NICAR 2016 Lightning Talks
Submit a talk by Feb. 21st. Voting begins Feb. 22nd (10 a.m. eastern) and closes Feb. 28th (11:59 p.m. eastern).
Submit a talk by Feb. 21st. Voting begins Feb. 22nd (10 a.m. eastern) and closes Feb. 28th (11:59 p.m. eastern).
2016 Lineup:
Thanks for pitching and voting, and congratulations to our 2016 speakers!
For more than a decade news apps and graphics developers loved using the word "interactive" to describe all kinds of new things they put on the internet. But what does "interactive" even mean? Slideshows, tooltips and d3.js? Have we reached the peak of the golden age of "interactives", or has it just started? Let's take five minutes to look at where we are and where we're going...
Admit it: you've copy and pasted a column of state names next to one of USPS codes... hundreds of times. You keep writing identical code to add the same state population numbers to different spreadsheets. Your nightmares echo with FIPS VLOOKUPs and CPI JOIN clauses. In this lightning talk I will introduce a tool and a project. The tool works. The project needs you. Together, we can forever vanquish the drudgery of the lookup table. Fans of csvkit or agate: please allow me to introduce you to `lookup`.
So you're not cut out to be one of those highfalutin programming databasers who can brag about their automated cloud bots bragging about their other automated cloud bots. But even if you're just a regular person with regular data needs, you can still learn and practice regular expressions to find and filter the kinds of things you already know you want to know from a pile of documents -- dates, names, money amounts -- and anything else that's a combination of the regular old alphabet and numbers 1 through 10.
In high school, math felt like a foreign language to me -- difficult to understand, stressful to learn and, above all, something to avoid. Now, not only is math a big part of my job, but it's enjoyable as well. I'll share several easy-to-adopt tips, tricks and habits, all of which increased my math fluency and will improve your understanding of any equations or formulas that you come across while you're reporting.
How can you map every census tract in the US -- thousands of shapes -- easily, without rendering map tiles or melting a computer? How can you style it dynamically? Find out how vector tiles can change your life. Or at least your maps.
With so much content to consume - Twitter conversations, listserv kerfuffles, the latest longform, Snapchat stories, hot takes and think pieces, the entire season of TV that just dropped on Netflix - where do BOOKS fit in? I'll talk about (1) why you should set the daunting goal of reading 52 books in 52 weeks rather than just resolving to "read more," (2) how you can achieve it, and (3) what things you'll learn (about the world and yourself).
Fresh off the European tour, it's cats AND stats. Stanley will demonstrate dichotomous variables. You'll learn about sampling from Gordon. Harley will give you tips for examining your data. It's not just cats. It's cats AND stats. Learn through the hilarious antics of the cats of NICARians.
When it is worth it to use scripting and automation to produce our news stories? Should your newsroom bother with writing these scripts? I’ll talk about what kind of automation has worked for stories in the past, where we can move in the future, and the kind of stories that will never be worth it to automate.
Building a case with analysis and visualization.
DON'T OPEN THAT TEXTBOOK. Statistics is a cycle of abuse and I'm here to make it stop. In five minutes I'll show you the refreshing new approach that everyone's talking about: randomization and visualization. Learn what types of problems statistics really solves and how to solve them with just a few lines of common sensel code. Get the right answer faster than you can say "Monte Carlo" and never memorize a formula again!
Google's AMP project will make the mobile web faster. It will also strike a death blow to innovation in digital journalism. Learn why it's so dangerous – and what we can do to prevent the AMPocalypse.
We tracked helicopters at the World Economic Forum in Davos Switzerland (http://qz.com/600590/) using a Raspberry Pi, a $20 USB stick, and some wire most typically found connected to your cable box. Let me show you how we built it and how we were not only able to find helicopters in the haystack of the 2 million data points it collected, but also who owned them.
No latlng, no address, no shape file? Have no fear, that doesn’t deny our chance making a map. I'll show you how I turned pages of scanned notes by the police into an interactive map in three days. Learn what kind of help you'll need and how to draw a map from scratch.
We spend a lot of time cleaning data in our programming language of choice. But is it the right one? Using Excel and R for two simple tasks, we (Sandhya Kambhampati and Christine Zhang) will show you the pros and cons of each. You be the judge.
Bret Victor, in his essay "Magic Ink," argues that interactivity in software interfaces is a "curse." He argues that interaction, for the user of information software, is a liability that should be minimized, extending Tufte's ideas about data visualization to software design. What can we as designers and developers of "interactives," "news apps" and software journalism tools learn from Victor's ideas about information and interface design?
There's a big gap between "student journalist" and "hardened newsroom coder," and existing solutions for bridging that gap (such as workshops and internships) don't seem to have eased the difficulty that many newsrooms have with finding talent. Certainly academia does not appear to be adequately addressing the need for skills. Can we create a solution for training outside of those systems, to create more experienced (and more diverse) journalists? This talk covers my experiences after posting a call for remote mentees to the NICAR-L list, and looks at other approaches taken from open-source software and other industries that might help.
You don’t have to be a computer scientist - or even great at coding - to use machine learning in your data stories. I’ll talk about some of the easiest ways to make stories more robust with simple machine learning methods, along with some examples.
I was in Paris on Nov. 13 when ISIS attacked the city. In this talk, I'll share some of my experiences during the attacks –– how I found out, how I stayed updated, how I figured out what could and could not be trusted. And in the days following: what did I learn about journalism –– as I went from a person covering the news, to someone living through it?
How can stories with lots of numbers, looonng explanations and dozens of interviews become compelling audio stories? How can I translate a one-year investigation into a four–minute piece? And more importantly, how can the print peeps who used to do that work at newspapers make the transition to radio and podcasting? Peggy Lowe, investigations editor at Harvest Public Media and KCUR, gives her top tips on moving from ink to air.
Did you now that eating egg rolls was linked to dog ownership? And potato chips to a higher SAT maths core? Well...not really. These relationships *are* statistically significant; but they were found by "p-hacking" data from nutrition studies. In this lighting talk I'll show you how to do regression analysis in R the evil way, by reworking a FiveThirtyEight story. By learning the "dark side" of statistics, hopefully you'll make smarter choices in running regressions.See (1) here: http://fivethirtyeight.com/features/you-cant-trust-what-you-read-about-nutrition/ and (2) here: https://github.com/fivethirtyeight/data/tree/master/nutrition-studies.
So many platforms, so little time. Do I use GIFs or create something in After Effects? Just take a screenshot of that chart or use the latest social media meme-making tool? The dirty secret about telling data stories on social is that you need to treat each platform in its own unique way and tailor the story accordingly. In this talk, I will present examples of data stories told in innovative ways on different social media platforms. By the end, you will walk away with ideas and simple tricks to bring back to your newsroom.
Have you ever waited for a script or long-running database query? Learn how I made my code run eight times faster by changing two lines, and you can too. Not working in Python? Also learn about GNU Parallel, which can speed up any repetitive command-line task.
Like many reporters, I used to spend a lot of time on my blog and Twitter criticizing government agencies for refusing to release data or taking too long to do so. Then, a couple years ago, I tried something different: Praising the few agencies that actually cooperated with my data requests. A case study in why carrots are as important as sticks in getting access to data.
Leaving business decisions to "the business people" is no longer a luxury traditional and computer-assisted journalists can afford. I'll talk about how hybrid journalists at NPR One use audience data to change NPR's 46-year-old audio storytelling strategy in a way that has the potential to take NPR's business model to the modern age - without resorting to cat videos.
Data journalism isn't always as objective as we'd like to think it is. We all know the importance of evaluating data sources for credibility. Let's turn that mirror inward and reflect on how our own individual biases affect the way that we select and present data.
Lindsey Cook spent a year talking to journalism students about coding. Despite an industry that is apparently clamoring for journalists with more tech skills, students (and the professors teaching them) aren't getting the message. In the talk, she'll detail the often depressing reasons why, as well as what working journalists can do to change them.
Programming and data analysis are already difficult enough without you diverting your brain energy to memorizing variable/function names or even just keeping focused as you switch between multiple applications. Learning a few keyboard shortcuts -- which can be used whether you're working or playing -- can free your brain for the more thoughtful part of computational work, and maybe even reduce risk of carpal tunnel syndrome from mouse/trackpad overuse.
Social media and partner platforms are increasingly important for news and media organizations, but building up a presence is so much more than just calling the youngest person into the room and asking them to explain Snapchat. We’ll discuss the do’s and don’ts for the most popular social media platforms, ranging from Mic's great Tumblr presence to the success of Vox’s Snapchat Discover, with tips and tricks from (gasp) youths.
Disney movies may seem like child’s play at first glance, but we can learn from a lot them about visual storytelling, choices about media formats, and accessibility. Here’s 7 things I learnt working my way through almost all 54 original Disney movies in the last 3 years.
What to do when manual data entry is the only option - and how code can still help
Ok, not really. But it is true that American tech companies are eating/have eaten the revenue base away from under the shaky structures of traditional African news media. Which means the death of democracy. Which means you’re eating your own children. Ick. So I’ll tell you how we’re trying to fix that. Or at least, how we’re trying to go down fighting. In five slides. Maybe four.
Dozens of non U.S. reporters attended NICAR last year and brought a lot of precious know-hows back home.Some techniques are ready-to-use (yes, I used them), but some are not applicable in their country due to custom, language, geometry, law, social taboo, FOIA(lack of it), state of open gov...I (we, hopefully) will show you several things that U.S. NICARians don't have to worry about but that foreign NICARians are tackling (there might be inverse cases), which might inspire something in U.S. data journalists.I want this to be a kind of collective presentation and needs collaborative inputs from international journalists/foreign correspondents if accepted.
"Mass Incarceration" has been a big news story for a long time, but unfortunately no one (myself included) really understands it. The problems with understanding it include incomplete or unreliable data, the temptation to generalize from biased samples to larger populations, and the difficulty of conceptualizing at the individual level--all problems that show up in other topics involving the numbers of millions of people over time. I'll explain as much of it as I can in five minutes.
Journalists are great contributors to the open-source mapping community. And the data in OpenStreetMap can be leveraged for our gain. You can also spin up a "map maker" and get started building your own maps for the internet, social media and that queer old thing called a news-paper. Jon Schleuss shows how the Los Angeles Times uses (and contributes to) OpenStreetMap.
From Asia to Africa, Latin America to the Middle East, many of our colleagues face Internet blackouts, imprisonment, threats and even death as they work to reveal what is happening in their countries. I’ll talk about why we should care and concrete steps we can take to support and work with them.
GitHub is much more than Octocat, pushing, pulling, merging and branching. I’ll talk about ways we’ve used GitHub at North by Northwestern, an online newsmagazine at Northwestern University, to build tools, teach web development, write better documentation and more!
What does this have to do with NICAR? Going beyond an understanding of how an ad support organization is funded, we need to take more ownership over the entire experience of our readers. A talk about what is happening with ads today, and more importantly what developers and others can do to have the tools and information needed to help their publication and the entire ecosystem improve the overall experience on our sites.
Everyone new to the world of journalism and data make mistakes. I know because I am an intern, fresh in the world of journalism, and a certifiable expert in making mistakes. I will give an overview of some common mistakes that I have made and how we can all avoid them in the future.
We build interactive stories and databases but often they live in the constraints of our own sites. What if we built more widgets that could live anywhere on the Internet? Widgets - the sole purpose of which was to inform and tell the story first and bring back people to our site later. And if we did build more of these, what is our audience for it? We will look at some existing examples and I'll show you what widgets I think could have been built with some major stories.
Facebook is opening Instant Articles to all publishers in April at F8. Learn what you need to do to be ready to launch Instant Articles for your publication.
The data visualisation field is full of very diverse tools – you can't possibly learn them all. (Although it seems everybody else does.) But with which tools should you start and with which ones continue? Which ones promise the most functionality with the flattest learning curve? And should you focus on learning one powerful piece of software or lots of smaller tools? I will offer a way to think about these questions, to finally end your "OMG SO MANY! FOMO!" problem.
We are going to surf some very useful extensions to play with your data like a champ in Jupyter notebooks. How to present you data, visualize it and share your results directly from your notebook.
Have you ever talked to a reporter from Europe? Did you ever cooperate with one? You definitely should. Because if you don't, that's not only a shame, you also miss a lot of sources for data and documents – lots of them regarding the US. And you also miss a lot of great stories as well. I'll tell you about the most important stories you can get by cooperating with your European colleagues – all stories with an American angle. And I will give you a list with special European resources for data and documents. Get your European-NICAR-experience in five minutes.
My non-judgmental spiel on why I parent like a hippy even though I work in tech and journalism. Bonus, free hugs.
With the evolution of the Smartphone and the rise of new technologies, such as Google Cardboard, we face the possibility of building amazing new virtual worlds in order to reimagine the stories we want to tell to our readers. However, it seems that the generation of this kind of resource is limited to a selected group of video game and graphics programmers. This Lightning talk breaks this myth and shows, with the help of simple tools, that if you can fill a simple spreadsheet, then you are capable of creating incredible Virtual Reality experiences.
DESCRIPTIONAn insight into how to quickly and effectively go about making (video) tools for journalist in the newsroom.quickQuote, takes the pain out of finding a video quote for a news article.We will consider how it was realised in under 3 months. Drawing components from existing open-source projects, Using a mixture of lean, agile approach leveraging HTML5, javascript, JQuery and rails.http://www.niemanlab.org/2015/09/a-new-tool-from-the-times-of-london-lets-you-easily-detect-and-capture-quotes-from-a-video/
Last year, the most horrific fires in human history broke out in Indonesia, the consequences of years of deforestation and negligence. What began shortly thereafter, and continues to this day, is a battle of accusation, with hard to access, often conflicting data - maps, concessions, GHG emissions, and health statistics - at the center. I'll explain how I, as a journalist, told stories about this muddled field and reflect on how to use data in places where the data is, often, the story itself.