African News Challenge funds data journalism and open government tech

The post-industrial future of journalism is already here. It’s just not evenly distributed yet. The same trends changing journalism and society have the potential to create significant social change throughout the African continent, as states moves from conditions of information scarcity to abundance.

That reality was clear on my recent trip to Africa, where I had the opportunity to interview Justin Arenstein at length during my visit to Zanzibar. Arenstein is building the capacity of African media to practice data-driven journalism, a task that has taken on new importance as the digital disruption that has permanently altered how we discover, read, share and participate in news.

One of the primary ways he’s been able to build that capacity is through African News Innovation Challenge (ANIC), a variety of the Knight News Challenge in the United States.

The 2011 Knight News Challenge winners illustrated data’s ascendance in media and government, with platforms for data journalism and civic connections dominating the field.

As I wrote last September, the projects that the Knight Foundation has chosen to fund over the last two years are notable examples of working on stuff that matters: they represent collective investments in digital civic infrastructure.

The first winners of the African News Innovation Challenge, which concluded this winter, look set to extend that investment throughout the continent of Africa.

“Africa’s media face some serious challenges, and each of our winners tries to solve a real-world problem that journalists are grappling with. This includes the public’s growing concern about the manipulation and accuracy of online content, plus concerns around the security of communications and of whistleblowers or journalistic sources,” wrote Arenstein on the News Challenge blog.

While the twenty 2012 winners include investigative journalism tools and whistleblower security, there’s also a focus on citizen engagement, digitization and making public data actionable. To put it another way, the “news innovation” that’s being funded on both continents isn’t just gathering and disseminating information: it’s now generating data and putting it to work in the service of the needs of residents or the benefit of society.

“The other major theme evident in many of the 500 entries to ANIC is the realisation that the media needs better ways to engage with audiences,” wrote Arenstein. “Many of our winners try tackle this, with projects ranging from mobile apps to mobilise citizens against corruption, to improved infographics to better explain complex issues, to completely new platforms for beaming content into buses and taxis, or even using drone aircraft to get cameras to isolated communities.”

In the first half of our interview, published last year at Radar, Arenstein talked about Hacks/Hackers, and expanding the capacity of data journalism. In the second half, below, we talk about his work at African Media Initiative (AMI), the role of open source in civic media, and how an unconference model for convening people is relevant to innovation.

What have you accomplished at the AMI to date?

Justin Arenstein: The AMI has been going on for just over three years. It’s a fairly young organization, and I’ve been embedded now for about 18 months. The major deliverables and the major successes so far have been:

  1. A $1 million African News Innovation Challenge, which was modeled fairly closely on the Knight Challenge, but a different state of intended outputs.
  2. A network of Hacks/Hackers chapters across the continent.
  3. A number of technology support or technology development initiatives. Little pilot projects, invariably newsroom-based.

The idea is that we test ideas that are allowed to fail. We fund them in newsrooms and they’re driven by newsrooms. We match them up with technologists. We try and lower the barrier for companies to start experimenting and try and minimize risk as much as possible for them. We’ve launched a couple of slightly larger funds for helping to scale some of these ideas. We’ve just started work on a social venture or a VC fund as well.

You mentioned different outputs in the News Challenge. What does that mean?

Justin Arenstein: Africa hasn’t had the five-year kind of evolutionary growth that the Knight News Challenge has had in the U.S. What the News Challenge has done in the U.S. is effectively grown an ecosystem where newsrooms started to grapple with and accepted the reality that they have to innovate. They have to experiment. Digital is core to the way that they’re not only pushing news out but to the way that they produce it and the way that they process it.

We haven’t had any of that evolution yet in Africa. When you think about digital news in African media, they think you’re speaking about social media or a website. We’re almost right back at where the News Challenge started originally. At the moment, what we’re trying to do is raise sensitivity to the fact that there are far more efficient ways of gathering, ingesting, processing and then publishing digital content — and building tools that are specifically suited for the African environment.

There are bandwidth issues. There are issues around literacy, language use and also, in some cases, very different traditions of producing news. The output of what would be considered news in Africa might not be considered news product in some Western markets. We’re trying to develop products to deal with those gaps in the ecosystem.

What were the most promising News Challenge entrants that actually relate to those outputs?

Justin Arenstein: Some of the projects that we thought were particularly strong or apt amongst the African News Challenge finalists included more efficient or more integrated ways to manage workflow. If you look at many of the workflow software suites in the north, they’re, by African standards, completely unaffordable. As a result, there hasn’t been any systemic way that media down here produced news, which means that there’s virtually no way that they are storing and managing content for repackaging and for multi-platform publishing.

We’re looking at ways of not reinventing a CMS [content management system], but actually managing and streamlining workflow from ingesting reporting all the way to publishing.

Some of the biggest blogs in the world are running on WordPress for a CMS. Why not use that where needed?

Justin Arenstein: I think I may have I misspoken by saying “content management systems.” I’m referring to managing, gathering and storing old news, the production and the writing of new content, a three or four phase editing process, and then publishing across multiple platforms. Ingesting creative design, layout, and making packages into podcasting or radio formats, and then publishing into things like Drupal or WordPress.

There have been attempts to take existing CMS systems like Drupal and turn it into a broader, more ambitious workflow management tool. We haven’t seen very many successful ones. A lot of the kinds of media that we work with are effectively offline media, so these have been very lightweight applications.

The one thing that we have focused on is trying to “future-proof” it, to some extent, by building a lot of meta tagging and data management tools into these new products. That’s because we’re also trying to position a lot of the media partners we’re working with to be able to think about their businesses as data or content-driven businesses, as opposed to producing newspapers or manufacturing businesses. This seems to be working well in some early pilots we’ve been doing in Kenya.

What were your takeaways from the Tech Camp? Was a hybrid unconference a good model for the News Challenge?

Justin Arenstein: A big goal that we think we’ve achieved was to try and build a community of use. We put people together. We deliberately took them to an exotic location, far away from a town or location, where they’re effectively held hostage in a hotel. We built in as much free time as possible, with many opportunities to socialize, so that they start creating bonds. Right from the beginning, we did a “speed dating” kind of thing. There’s been very few presentations — in fact, there was only one PowerPoint in five days. The rest of the time, it’s actually the participants teaching each other.

We brought in some additional technology experts or facilitators, but they were handpicked largely from previous challenges to share the experience of going through a similar process and to point people to existing resources that they might not be aware of. That seems to have worked very well.

On the sidelines of the Tech Camp, we’ve seen additional collaborations happen for which people are not asking for funding. It just makes logical sense. We’ve already seen some of the initial fruits of that: three of the applicants actually partnered and merged their applications. We’ve seen a workflow editorial CMS project partner up with an ad booking and production management system, to create a more holistic suite. They’re still building as two separate teams, but they’re now sharing standards and they’re building them as modular products that could be sold as a broader product suite.

The Knight News Challenge has stimulated the creation of many open source tools. Is any of that code being re-used?

Justin Arenstein: We’ve tried to tap into quite a few of them. Some of the more recent tools are transferable. I think there was grand realization that people weren’t able to deliver on their promises — and where they did deliver on tools, there wasn’t documentation. The code was quite messy. They weren’t really robust. Often, applications were written for specific local markets or data requirements that didn’t transfer. You actually effectively had to rebuild them. We have been able to re-purpose DocumentCloud and some other tools.

I think we’ve learned from that process. What we’re trying to do with our News Challenge is to workshop finalists quite aggressively before they put in their final proposals.

Firstly, make sure that they’re being realistic, that they’re not unnecessarily building components, or wasting money and energy on building components for their project that are not unique, not revolutionary or innovative. They should try and almost “plug and play” with what already exists in the ecosystem, and then concentrate on building the new extensions, the real kind of innovations. We’re trying to improve on the Knight model.

Secondly, once the grantees actually get money, it comes in a tranche format so they agree to an implementation plan. They get cash, in fairly small grants by Knight standards. The maximum is $100,000. In addition, they get engineering or programming support from external developers that are on our payroll, working out of our labs. We’ve got a civic lab running out of Kenya and partners, such as Google.

Thirdly, they get business mentorship support from some leading commercial business consultants. These aren’t nonprofit types. These are people who are already advising some of the largest media companies in the world.

The idea is that, through that process, we’re hopefully going to arrive at a more realistic set of projects that have either sustainable revenue models and scaling plans, from the beginning, or built-in mechanisms for assessments, reporting back and learning, if they’re designed purely as experiments.

We’re not certain if it’s going to work. It’s an experiment. On the basis of the Tech Camp that we’ve gone through, it seems to have worked very well. We’ve seen people abandon what were, we thought, overly ambitious technology plans and rather matched up or partnered with existing technologists. They will still achieve their goals but do so in a more streamlined, agile manner by re-purposing existing tech.

Editors’s Note: This interview is part of an ongoing series at the O’Reilly Radar on the people, tools and techniques driving data journalism.

Beth Noveck: think about how to open up the API of government

Former White House deputy CTO for open government Beth Noveck recently gave a TED Talk on a “more open source government. For more perspective in the vein of “open,” read CNN’s TED summary, “What if you could make anything you wanted?” Noveck’s talk is embedded below:

“…start by teaching young people that we live, not in a passive society, a read-only society, but in a writable society, where we have the power to change our communities, to change our institutions, that’s when we begin to really put ourselves on the pathway towards this open government innovation”

Chicagobuildings.org maps vacant and abandoned buildings using open government data

One of the minds behind the Look at Cook open government data visualization app is at it again. Derek Eder wrote in this week to share another Web app he just launched (ChicagoBuildings.org) and a reminder about what’s happening in Chicago in this space.

This Web app takes 311 reports about vacant and abandoned buildings from the Chicago and visualizes them onto a searchable map. “It’s specifically set up to pull data from Chicago’s data portal,” said Eder, linking to the 311 service requests of vacant and abandoned buildings dataset.

Eder shared more about how mapping Chicago’s vacant buildings in a blog post earlier this week. The results are unsurprising: there are many more vacant buildings in areas with high poverty rates.

Eder said that the app could be used by other cities, depending on how they store or format their data. The code for
Chicago Buildings is on Github. On that front, he says that Chicago “isn’t using Open 311 yet, so this site isn’t either. That being said, it wouldn’t be too hard to hook up the same interface to a different data source.” Code for America will help Chicago to implement Open311 in 2012. Eder shared that he wrote a script that converts Socrata to Google Fusion Tables that could be modified for this purpose.

ChicagoBuildings.org is one of a growing number of civic applications that have come out of Chicago’s open government initiative. As Eder made sure to point out, his app is a finalist in the Apps for Metro Chicago contest, along with 9 other apps, including iFindItChicago and Techno Finder.

In the video below, Elizabeth Park, the creator of IFindit Chicago, talks about how she was inspired to build the team that created an Android app to help homeless and lower income citizens find resources like as shelters, medical clinics,and food pantries.

Voting for the winners ends this Friday, October 14th, so check out the community round entries and weigh in.

As a reminder: If you have open government news to share, you can always find me at @digiphile on Twitter, where I share my email address, alex@oreilly.com.

NASA to launch inexpensive Android ‘phonesats’ into space

Android….in….space! This morning, Will Marshall of NASA showed the Android Open Conference plans for a sub-$10,000 Nexus One “phonesat.” Given that the cost of satellites usually measure in the tens or hundreds of millions or so, that’s a rather spectacular cost savings.

Marshall says that this will be the fastest processor to govern a satellite. For reference, Mars Explorer used a 33MHz processor. It sends signals back via amateur radio packet system, rather than ground tracking. The launch video is embedded below:

Great anecdote: when a launch failed, the entire payload fell without parachute into desert. The data was left intact.

William launching 3 Android phonesats in December, in space for 3 weeks. As O’Reilly Media’s Gina Blaber pointed out, they’re “iterating Silicon Valley-style.”

For more, check out the short documentary below about the PhoneSat suborbital test launch in the Black Rock desert:

PhoneSat Rocket Launch Documentary from Ben Howard on Vimeo.

You can follow @NASA_Phonesat on Twitter — there’s no official website yet – and, according to Marshall, eventually check out code on Github, where NASA is open sourcing some software behind it. (And yes, that’s a big deal.)

Open data and maps tell the local story of unemployment and recovery spending

Washington-based DevelopmentSeed continues to tell dazzling data stories with open source mapping tools. This week, they’ve posted a map of the local impact of unemployment and recovery spending. The map visualizes unemployment rate changes at a county level and folds in total economic recovery spending by the government under the American Recovery and Reinvestment Act of 2009. In the map embedded below, red corresponds to an increased unemployment rate and green corresponds to a lower unemployment rate or job growth. Counties that received less than $10 million dollars in recovery spending have a white pattern.

David Cole explains more in a post at DevelopmentSeed.org:

Over the last year, we see that unemployment dropped in 58% of counties by an average of 0.25 percentage points. On average the Recovery Act funded 31 projects at a total of $24,131,582.47 per county. Nationally this works out to about $282.66 in recovery spending per person.

Overall, it’s impossible to tell for sure how much recovery spending improved the economic situation, because we just don’t know how bad things could have been. It may be the case that without spending, this map would have a lot more red. Or maybe not. What’s interesting here is the local impact and information we are able to see from processing a few sets of open data. Check out how your county is doing compared to its surroundings. How about compared to a more or less urban county nearby?


Strata Conference New York 2011, being held Sept. 22-23, covers the latest and best tools and technologies for data science — from gathering, cleaning, analyzing, and storing data to communicating data intelligence effectively. Save 20% on registration with the code STN11RAD

US DNI releases government open source handbook at MILOSS 2011

The United States Director of National Intelligence released “GOSS for Govies” during this year’s U.S. military open source software 2011 Working Group 3 in Atlanta, GA. The livestream for the MIL OSS conference is embedded below:

The handbook, which is embedded below, offers a guide to doing open source projects inside of government, based upon the DNI’s experience working on the Ozone Widget Framework.

US DNI Government Open Source Handbook