Tag Archives: Osama bin Laden

Using Data to Find Bin Laden

Map picture

Where they thought he was and where he was.

Great piece — Geographers Had Predicted Osama’s Possible Whereabouts – ScienceInsider (thanks Daily Kos- Geographers predict Osama’s location) which tells the story of Thomas Gillespie, a UCLA geographer

who, along with colleague John Agnew and a class of undergraduates, authored a 2009 paper predicting the terrorist’s whereabouts, were none too shabby. According to a probabilistic model they created, there was an 88.9% chance that bin Laden was hiding out in a city less than 300 km from his last known location in Tora Bora: a region that included Abbottabad, Pakistan, where he was killed last night.

Here’s their original paper: web.mit.edu-mitir-2009-online-finding-bin-laden.pdf. It’s not as if these guys identified the town correctly (and the Science article has had to backtrack on some of its original assertions and the comments aren’t kind), but they got a lot of things right: They figured out he was much likely to be in a house than a cave, and in a relatively large town rather than a village, and that he was in Pakistan rather than somewhere else. They also predicted the kind of building he would be living in. In the end they were less than 300 km off.

Not bad, when you look at what the CIA was saying about him before (of course, they may have been trying to put people off the scent, but we know that it was only earlier this year that they had an idea he might be in the house:

Osama bin Laden’s Death on Twitter

(Updated timeline to include subsequent accounts)

There was, by all accounts, no Internet or phone access to Bin Laden’s compound. Had there been, might he have known about the attack in advance from social media?

This depends on what was being said on twitter, and when. Although lots of people in Pakistan are on Facebook, twitter would have been more useful. There’s no clear timeline yet about when the US launched its attack on the compound. But had Osama’s people been monitoring the keyword ‘abbottabad’ (or people who had previously mentioned the word), which would have been smart, they would have known that something was afoot:

image

notice that ReallyVirtual’s tweets are half an hour before the first news reports of the crash. His first is at 00:58 am local time:

twitter guy

his second seven minutes later:

image

But would that have been sufficient warning?

Almost certainly not. By then the operation was already over, I believe. Here’s the timeline as best as I can figure (all times are Pakistan time, i.e. GMT +5, tweets come from ReallyVirtual unless stated):

00:00 (just past) Seal helicopter take off from Islamabad?
In Pakistan, it was just past midnight on Monday morning, and the Americans were counting on the element of surprise. As the first of the helicopters swooped in at low altitudes, neighbors heard a loud blast and gunshots (NYT)
00:35 ISL 5/2/11 First helicopters arrive on the scene, according to Pakistan news blog
NYT puts it, logically, a little bit earlier, as it called it a 40 minute raid)
00:32 Obama returns to the Situation Room for additional briefing. (Timeline- The Raid On Osama Bin Laden’s Hideout – NPR)
00:50 Obama first learns that bin Laden was tentatively identified. Shortly after the raid, Pakistani leaders are briefed of the actions. (NPR)
00:58 Helicopter hovering above Abbottabad at 1AM (is a rare event).
01:05 Go away helicopter – before I take out my giant swatter :-/
? One of their helicopters stalled and could not take off. Rather than let it fall into the wrong hands, the commandos moved the women and children to a secure area and blew up the malfunctioning helicopter. (NYT)
01:09 A huge window shaking bang here in Abbottabad Cantt. I hope its not the start of something nasty :-S
01:10 As they took off at 1:10 a.m. local time, taking a trove of documents and computer hard drives from the house, the Americans left behind the women and children. A Pakistani official said nine children, from 2 to 12 years old, are now in Pakistani custody. (NYT)
01:30 Pakistan News: Helicopter Crashed in Abbottabad, Pakistan (story here, appears about a minute earlier)
   
01:30 No one is picking phone in Abbottabad, not even the landlines. (m0chin)
01:38 Just talked to family in Abbottabad, say they heard three blasts one after another, don’t know what really happened. (m0chin)
01:43 Hello sir, any update on the blasts? What has really happened? (m0chin)
   
01:44 all silent after the blast, but a friend heard it 6 km away too… the helicopter is gone too.
01:45 OMG :S Bomb Blasts in Abbottabad.. I hope everyone is fine :(  (han3yy)

Here’s a timeline courtesy of tiki-toki:

Seems that despite the fact that twitter broke news of the attack, the guys in the compound wouldn’t have been any better of if they’d been following it.

The New Newswire: a Dutch Student Called Michael

Twitter is now a news service in its own right. ReadWrite Web, an excellent website dedicated to Web 2.0 stuff, points out that the recent earthquake in England–not that unusual in itself, apparently, but rarely actually strong enough to be felt by humans—was reported first by Twitterers and by a Twitter-only news service called BreakingNewsOn (www.twitter.com/BreakingNewsOn): 

This story broke over Twitter in the past half hour, and nothing is up yet on the BBC sites, the Guardian, or the Telegraph. This story is breaking live on Twitter.

Looking at the situation a few hours later, it’s certainly true that mainstream websites have been a bit slow with the story. From what I can gather, the timeline is something like this (all times are in GMT):

Quake hits south of Grimsby 00:56  
First tweets 00:57  
BreakingNewsOn 00:59 (“Unconfirmed reports of earthquake in London”)
BreakingNewsOn 01:01 (“Reports of earthquake, working to confirm”, followed by lots of tweets)
BreakingNewsOn 01:10 (confirmation from European-Mediterranean Seismological Centre)
Dow Jones Newswires 01:29 (quotes BBC report)
Associated Press 01:30 (garbled alert)
Reuters 01:36 (“Quake shakes Britain, no casualties reported”)
AFP 01:45 (“Moderate quake shakes Britain”)
BBC twitter feed 01:56 (“Tremors felt across England”)

There may be some holes in here: I don’t have the exact time when the BBC website first carried the story, but I’m guessing it’s a few minutes before the wires. And this is not the first BreakingNewsOn has been ahead: It was, according to some reports, first on the Benazir Bhutto assassination, although I’ve not been able to confirm that. 

So who or what is BreakingNewsOn, and how does it scoop the big guys on their own turf? The service is actually pretty much one guy, a 20-year old Dutch student called Michael van Poppel, according to this interview by Shashi Bellamkonda. He is a news junkie, and makes money from it too, doing something called web-trawling—searching the net for stuff he can sell to the big players. (He was the guy who last September dug up a videotape of Osama bin Laden, which he then sold to Reuters.) 

Van Poppel works with a couple of other people and is clearly experienced and voracious in hoovering up web content. But it’s also about citizen journalism, crowd sourcing, whatever you want to call it: in the case of the UK quake, the first alerts actually came from witnesses, who twittered about the jolts they felt; it was BreakingNewsOn’s skill in harvesting that information, and staying sufficiently close to its readers for them to think to share their experience, that led to the fast turnaround. 

Of course, there’s much about this that is new. Everyone is now a reporter, if they find themselves in the middle of news. And everyone can be a media publisher: In this case it’s one 20-year old student with a twitter feed and an Internet-connected computer. And, finally, everyone can now subscribe to that once holiest-of-holies: a newswire service that updates in real time. Only now it’s not called a Reuters terminal or a Bloomberg but Twitter. 

But behind that, not much has changed. I’ve covered a few quakes in my time, and it’s all about finding the stuff out quickly by getting it out quickly. Nothing much has changed. No one was injured or killed, and it sounds like there was no falling masonry or damage to buildings. But that’s no excuse: earthquakes are news, and especially if they’re the strongest in the country for more than two decades

Twitter is perfectly suited for breaking news, because it’s all about short pithy sentences and updates. As ReadWrite Web points out, during the California wildfires last year, Twitter and other citizen journalism tools were used by people on the ground, scooping the mainstream press. And all this offers some lessons for the mainstream press that it would be wise to absorb: 

  • Mainstream media cannot afford to be slow off the mark on stories like this, since their value to high-paying subscribers is intimately tied to their speed;
  • Alert streams are no longer the province of market traders;
  • Traditional media needs to find a way to work with these new sources of news, or else find a way to add value that such services cannot. In this case it could have been finding a way to reflect in the headlines the unusual nature of this event;
  • Traditional media has to both monitor these new sources of news–the tweets from ordinary folk surprised to be shaken awake by a tremor—and work with them to ensure that they, too, benefit.

Some might say that what van Poppel does isn’t news. I’d contest that. He did everything right in reporting the story: it’s big enough an event to merit an “unconfirmed” snap, a quick follow-up which contains what we old newshounds would call an advisory letting subscribers know what he’s doing and to expect more. When he got confirmation he put out, all within 10 minutes. That’s a time-tested, old-fashioned and reasonable news approach. He leveraged the new media, but he showed an understanding of news values and what his readers needed. 

Kudos to him. We all could learn a lesson.

(An extended version of this post is available for publication to newsprint media as part of the Loose Wire Service. More details here, or email Jeremy Wagstaff directly.)

The Digital Fallout Of Journalistic Plagiarism and Fakery

How do you correct the Internet?

All these reports of plagiarism and fakery in U.S. journalism — at least 10, according to the New York Times — raise a question I haven’t seen addressed elsewhere. What should newspapers and other publications which have carried the reports do about setting the record straight?

A USA Today report says of disgraced reporter Jack Kelley that it has “found strong evidence that Kelley fabricated substantial portions of at least eight major stories, lifted nearly two dozen quotes or other material from competing publications, lied in speeches he gave for the newspaper and conspired to mislead those investigating his work.”

Here’s a taster: ”An extensive examination of about 100 of the 720 stories uncovered evidence that found Kelley’s journalistic sins were sweeping and substantial. The evidence strongly contradicted Kelley’s published accounts that he spent a night with Egyptian terrorists in 1997; met a vigilante Jewish settler named Avi Shapiro in 2001; watched a Pakistani student unfold a picture of the Sears Tower and say, “This one is mine,” in 2001; visited a suspected terrorist crossing point on the Pakistan-Afghanistan border in 2002; interviewed the daughter of an Iraqi general in 2003; or went on a high-speed hunt for Osama bin Laden in 2003.”

That’s quite a lot of correcting to do. USA Today says it will withdraw all prize entries it made on Kelley’s behalf (including five Pulitzer nominations) and “will flag stories of concern in its online archive”.

But is that enough? Correcting the “online archive” would have to include all secondary databases such as Factiva (part-owned by Dow Jones, publisher of the Far Eastern Economic Review, The Wall Street Journal, and my employer; There are 1,495 USA Today stories with Jack Kelley’s name either on them or in them prior to this year). Strictly speaking, it should also include all Internet copies of those stories on the Internet (a Google search of [“Jack Kelly” and “USA Today”] threw up 3,470 matches; while many of those are accounts of the plagiarism charge, many precede that). And what about blog references to Kelley’s stories?

I’ll take an example. In 2001 Jack Kelley wrote about a vigilante Jewish settler named Avi Shapiro in 2001. According to USA Today, this was one of the stories where “the evidence strongly contradicted Kelley’s published accounts”. That story has been posted on dozens of websites (I counted 60). Who’s going to correct, or raise flags on all those?

Then there’s the doubt. With Kelley claiming, according to the USA Today report, that he was “being set up”, there’s no way that even a serious investigation by the paper (which included a eight-person team, a 20-hour interview with Kelly by three veteran journalists from outside the company and extensive use of plagiarism-detection software) is going to confirm with any sense of certainty what was faked or plagiarised. So what, exactly, do you correct? Do you delete his whole oeuvre?

It’s a tough one, and perhaps a sober reminder for journalists (and bloggers) using the Internet as a source that it’s not just emails that appear to come from our bank that we need to double check. Is there a technological solution to this? A digital watermark or trace that can allow someone to instantly correct a story, or at least notify those hosting the material that there’s a problem?