Tags : Tech Tips

Matching for dead registered voters

My news organization, Texas Watchdog, found thousands of dead people on our local county's voter rolls and dozens whose names may have been used to cast ballots after their deaths. We couldn't have conducted the investigation without the IRE and NICAR Boot Camp training in Columbia, Mo. Our story prompted local officials to remove two voters from the rolls whose names were used to vote after their deaths, and the story generated interest on the Drudge Report and Instapundit Web sites. Officials in Harris County, which encompasses Houston, have promised to meet with us post-election and have pledged ...

Read more ...

Overcome IT roadblocks with Amazon data services

It has become a common refrain among journalists who want to develop data-rich Web applications: What good is learning programming skills if our newsroom IT sentries won’t give us a server to work on? Fortunately for aspiring developers, a set of tools known as Amazon Web Services – offered by book-selling titan Amazon.com – is slowly gaining traction in newsrooms nationwide. It offers CAR specialists and Web developers a low-cost detour around IT-imposed roadblocks while providing robust infrastructure that won’t buckle under heavy traffic. Two services in particular are worth your attention: an elastic computing system known as the ...

Read more ...

Flying with SQL subqueries

In most cases, when you want to match data from one table in a database to another, a common approach is to use a JOIN clause in your query. For example, if you had a list of candidates in one table, and a list of donations made to candidates (with one field having a candidate ID in it), you'd probably either use JOIN clause or a WHERE clause to match them together: (pseudo-code) SELECT * FROM candidates, donations where candidates.candidate_id = donations.candidate_id; But simple joins aren't the only technique available to journalists trying to learn more about their ... Read more ...

Data scrubbing ID's college football players with criminal records

Catching a 20-year-old college football player drinking might be as easy as stumbling into any busy bar near campus. Trying to find every player who has run afoul with the law over eight years takes more work. Producer David Lubbers and reporter Steve Delsohn were already working on a story about Penn State players when I joined ESPN in May. (Watch the report on ESPN's "Outside the Lines.") They asked if I could help them answer a seemingly simple question: How many Penn State football players have been in trouble with the law since 2000? A variety of lists ...

Read more ...

Death and the UNION clause

Sometimes we need to get information from a database where the information is, for whatever reason, not stored in one table but broken into smaller tables with an identical structure. A good example of this is our Social Security Administration Death Master file. In essence, this is a simple database, with only a few fields (such as the person's name, when they were born, when they died, etc) -- but because it's an effort to store this information for every US citizen who died since 1937 (by no means a perfectly successful effort, but I digress), what could be ... Read more ...

Scraping barriers and how to avoid them

In early January, a short thread on NICARL, a computer-assisted reporting Listserv operated by IRE and NICAR, highlighted a bizarre step taken by the Seattle Fire Department to shield its response time data from mashup artists. Intending to obstruct automated Web scrapers, Seattle fire officials reformatted the public response data available on their Web site from HTML into a JPEG image. The theory, according to the officials behind the decision, was that malevolent users could harvest the data and exploit it for nefarious ends. "Our intent is to enhance the safety of personnel and the public but still provide information ...

Read more ...