectively, either, but I'll leave that to other pundits at other forums/news sites as, I think you'll agree, it is not TechTarget's, nor my, domain. [If a TechTarget editor would like to contact me regarding latitude here, I'll feel more comfortable exploring this further, as Tom's points are very valuable, and this discussion is important to the safety and security of us all]. This article is to help promote Big Data, and BD can be used in a variety of solutions, to include business; to include the thwart of terror or minimization of it. My main thrust is to promote Big Data for business leverage, but the proximity in time made me treat the subject for the Boston event: Maybe someone at the FBI will read the post and be influenced in a positive way - one never knows. Thanks for commenting, good thoughts all around.
    125 pointsBadges:
  • David Scott
    Thanks for the reinforcement David regarding the use of this type of data. I figure if the government has this data anyway, they may as well wag it for all it's worth - there are two specific types of terror-event that I'm deeply concerned about that make getting this right absolutely imperative.
    125 pointsBadges:
  • LoserInTheMiddle
    These are, essentially, government bureaucrats.  Not the agents in the field or even their immediate supervisors - but the support staff "behind the scenes" - those bureaucrats.

    And you and I both know - along with most of the regular readers on this site - that large, highly-structured organizations without an overriding incentive to change are not easily given to innovative solutions or "new" ideas.

    Big Data is 21st Century, although the beginnings of it date back several decades.  Only in the recent past has "big data" come into its own in business.
    So, in the eyes of a government employee whose been "with the bureau" even for just 20 years. "Big Data" is "new". 

    And therefore unproven, untrusted, unknown, and "needs more study" before a "pilot project" can be included in the budget request for "next fiscal year".

    Seriously - this is how bureaucrats THINK.  It doesn't matter to them how many more die in IED attacks on our soil - they've got to have their "process" followed and to hxll with the American People.

    Bureaucrats is the dirtiest swear word I know.
    10 pointsBadges:
  • David Scott
    This should prove interesting to all who view it:
    125 pointsBadges:
  • TomLiotta
    Again, what evidence is there that the name did not come up on Day 1? And if it did, how many other names would also come up?   There simply was no reason to connect the event with a specific individual at the exclusion of many others in such a metropolitan region on such a busy day. Now, if a piece of shrapnel was found with a Chechen freedom symbol on it or some similar indicator, then the potential list might be trimmed greatly.   Beyond that, thanks for the link to the chart. It matches with everything I've run into about the case and adds some details that seem plausible. The bit about the misspelled name points to examples of why such lists can be trouble. (Remember Sen. Kennedy being delayed due to a misidentification?)   Maybe more interesting, and maybe more troublesome, the FSB seemed to have more information about someone who'd been legally in the U.S.A. for many years than U.S. agencies had. How would they know? And why were there no responses from them after requests for follow-up details? Would it compromise their information gathering activities?   To me, that all starts getting to the point of Big Data. There is actually too much data and too little "information" available in this kind of case. It's easy to draw connecting lines from event to perpetrator after the identity is known.   But before then, there are far more possible lines than we will ever have information to make judgments about. None of the data necessary for judgment is publicly available, only conclusions are known. How many lines can be drawn from 'two bombs in Boston' to possible suspects within a few hours drive of Boston? We simply don't know, but I suspect it's actually a fairly complicated set. There wasn't even a motive to help narrow things down.   How can "Big Data" help with too few data points?   As for "bureaucrats", I'd say that that's off into political discussion. I could go much farther into that area, but I'd guess it's too far off of the BD trail.   Tom
    125,585 pointsBadges:

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

Share this item with your network: