# Paul Revere as an example of the trouble with the collection of metadata **URL:** \ **Category:** In My Humble Opinion\ **Created:** [June 18, 2013, 3:24pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275 "2013-06-18T15:24:21Z")\ **Posts on this page:** 11\ **Page:** 1
**Author:** ![Quartz](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/quartz/32/267_2.png) [@Quartz](https://boards.straightdope.com/u/Quartz)\ **Post date:** [June 18, 2013, 3:24pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/1 "2013-06-18T15:24:21Z")
With all the brouhaha about the NSA spying on everyone, I found [this article](http://kieranhealy.org/blog/archives/2013/06/09/using-metadata-to-find-paul-revere/) of interest. It shows how Paul Revere was a person of interest merely using metadata. ---
**Author:** ![Greg\_Charles](https://avatars.discourse-cdn.com/v4/letter/g/839c29/32.png) [@Greg\_Charles](https://boards.straightdope.com/u/Greg_Charles)\ **Post date:** [June 18, 2013, 4:56pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/2 "2013-06-18T16:56:42Z")
That’s pretty interesting. Ironically though, I enjoyed the data and it analysis (which I assume is genuine) more than the satire. ---
**Author:** ![Smapti](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/smapti/32/17938_2.png) [@Smapti](https://boards.straightdope.com/u/Smapti)\ **Post date:** [June 18, 2013, 5:33pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/3 "2013-06-18T17:33:31Z")
So the argument here is that police shouldn’t use evidence available to them to stop anti-government insurgents from organizing terrorist militias to engage in armed resistance against a lawful use of police power? Color me unimpressed. ---
**Author:** ![Ludovic](https://avatars.discourse-cdn.com/v4/letter/l/7ab992/32.png) [@Ludovic](https://boards.straightdope.com/u/Ludovic)\ **Post date:** [June 18, 2013, 5:49pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/4 "2013-06-18T17:49:55Z")
> [@Smapti](#): > > So the argument here is that police shouldn’t use evidence available to them to stop anti-government insurgents from organizing terrorist militias to engage in armed resistance against a lawful use of police power? > > Color me unimpressed. I’m reminded of the snippet I heard on NPR recently when William F Buckley was talking to a Black Panther, who before the talk started asked _Buckley_ a question: would you have supported the American Revolution. Buckley demurred, saying that he was unsure as he was conservative by nature, at which point his guest expressed his admiration for this admission rather than the expected display of hypocrisy. ---
**Author:** ![UncleRojelio](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/unclerojelio/32/3160_2.png) [@UncleRojelio](https://boards.straightdope.com/u/UncleRojelio)\ **Post date:** [June 18, 2013, 7:07pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/5 "2013-06-18T19:07:24Z")
> [@Smapti](#): > > Color me unimpressed. Yeah, a national gun registry would be more effective. ---
**Author:** ![etv78](https://avatars.discourse-cdn.com/v4/letter/e/d78d45/32.png) [@etv78](https://boards.straightdope.com/u/etv78)\ **Post date:** [June 18, 2013, 7:22pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/6 "2013-06-18T19:22:57Z")
> [@Greg\_Charles](#): > > That’s pretty interesting. Ironically though, I enjoyed the data and it analysis (which I assume is genuine) more than the satire. +1 Loved it! ---
**Author:** ![Chessic\_Sense](https://avatars.discourse-cdn.com/v4/letter/c/7c8e57/32.png) [@Chessic\_Sense](https://boards.straightdope.com/u/Chessic_Sense)\ **Post date:** [June 19, 2013, 1:14am UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/7 "2013-06-19T01:14:25Z")
I love the math, and the writing is tickling, but I have to say his analytical acumen needs some polishing. Take this, for example: > [@](#): > > We started with a list of memberships in various organizations. But now suddenly we do have a social networke of individuals, where a tie is defined by co-membership in an organization He slips up in three ways. One, he assumes he has a full roster of organizational membership. Two, he assumes each person’s entry is accurate. Three, he defines, without support, that co-membership means they know each other. The issue with this is that you’ll often find yourself with circular conclusions. Let’s say I took a phone book for a small city and along with other data, I made the same roster. If I assume that the phone book has given me every person in the city, I’d be making a big mistake. There are unlisted numbers, so people that should be in the book are missing, and people die, so people in the book shouldn’t be there. Basically, you end up proving- not what the city looks like- but that the city looks like your book. Taken as a whole, you’ll find yourself missing nodes, missing connections, and having false connections. This is a BIG deal because the conclusion that Paul Revere was a “person of interest” was based not only on his ties to others (what if he doesn’t really know all those people?) but also on his unique position as the only person linking the groups (what if there are a dozen others just like him that didn’t get their names in the book?) Finally, what if someone that you think is isolated to only a single group or maybe two _actually_ belongs to many more of the groups? Taking all this into account means you’ve got poor Paul not really connected at all, and you’re missing a slew of “Paul-alikes” that dilute his importance. So essentially, the author cheated by granting himself amazing data to start with. He also cheated in another way - he looked at Paul’s records without a warrant. Naughty, naughty. ---
**Author:** ![BigT](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bigt/32/12044_2.png) [@BigT](https://boards.straightdope.com/u/BigT)\ **Post date:** [June 19, 2013, 1:49am UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/8 "2013-06-19T01:49:54Z")
> [@Chessic\_Sense](#): > > Taking all this into account means you’ve got poor Paul not really connected at all, and you’re missing a slew of “Paul-alikes” that dilute his importance. So essentially, the author cheated by granting himself amazing data to start with. > > He also cheated in another way - he looked at Paul’s records without a warrant. Naughty, naughty. Wasn’t it all public information, which would not have needed a warrant? And the fact that the data isn’t reliable is kinda the entire point, is it not? Revere has now been proven as a person of interest, so they can now get a warrant. Even though there are many flaws in said proof. ---
**Author:** ![Gatopescado](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/gatopescado/32/5711_2.png) [@Gatopescado](https://boards.straightdope.com/u/Gatopescado)\ **Post date:** [June 19, 2013, 5:15am UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/9 "2013-06-19T05:15:36Z")
“Person of Interest” all depends on which side your on. What is your point? I’m sure Eisenhower was a “Person of Interest” in Germany on June 4th, 1944, “metadata” notwithstanding. I guess I just don’t get it? ---
**Author:** ![delphica](https://avatars.discourse-cdn.com/v4/letter/d/e9a140/32.png) [@delphica](https://boards.straightdope.com/u/delphica)\ **Post date:** [June 19, 2013, 4:11pm UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/10 "2013-06-19T16:11:43Z")
I don’t think I get it either, but I very much enjoyed reading the article. I’m pretty liberal, I LOVE Paul Revere and have long had a serious crush on David Hackett Fischer. I thought the writing style of the article was delightful (“I have an upcoming EDWARDx talk about it. You should come.” had me laughing like a drain.) But it doesn’t really work as a good example of THE TROUBLE with metadata, it’s more like really good evidence that this kind of data collection and analysis yields awesome results. Paul Revere SHOULD HAVE BEEN a person of interest to the government of the time. Fischer’s book on Paul Revere makes the case, btw, that he was dangerous/successful not because of his midnight ride, but because he knew every man, woman and child in the colonies with a special emphasis on those connected to the revolutionary ferment, and actively cultivated these networks. It’s a terrific read. ---
**Author:** ![Chessic\_Sense](https://avatars.discourse-cdn.com/v4/letter/c/7c8e57/32.png) [@Chessic\_Sense](https://boards.straightdope.com/u/Chessic_Sense)\ **Post date:** [June 20, 2013, 2:10am UTC](https://boards.straightdope.com/t/paul-revere-as-an-example-of-the-trouble-with-the-collection-of-metadata/661275/11 "2013-06-20T02:10:30Z")
> [@delphica](#): > > I don’t think I get it either, but I very much enjoyed reading the article. > … > But it doesn’t really work as a good example of THE TROUBLE with metadata, it’s more like really good evidence that this kind of data collection and analysis yields awesome results. > [@Gatopescado](#): > > “Person of Interest” all depends on which side your on. What is your point? I’m sure Eisenhower was a “Person of Interest” in Germany on June 4th, 1944, “metadata” notwithstanding. The point here, folks, is that metadata is A LOT of information that is very useful. The author isn’t making an argument about the recent news, I don’t think, but rather rebutting the defense “Who cares? It’s just _metadata_! It’s not like they know the content.” > [@BigT](#): > > Wasn’t it all public information, which would not have needed a warrant? > > And the fact that the data isn’t reliable is kinda the entire point, is it not? Revere has now been proven as a person of interest, so they can now get a warrant. Even though there are many flaws in said proof. 1. You can’t just go on a fishing expedition to find “persons of interest.” You can’t just go crunching data on a massive table like that. You can get in serious trouble for doing that, _especially_ when US Persons are involved. 2. My point is that he cheated in the setup. He gave himself reliable data and open access to it. He was only able to prove Revere as a person of interest because he knew ahead of time that he would be. Whoever is [red-teaming](http://www.govinfosecurity.com/interviews/red-team-primer-i-244) you would go “How do you know there aren’t a hundred Revere-alikes that you haven’t captured in your dataset? And how do you know the reliability of these connections? … Wait, wait, you’re just _assuming_ he knows everyone in the entire organization?! And you’re just _guessing_ you’ve got everybody on the map? BWAHAHAHA!” and your commander would be embarrassed for you.