Recent Posts

Why does YouTube have a longer lifespan than other platforms?

5 minute read


When trying to reach a mass audience, what's the best platform to share your content? Well, the obvious answer is as many places as you can. But according to a post by bitly analyzing traffic patterns, links shared on YouTube have a lifespan of 7.3 hours, compared to 2.8 hours on Twitter and 3.4 hours on Facebook. Why such the disparity? Why does YouTube have such a longer lifespan?

Is it because video has a longer lifespan than all other forms of content? Or is it because YouTube has a different user-experience than other social media platforms? While YouTube content is slower to peak, it lasts far longer in the online ecosystem than content posted on other social media platforms such as Twitter and Facebook. The most obvious answer for the cause of this phenomenon would be that video is a medium that inherently captures our attention for a longer, and slower, period of time. We tend to go back, rewatch and share video more than we do text-based content, causing video to have a longer lifespan.

But there's also another possible explanation for YouTube's lengthier half-life. It could just be the nature of YouTube's network structure. Facebook and Twitter are more of aggregators than YouTube, which is a platform for user-generated content rather than just a portal. So, because of their vast user base and high rate of captivity, Facebook and Twitter by their nature attract attention quicker. But that attention is often only surface attention, which is possibly a reason those networks have a shorter half-life than YouTube. People go to YouTube videos more frequently as a destination, whereas other social media platforms only act as a portal.

Critique: “Salubrious Nation: a game-y look at U.S. health”

10 minute read

In keeping with our recent weekly reading about the growing 'gamification' of data, I wanted to focus my critique this week on a map-styled data-driven game made my a group of researchers at Rutgers University called Salubrious Nation. The game attempts to engage users more deeply with public health data by luring them in with an addictive system of points and rewards.

In terms of functionality, the game play operates fairly simply. A map presents demographic data about every county in the 48 states of the continental U.S. The game then chooses one county at random and asks the player to guess a public health statistic about it, like binge drinking, teenage pregnancies, diabetes, obesity rates, etc. The game features two types of interaction: the user can mouse over any county to see demographic data about it (population, poverty rate, life expectancy, etc.), and a slider at the top to enter the player’s guess for the county up for play. As you moved the slider up and down, you can get hints about how close you are by looking at whether the surrounding counties are above or below the value you've chosen. Based upon how close the player's guess is to the actual statistic, the player earns a corresponding amount of points. After eight rounds of the play, the game ends, and the player is told how his or her performance matches up to others who've played the game before. Apparently, I scored higher than 62 percent of other players. Woohoo! Just enough of a dopamine rush to get me to play again.

What's cool about this game is that it makes data something to get immersed in for the fun of it, and you learn along the way. Over time, you begin to notice patterns emerging as you learn the tricks and strategies of the game. You figure out that the Western half of the country tends to have a higher rate of binge drinking. You learn that diabetes and obesity is the worse in the South. As one of the game's creator, Nick Diakopoulous, explains, the gamification of health data provides a good opporunity for users to focus on data they might otherwise ignore: "Considering the selective attention issue, where people are more likely to pay attention to things that they already agree with, this result suggests an opportunity to get players to look at aspects of the data that they might not otherwise be inclined to look at."

I can only find a few possible qualms with the game. One is that it operates off of flash, meaning that it can't be run on most smartphones or tablets. Another is that the yellow-to-orange color scheme seems to be a bit disorienting on the eyes. Perhaps the developers would've been wiser to choose softer colors – possibly even a red-to-green graduated scale with a neutral middle value. Another thing that irked me, although I see little simple solution, is that county-level guessing seems almost so geographically-specific that it's hard for most people (including myself) to have much knowledge of which specific counties in Oklahoma or Kansas have the highest obesity rates.

Why news organizations should stop differeniating blogs from articles

13 minute read

Andy Boyle (@andymboyle) of The Boston Globe made an┬áimpassioned┬áplea to news organizations earlier this week that they stop differentiating between blogs and articles because they’re both equally forms of content. Someone’s been needing to put this into writing for a while now, and I’m glad Andy said it so eloquently.

INTERACTIVE: Why is the South the most obese part of the country? Five theories

17 minute read

This map displays the obesity rate of each U.S. state in 2010. The darker shade red represents a higher percentage of obese residents, while the green represents states with lower obesity rates. Click on each state to see the exact totals of each state's obesity rate.* 

Southerners need to lay off of the Crisco, cut back on the processed foods and start spending more time on the treadmill to fight the growing epidemic of obesity, experts say.

According to 2010 data released by the Centers for Disease Control and Prevention, the South is the most obese region in the nation, with about one in three of its residents classifying as chronically obese. That's far greater than the entire nation, where the figure is closer to one in four.

Of the 10 states with the highest rates of adult obesity, eight of them are in the South: Mississippi, Alabama, South Carolina, Kentucky, Louisiana, Texas and Tennessee. And that's only assuming you don't count West Virginia as being "southern."

Across the nation, the epidemic has grown worse in recent years. Twelve states now have obesity rates higher than 30 percent, compared to four years ago when only one state, Mississippi, ranked above the 30 percent threshold. The only state in the Deep South without an obesity rate of more than 30 percent today is Georgia, but that appears to be primarily because the more physically fit population of metro Atlanta offsets the rest of the state's obesity.

But what's making the South –– the region CDC Dr. William Dietz has dubbed "the heart disease and stroke belt" –– more chubby than the rest of the nation? Here are five possible explanations:

1. High poverty -– The South may be obese, but it's also poor. With a poverty rate of 14 percent, the South is easily the most impoverished region in the country. And according to data from the USDA, states with a higher poverty rate also tend to have a higher number of obese citizens. Experts say that's because people with a low income are more likely to purchase high-calorie inexpensive processed foods, which contribute to weight gain. "If you overlay a map of obesity onto a map of poverty, the two very clearly correspond," said David A. Davis, a professor of Southern Studies at Mercer University who has conducted extensive research on southern foodways. "The southern diet is a diet of poverty, and it's one based on cheap, fatty processed foods."

2. The "grocery gap" – Because the South is largely rural, many residents don't have quality access to fresh fruits and vegetables, and are forced to drive long distances to find anything healthier than potato chips and sodas at roadside gas stations. All five states with above average of what the USDA calls "food insecurity" levels are located in the South: Texas, Arkansas, Mississippi, Georgia and North Carolina. What's more, it's significantly more expensive to purchase low-fat items in the South than in the rest of the nation. For example, Georgia, Mississippi, North Carolina and Virginia all topped the list of states where it costs the most to buy low-fat milk, USDA data says.

3. The grease-fed "southern" culinary tradition -– One of the easiest explanations for the South's staggering obesity rates is the region's tradition of fried chicken, sweet tea and gravy on top of everything – or what's commonly referred to by non-southerners as the "Paula Deen" effect. "To me, it's simply a cultural habit regarding what we eat, not an issue of poverty," says Andy Breck, director of the Center for a Better South, a nonprofit group based out of the University of South Carolina that seeks to raise awareness about ongoing issues facing the region. "People are fat in Mississippi. People are fat in South Carolina. People are fat in Alabama. There's got to be something going on. And it's not just poor people. It's middle and upper-class folks who grotesquely overeat, because that's all they've ever known to do."

4. Lack of physical activity–– Southerners also tend to be less physically active than the rest of the country, burning off fewer calories and retaining more body fat, USDA data says. All five U.S. states where less than 60 percent of adults met the USDA's recommended physical activity guidelines in 2008 were located below the Mason Dixon Line: Louisiana, Mississippi, Alabama, Tennessee and Kentucky. Some researchers have speculated that the South's lack of physical activity may not be so much sheer laziness as it is a lack of access to places to exercise. Few rural areas have fancy private gyms for southerners to burn off their extra calories, and most of the year it's just too plain hot in the South to exercise outdoors.

5. Lack of quality education – Perhaps at the heart of the southern obesity epidemic, however, is the region's crippling lack of quality public education. "I don't buy the fact that the South is fat because of traditional southern foodways," said Davis, who teaches classes on southern poverty and culture and has written numerous articles on the subject. "To me, it's more of a poverty and an educational problem. If we don't educate people, especially in terms of health education, we're going to keep having obese citizens."

*Source: The Centers for Disease Control and Prevention, April 3, 2012


8 minute read

Lately I've been trying to get my feet wet with Django, an open-source Python web framework that's well-suited to producing complex news apps under  tight deadlines. I haven't had enough free time yet to get into the nitty gritty of it, but I'm getting there slowly. What first piqued my interest in Django was a brilliant news app I ran across a couple of months ago called Curbwise, which was built with Django by the news developer team at the Omaha World Herald/

Curbwise advertises itself as "your one-stop shop for the latest on real-estate in Douglas County." But Curbwise is much more than your standard, run-of-the-mill real-estate section of most local news websites. It allows the user to fine-tune what neighborhoods he or she wants to view, and compare demographic data and housing prices side-by-side. Using a complex, clickable system of Google maps with a clean design and corresponding tables, you can drill-down to see all sorts of individual data charted out in an appealing red color-scheme, along with a listing of houses that are currently on the market in the neighborhood. You can even click on individual properties to see the historical and current valuations not only of the property in question,but of all the properties nearby. The warm yellow used to display the property tracts on the map invites the user to mouse over all the houses to see highly stylized infoWindows with more information. It's really hard to find anything about the navigation, interface and design to complain about. The only thing that might possibly make the app better is adding interactivity to the static charts on the neighborhood and property pages.

Obviously, all of this data is of immense value to users on an evergreen basis, not just a transitory news cycle. What's also impressive is that it's useful for both interested home buyers looking to browse the marketplace and for current home-owners who want to see the valuations of their home compared to nearby homes. For a small fee, the app even lets you download a custom report with all of that information contained within it upon entering your address. And, just in case a homeowner suspect his or her home may be overvalued, the interface includes a handy guide to protesting your valuation with local government agencies.

On a whole, Curbwise is the epitome of a solid, innovative app built by a news organization that works to protect consumers and inform the public. Even better, the money made off custom report sales provides the paper with an additional revenue stream that likely helps offset the loss in print advertising in recent years.

On the importance of localism

less than 1 minute read

A decade before the rise of the Internet set in motion the disruption of legacy news business models, Kaniss foresaw the growing need for local and regional news to unite increasingly fragmented, suburbanized communities.

Critique, “French wine map shows the best vintage, from 1978 to 2011”

4 minute read

It's nearing the end of the week, so what better way to relax than with a good bottle of wine and some leisure reading? Problem is, I'm not very skilled at buying wine that tastes any good. I always end up paying more for the bitter, expensive stuff. Fortunately,  there's a pretty cool news app for that. The Telegraph UK's recent interactive app on French wine ratings allows users to browse through the years to see which regions of the country produced the best-tasting wines in each year. With a handy HTML5 slider on the bottom, the user can locate the year of the bottle while in the store, then match that up with the region the bottle was harvested in. Then by mousing over the corresponding region labels, users can get an idea of that year and region's quality as rated by conniseaurs.

The only thing that concerns me is this app's use of mapping when mapping was not required. Granted, the geographical component to this topic is very important and likely justifies a map. But at the same time, the map feels bare with only the wine regions colored and the rest of France empty. What stands out more than anything, however, is the app's use of color. The deep red and pink colors combined with the light green shading not only represents white and red wine visually, but it also gives the app an aesthetically appealing and bright color scheme against the canvas-colored backdrop.

Critique, “Why is Her Paycheck Smaller?”

4 minute read

For my final critique, I decided to look at a more straightforward and well-known visualization on gender wage gaps created by The New York Times back in 2010. The "Why is Her Paycheck Smaller" visualization shows how simple, mostly static scatter plots can sometimes be the most efficient and informative way to tell a story.

Functionality-wise, the visualization is not terribly impressive. Not only does it run on clunky, often-inoperable Flash, but it has little in terms of interactivity. All you basically do is click on each of the occupations to see where the dots for that occupation fall, and then mouse over the dots to see more specific information. The clean, crisp design, on the other hand, makes the colored dots stand out, basking in the surrounding minimalism. The notations help explain possible outliers without cluttering the graph, and the charts on the bottom right put the data into a larger context neatly and concisely.

For its time, this visualization probably was cutting-edge. But despite its less sophisticated technologies looking back now, it communicates just as powerfully as any of the best visualizations do in 2012. The "Why is her Paycheck Smaller" visualization shows that, no matter what technology, good charting, design and editing makes for a strong story. It's easy to get caught up in the technologies, but sometimes less is more.

Should data viz be a specialty or a commodity skill in the newsroom?

8 minute read

An interesting question came up at last Wednesday's Doing Data Journalism (#doingdataj) panel hosted by the Tow Center for Digital Journalism here at Columbia's J-School: Should there be data specialists in the newsroom, or can everyone be a data journalist? For New York Times interactive editor Aron Pilholfer, who participated in the panel, the question is not so much should everyone do data as will everyone do data. And for Pilholfer, the answer to that question clearly seems to be no:

I kind of naively thought that at one time you could train everybody to be at least a base level of competency with something like Excel, but I'm not of that belief anymore. I think you do need specialists.

I've always hated the idea of having technology or innovation 'specialists' in a work environment that should ideally be collaborative. So, at first I tended to disagree with Pilholfer's argument. But what won me over was the reasoning behind his claim. For Pilholfer, it's not that the technology, human talent or open source tools aren't there for everyone to scrape, analyze and process data –– in fact, it's now easier than ever to organize messy data with simple and often free desktop applications like Excel and Google Refine. The problem is that there's a cultural lack of interest within newsrooms, often from an editorial level, to produce data-driven stories. As Pilholfer says in what appears to be an indictment of upper-level editors for disregarding the value of data,

The problem is that we continue to reward crap journalism that's based on anecdotal evidence alone . . . But truly if it's not a priority at the top to reward good data-driven journalism, it's going to be impossible to get people into data because they just don't think it's worth it.

I totally agree, but with one lurking suspicion. As with the top-level editors, many traditional users –– or 'readers,' as one might call them –– still at least think they like to read pretty, anecdotal narratives, and tend not to care as much whether the hard data backs them up. In other words, it’s an audience problem just as much as it is a managerial or institutional one. Some legacy news consumers just still aren't data literate. Because they're not accustomed to even having such data freely available to them, they don't even value having it. As the old saying goes, "You can't miss what you never had." Yet as traffic and engagement statistics continually confirm, as soon users have open data readily available to them through news apps and data visualizations, they spend more time accessing the data than they do reading the print narrative.

Aron Pilholfer at #doingdataj

less than 1 minute read

Totally agree, but harbor the lurking suspicion that many traditional readers still like to read pretty narratives and don't care as much if the facts back them up. In other words, it's an audience problem just as much as it is an editorial one.