Depth Reporting

Trade in your clunker data for new data

2009-11-06T11:59:00.004-05:00

You can download data on 677,081 cars traded in the Cash for Clunkers program from Cars.gov. The AP has already analyzed it and found that one in seven of the new vehicles purchased in exchange get only a few miles per gallon more than the so-called clunkers:

The single most common swap — which occurred more than 8,200 times — involved Ford F150 pickup owners who took advantage of a government rebate to trade their old trucks for new Ford F150s. They were 17 times more likely to buy a new F150 than, say, a Toyota Prius. The fuel economy for the new trucks ranged from 15 mpg to 17 mpg based on engine size and other factors, an improvement of just 1 mpg to 3 mpg over the clunkers.

Some of the data that would make it easier to spot fraud -- like the names of the consumers who traded in the clunkers -- has been excised.

I downloaded and played with it a bit. You need to be cautious because there's some data strangeness in there, such as multiple clunker odometer readings of 9,999,999 miles, and clunkers whose model year is listed as 2008, but here's my quickie state-by-state visualization on Many Eyes of clunkers per 10,000 people:

There are more clunkers per capita in the North than in the South and West. I'm guessing that has something to do with the amount of salt on the roads.

Here's the number of clunkers traded in by model year, which just makes all the more obvious that I missed a golden opportunity to trade in my 1995 Saturn. Turns out 1995 was the peak clunker year.

Tens of thousands of military Social Security numbers published on Library of Congress Web site

2009-11-05T10:19:00.005-05:00

Until this week, you could find the Social Security numbers of tens of thousands of current and former soldiers, sailors and airmen on the Web site of the Library of Congress.

This included Social Security numbers for a former operational commander for Iraq, a former commander of U.S. Central Command, a former vice chief of Naval Operations, a former director of the National Security Agency -- and Colin Powell, the former chairman of the Joint Chiefs of Staff and Secretary of State for President George W. Bush.

The Social Security numbers were in online versions of the Congressional Record from the late 1980s and early 1990s available via a beta search tool on the library's Web site, known as THOMAS. The Social Security numbers were published with the soldiers, sailors and airmen's names when they were nominated for higher ranks.

Because the numbers were first printed long ago, most of the soldiers, sailors and airmen appear to be retired. Some are now prominent private citizens. One, for example, now sits on the South Carolina Supreme Court.

The Social Security numbers were there until I told the Library of Congress about them this week. I found them while using the beta search tool, which has since been taken down.

I didn't attempt to find every page with a Social Security number on it, but from the pages I did find, I extracted 31,144 unique Social Security numbers.

I shared three examples of the pages I found with the library so they could remove the Social Security numbers before I wrote about them.

John Sayers, a spokesman for the Library of Congress, replied in an email message Wednesday that "We've been aware of this problem with some parts of the Congressional Record on THOMAS for some time now, and have been making special efforts to find these problem pages and scrub them of any SSN material."

"Obviously, these were three pages we missed," he wrote.

He promised to tell the library's "data folks," who he said would "redouble their efforts and make sure we've got all these scrubbed."

Sayers also promised to explain further how these pages were missed. He wrote that "about four years ago we took a thorough look at all our online public data with an eye to finding and removing any Social Security numbers that might appear there ... A scrub was done at that time (zeroes put in place of the actual numbers)."

"As you may know, we have millions of searchable records and web pages online, in addition to the Congressional Record, where the numbers you cited originally appeared," he wrote. "It's a lot of information to check that isn't always clearly marked with a signpost that says 'SSN Numbers Here.'"

Sorry to say, it took a while to even get the library's attention.

I initially reported the Social Security numbers through the Library of Congress' beta comment form. I did it that way because I wanted to see how the library would respond if just anyone reported it, not necessarily a journalist. I used my name and email address, but didn't identify myself as being associated with a newspaper.

I never received a response.

A week later, I reported them again using the library's regular comment form.

Again, no response, other than a generic, automated email reply that my message had been received

Finally, I emailed the public affairs office using the email address published on the library's Web site, this time identifying myself as an editor at The Courier-Journal.

Again, no response.

I didn't get a reply until I called and left Sayers a voice mail message, two days after sending the email to public affairs. Sayers later wrote that email to the public affairs office goes to him but he never saw it, perhaps because of "a blip on the system" or because he accidentally deleted it.

"... my sincere apologies again for the fact that no one got back to you sooner on this," he wrote. "It is our policy, when notified by a user that there are SSNs displayed anywhere on our websites, to remove them. We should have responded to you immediately."

I never contacted any of the people whose Social Security numbers were exposed directly. I'm saying these are their Social Security numbers based on the uniqueness of their names, where the Social Security numbers were issued and how their careers paths compare to the promotions listed in the Congressional Record. I could be wrong in a specific case or two mentioned above, although I doubt it.

Personally, I think the issue of identity theft is overhyped. I'm writing about this because not everyone agrees.

Recently the Veterans Administration spent millions to notify veterans and agreed to pay $20 million in a court settlement after a laptop and external drive was stolen from an employee's home.

That strikes me as absurd. The onus should be on the retailers, banks, credit card companies and others who make use of Social Security numbers to have robust defenses against fraud. You shouldn't be able to lose your identity just because your Social Security number or other personal identifiers become public.

If nothing else, this shows how ubiquitous Social Security numbers are, and just how futile it may be to pass laws to hide them or to pretend that by banning their public display it will somehow protect us from identity theft.

Just this January Senator Dianne Feinstein of California introduced the Protecting the Privacy of Social Security Numbers Act, which would amend the federal criminal code to prohibit the "display, sale, or purchase of Social Security numbers without the affirmatively expressed consent of the individual."

In her floor speech introducing the legislation, which is similar to bills she introduced in past sessions, Feinstein said:

Use of these numbers has expanded well beyond their original purpose. Social Security numbers are now used for everything from credit checks to rental agreements to employment verifications, among other purposes. They can be found in privately held databases and on public records--including marriage licenses, professional certifications, and countless other public documents--many of which are available on the Internet.

Once accessed, the numbers act like keys--allowing thieves to open credit card and bank accounts and even begin applying for government benefits.

Feinstein then shared this anecdote:

One thief stole a retired Army captain's military identification card and used his Social Security number, listed on the card, to go on a 6-month, $260,000 shopping spree. By the time the Army captain realized what had happened, the thief had opened more than 60 fraudulent accounts.

If only she knew.

#depth on Twitter

2009-06-27T11:05:00.001-05:00

You can find links I used to share more often on this blog by searching for the #depth hashtag on Twitter.

Follow me on Twitter, Delicious and Google Reader

2009-06-20T11:03:00.001-05:00

If you’re interested in the kinds of things that interest me on this blog, you can follow me on Twitter, Delicious and Google Reader. I serve an unsteady stream of links that often never make it here. All have RSS feeds you can subscribe to independent of Depth Reporting. My latest updates to those services are also available on Depth Reporting’s right rail. There’s some cross-posting going on, so forgive me for the dupes.

Visualizing U.S. Supreme Court ideology

2009-06-18T09:28:00.001-05:00

The Ideological History of the Supreme Court, 1937-2007, is a nice visualization designed by Alex Lundry that uses scores developed by two political scientists to measure where the justices have stood on an “ideological continuum” from liberal to conservative. You can download the visualization in spreadsheet form, play with it and see how it was done. Boiling a judge’s judicial philosophy down to a score is a tricky business, and not everyone likes it. You can get a flavor of that by reading The Distorting Slant of Quantitative Studies of Judging, by a law professor.

[via InfoAesthetics]

“100 Awesome Open Source Tools for Writers, Journalists, and Bloggers”

2009-06-17T21:21:00.001-05:00

…. from OnlineCourses.org. Actually, some are just so-so (and at least one, Evernote, isn’t open source), but at least they’re free.

Temporarily laid off, again

2009-06-07T08:50:00.001-05:00

Tomorrow begins my second mandatory exile from the Gannett Company. I’ll return June 15.

Data.gov is lame (so far)

2009-05-22T08:03:00.001-05:00

Data.gov has come out of the gate lame. This is the Obama administration's new site for sharing executive branch data in formats friendly to programmers and stat junkies. It's got the Sunlight Foundation goo-goos excited: They're offering a $10,000 prize for the best application and visualization using Data.gov's data as a way of encouraging the government to do more of this sort of thing. Craig Newmark of Craigslist, who's putting up some of the prize money, called it "a genuinely big advance in grassroots democracy." I'd like to see more government data sharing too, but to say that "we've just taken an unprecedented first step into the Era of Big Open Government," as Jake Brewer of Sunlight wrote on the Huffington Post, is to believe that the promise of something is the same as delivering it. Rounding up a tired collection of not very interesting data sets already available on other government Web sites (The Residential Energy Consumption Survey and Patent Application Bibliographic Data, together at last!) and serving them up on a not unattractive but cramped search page is barely evolutionary, much less revolutionary. Marshall Kirkpatrick of ReadWriteWeb lamented that it was "a bit of a let down" and noted that the 47 data sets there now "lean toward the least controversial matters." He searched for keywords like food, prisons and drugs -- and came up empty. It’s the tough stuff, the stuff that challenges real vested interests, that matters, of course. It reminds me of the poor marks given to Kentucky's OpenDoor Web site, which has made a show of making state spending transparent, while failing thus far in the execution. If you tell me you've spent so much for widgets, but not who's supplying the widgets and at what terms because, golly gee, doing so would piss the widget supplier off, you haven't really told me anything useful, much less praiseworthy. Yes, intentions are good, and Data.gov is promising to listen to suggestions and get better, but I'll save my huzzahs for meaningful results.

Use Twitter to publish breaking news to your Web site

2009-05-10T20:08:00.001-05:00

Matthew Stoff and David Durrett of The Daily Sentinel, a newspaper in East Texas, explain how to use Twitter to publish breaking news from the field to any Web site. "If you can send an e-mail, you can post breaking news," they say:

With this method, a writer or editor can use a Twitter account to publish breaking news directly to a Web site without any technical skills required. After a one-time setup, headlines or breaking news updates sent with this method will appear seamlessly on the Web site, matching the appearance of existing styles on the page. The headline can be easily replaced or removed without any technial knowledge. And the publishing occurs entirely via e-mail so that reporters with e-mail capable cell phones can send updates directly to the Web site from the field using this technique.

Bankruptcy Research Database

2009-05-10T19:59:00.001-05:00

UCLA law professor Lynn M. LoPucki's free Bankruptcy Research Database "provides data for much, if not most, empirical work" on the subject of large company bankruptcies, according to his faculty bio. Information on bankrupt firms includes debtor names, annual sales, assets, number of employees, the filing city, the disposition of cases and more.

Data quality in the WebBRD exceeds that available in any other business bankruptcy resource. Data are gathered from a variety of sources, the most important of which are the bankruptcy courts' files (on PACER) and the debtors' filings with the Securities Exchange Commission.

Data are carefully collected and checked according to Bankruptcy Research Database protocols. Since the founding of the BRD in 1994, I have personally checked each piece of data against its source at the time of entry.

On his frequently asked questions page he explains the database's purpose:

The database itself is an experiment in systems research. My theory is that improving the flow of information in any social system has a marked, positive effect on the operation of the system. If my theory is correct, wide availability of the BRD will change the manner in which the bankruptcy courts and professionals process the cases of large, public companies -- and the change will be an improvement. ... Also, I love to collect and explore good data.

Discogs

2009-05-10T19:49:00.001-05:00

Discogs is "a community-built database of music information." You can download all the site's data or access it programmatically.

Is it a good thing that photojournalism traffics on the dark side?

2009-04-29T22:41:00.001-05:00

Mindy McAdams viewed this fine tale of addict woe and wondered if it's a good thing that photojournalists prefer images of illness, death and depravity:

I fully acknowledge the value in taking us where we do not want to go. We need to see scenes of poverty, death on the battlefield, victims of natural disasters. This is the duty of journalism, to take us there.

But I’m having a problem with stories about sick people, people with cancer or leukemia, and addicted people (also a sickness) that just take us down a hole and leave us there. I also think that far too many stories by student photojournalists are focused on these subjects. I start to think, “Oh, no, not another sweet little child with cancer …” when the slideshow starts.

Shouldn’t we be looking for other stories? What does anyone in the audience learn from seeing yet another child with a disease, and the grieving parents, the valient fight? Effort be damned — what is the value to the public in these tragic tales? Is this what journalism is for?

Typically we don't learn anything. It's an exercise in cheap emotion and voyeurism. And yes, that's one of the things journalism is for.

Freerisk

2009-04-28T21:15:00.001-05:00

"Financial hackers" are creating a free, open store of financial data to make it easier to spot fraud or companies at risk of default. They've also created a Google Group for people "interested in hacking the financial system for the greater good."

Interactive public data from Google

2009-04-28T21:11:00.001-05:00

You can now type "unemployment rate Kentucky" or "population indiana" in the Google search box and get an interactive chart that allows you to do comparisons with other states. Says Google:

... we have been working on creating a new service that make lots of data instantly available for intuitive, visual exploration. Today's launch is a first step in that direction. We hope people will find this search feature helpful, whether it's used in the classroom, the boardroom or around the kitchen table. We also hope that this will pave the way for public data to take a more central role in informed public conversations.

TechCrunch suggests it was deliberately intended to steal the thunder of a soon-to-be-launched search engine from the maker of Mathematica.

It also can't be good news for Many Eyes, Swivel or Data360.

Spreadtweet

2009-04-28T21:00:00.001-05:00

If your boss scolds you for Tweeting at work, consider Spreadtweet:

So, you work at a big corporate, huh?
And you're not allowed to use Twitter...
Wouldn't it be awesome if there were a Twitter tool
that looked just like Excel?

A newspaper is a composition, like music

2009-04-10T11:37:00.001-05:00

The front page of today's Courier-Journal was hand-drawn by a Turkish artist, an experiment in public art I love. If only we were as inventive every day.

It reminded me of this TED talk by Jacek Utko, a Polish designer who said there's “no practical reason for newspapers to survive,” then went on to describe how superb design helped lift the circulation of a chain of European newspapers. “We were treating the whole newspaper as one piece, as one composition,” he said, “like music.”

You can live in a small, poor country, like me. You can work for a small company, in a boring branch, you can have no budgets, no people, but still can put your work to the highest possible level. And everybody can do it. You just need inspiration, vision and determination. And you need to remember that to be good is not enough.

LLRX.com: “Competitive Intelligence - A Selective Resource Guide”

2009-04-08T09:15:00.000-05:00

LLRX.com recently updated its “competitive intelligence” resource guide.

Glassdoor.com

2009-04-07T09:29:00.001-05:00

… offers “Salaries, Ratings, and Reviews posted anonymously by employees” about the companies they work for.

Glassdoor is completely free — there are no subscription or report fees. Instead we use a simple "give-to-get" model.

Give us your review or salary for a current or former employer.

Get access to all our reviews or salaries.

That seems fair, right? If you're interested in seeing reviews, give us a review of your own and you'll get access to all the reviews posted by our employee community for a full year — all for free! The same goes for salaries. To see what we're talking about, check out employee reviews or salaries for one of our Sneak Peek companies or get started by searching for your company or job title. Chances are we have you covered.

Forbes.com calls it “Workplace Porn.”

Free Mathematics Books

2009-04-07T08:51:00.000-05:00

“ … an alphabetical list of online mathematics books, textbooks, monographs, lecture notes, and other mathematics related documents freely available on the web.”

Little Red Riding Hood, reinterpreted

2009-04-03T17:46:00.000-05:00

Slagsmålsklubben - Sponsored by destiny from Tomas Nilsson on Vimeo.

Launching a nonprofit news site “no picnic,” MinnPost CEO says

2009-03-24T10:31:00.001-05:00

Joel Kramer is a former Minneapolis Star Tribune editor and publisher who is the founder of MinnPost.com, a Web news startup that promises “a thoughtful approach to news.” His recounting of lessons learned after the first year is refreshingly specific:

Even for our serious audience, we’ve learned that $600 spent on one long story produces a lot less traffic than $600 spent generating six to 12 shorter items. We still do longer stories every day, including many that combine in-depth reporting and analysis with personal voice.

But a careful reader of our site over the past year will note that we have a great many more short, quick hits, published all day long. So while we are spending less on news today than a year ago, our traffic has more than doubled during that time.

As traditional media organizations shed workers, MinnPost.com and similar sites have fielded so many calls seeking advice on how to do something similar Kramer says they’re considering forming "a consortium of nonprofit regional online news sites" to help others get started.

Nonprofit, by the way, doesn’t mean no money down:

Many of the callers tell me they have no start-up funds in hand yet. “Well,” I say, “I’d start by getting some.”

Jack Shafer: “The pharm-party meme refuses to die”

2009-03-24T09:23:00.001-05:00

A lack of evidence never stops a journalist from declaring a trend. Slate’s Jack Shafer makes the point about alleged pharm parties, a “drug bacchanalia in which teenagers meet up to dump the pills they've pilfered from their parents' medicine cabinets into a collective bowl”:

In recent months the Petoskey News-Review, the Las Vegas Sun, the Herald Bulletin (Anderson, Ind.), the Rapid City Journal (S.D.), the Oklahoman, the Salisbury Post (N.C.), the Long Island Press, the Caspar Journal, the Times & Transcript (New Brunswick, Canada), the Columbian (Vancouver, Wash.), the Reading Eagle (which also calls the drug mix a "fruit salad"), the Paducah Sun, the Salt Lake Tribune, the Charlotte Observer, and other newspapers have given publicity to pharm parties without visiting one.

The difference between a million and a billion

2009-03-20T09:33:00.000-05:00

Two tools for checking on the health of banks

2009-03-19T10:05:00.001-05:00

BankTracker, from the Investigative Reporting Workshop at American University, has received more than 700,000 pageviews since its unveiling this week. BankTracker uses FDIC data to calculate the troubled asset ratios for all of the nation's banks, comparing them to the national median. An explanation of their methodology says they don't attempt to "value the non-loan assets that may also be causing bank problems, such as mortgage-backed securities, collateralized debt obligations, etc." Needless to say, such assets are not an insignificant factor in the current financial crisis.

Bankrate.com, which calls itself "the Web's leading aggregator of financial rate information," offers an older, more in-depth tool that provides "Safe & Sound ratings" for banks, thrifts and credit unions. Bankrate says it "continually surveys approximately 4,800 financial institutions in all 50 states in order to provide clear, objective, and unbiased rates to consumers." It says their system applies 22 tests to measure each institutions capital adequacy, asset quality, profitability and liquidity. "Individual performance levels are determined from publicly available regulatory filings and are compared to asset-size peer norms, industry standards and key absolute benchmarks," the site says.

[via IREPLUS-L]

Furloughed

2009-03-08T18:38:00.001-05:00

My one-week furlough from The Courier-Journal begins tomorrow.

Furlough, of course, is a polite if Orwellian way to say I've been laid off for a week.

Amusingly, it has other meanings. Wordnet says one is to "grant a leave to," as in "The prisoner was furloughed for the weekend to visit her children."

Another is "a temporary leave of absence from military duty."

Prison. Military duty. This is what journalism has come to.

It is not, however, how I envision my future, so I'm going to use the week to learn some new skills that will make me more relevant in a depression economy -- say, street corner apple selling, breadline waiting, odd jobbing and possession hocking.

Or Flex. I don't really know. I may just scrape paint off the old carriage house out back. I'm going to let my mood be my muse.

See you on the other side.