[danso]

I currently teach journalism at Stanford, with a focus on computational methods and programming. It's a good school and I like working with students, and I can work on research and my own learning. But I'm not used to the pace and lack of deadlines and I don't like the Bay Area compared to New York. My favorite job at this point was working as a developer at ProPublica, for the kind of projects we did and for being in New York.

[chubot]

Hm that's interesting -- what kind of programming are you teaching? Is it for journalists per se or the people who work with them?

I would imagine there is some overlap between journalism and "data science". Many articles about the world would benefit from statistics derived from online data sets to back them up. But I wonder if this is really journalism, or if it's the primary source which the journalist relies on?

There also seems to be a lot of interest in journalism and computer security/privacy, because journalists are targets for spying. But that doesn't sound like what you're talking about.

[danso]

I would define the kind of programming I focus on to be general scripting...the coding to do a variety of information gathering (e.g. scraping) and publishing (e.g. visualization)...I was mostly a Rubyist but decided to go with Python for teaching because it's a lot more consistent.

The other focus I have is on SQL, which I make all of the students learn in the required course I teach. For many students, it's the first programming they've ever done...and it's directly applicable in data journalism work. Here's a couple of sample midterms:

- http://www.padjo.org/2014-10-23/

- http://2015.padjo.org/assignments/midterm-wsj-medicare-walkt...

[mylons]

sounds like you might be the compjour.org instructor? if that's the case, big fan! followed along on my own and it got me interested in journalism.

how would someone who's been doing distributed systems and genetics for 8 years make the switch to something equivalent of five thirty eight, or other data driven outfits?

[danso]

I think the quickest way to make the transition is to show skill in visualization. Data journalists share the same kind of analytic skills and processes as other kinds of data practitioners, but with an emphasis on publication and design. Visualization also happens to be the easiest way to reach and impress a large audience.

The other part of it is to demonstrate a good sense of the state of civic affairs. You can actually do quite well with average talent in programming and statistical skill...the bottleneck when it comes to data journalism is knowing the quality and quantity of data that is actually kept and produced by civic entities. Basically, being a good researcher will pay immense dividends...all of my best projects have literally been just being able to find or being aware of information that has already been publicly made available (and the legal/institutional reasons for the precedent).

IQuantNY is one of my favorite examples of non-traditional journalists doing great things in civic data analysis using already publicly available datasets: http://iquantny.tumblr.com/