TroySchneider.com

Brookings Maps Manufacturing in America

Troy K. Schneider — Thu, 17 May 2012 20:39:12 +0000

Largely under the radar (or at least under my radar), the Brookings Institution's Metropolitan Policy Program has been producing some very interesting data visualizations to complement their research. The latest digs deep on manufacturing in the United States -- detailing not just the geographic distribution of plants, but also industry specialties, wage variations, plant sizes and more.

Unfortunately, it's not particularly share-able -- I'd love to drill down on wage variations in the midwest, and embed that view in a blog or article -- but it it's very data-rich, and nicely done all around. Definitely worth checking out.

Cross-posted at http://data.newamerica.net

Data Without Borders DC DataDive: March 2-4 at New America

Troy K. Schneider — Thu, 23 Feb 2012 18:53:15 +0000

My New America colleagues and I are excited to be working with Data Without Borders to host their Washington, DC, DataDive.

More details are at http://datawithoutborders.cc/events/dcdatadive/ -- learn more, and sign up to attend!

Introducing New America's Map of the Week with Slate

Troy K. Schneider — Fri, 10 Feb 2012 00:00:26 +0000

Today New America launched a new "Map of the Week" dataviz feature with Slate. Each Thursday, we'll be working together to map some policy or social story of note.

The first edition details the explosion of mobile-phone subscriptions around the globe -- and hints at how those growth rates alone aren't enough to transform development efforts in Africa and elsewhere. (These issues were explored in more detail today at our Global Assets Project/Open Technology Initiative event, "Mobile Disconnect: Can Mobile Solutions Really Combat Global Poverty?").

For the complete picture, go to Slate.com.

There are lots of interesting maps in the pipeline for the coming weeks, but we're always on the lookout for additional ideas. So if you have a dataset that deserves attention, please let us know.

Cross-posted on http://data.newamerica.net

On the Drug Wars in Mexico -- and the Challenges of Working with Real-World Data

Troy K. Schneider — Mon, 06 Feb 2012 18:42:33 +0000

Last week, FlowingData's Nathan Yau featured this great interactive map by Diego Valle-Jones. The map, with its rich data and myriad filters and options, is well worth showcasing -- be sure to check it out at http://www.diegovalle.net/drug-war-map.html.

What really caught my eye, however, were the explanations that Valle-Jones included, and what they and the map together reveal about the real-world challenges of in-depth data visualization.

Among the things worth noting:

Consistent, complete data are rare.

Valle-Jones writes that not all the homicides were reported with a specific date -- which is something of a problem when trying to display time-series data! -- and explains that such cases are "assumed to have occurred on the month they were registered." A similar best-guess effort is made to adjust for under-counting, and for homicides that took place in one year but were reported in the next. Care is taken to explain what decisions were made and why -- but that's hard to do in map legend itself. And even with accompanying blog post, I would wager that most people who look at the map will still assume the numbers are some sort of God's Truth.

New America has similar issues with its tracking of U.S. drone strikes -- our map is the definitive source for those without a CIA security clearance, but it's not like the government releases geo-coordinates for each strike. We follow the reports carefully, screen out the uncorroborated stories, and paint as accurate a picture as we can. People need to remember: Nature may abhor a vacuum, but the world at large carries a pretty serious grudge against well-structured datasets.

One map can show only so much.

Valle-Jones' project is pushing the limit of what an ordinary viewer/reader can absorb; his map is not so much a straight visualization as it is a a full-blown data browser. That's not a critique -- the world needs both, and the latter was clearly Valle-Jones' intent. But with the newest mapping tools and javascript libraries allowing us to cram ever more data into a single presentation, it's important to remember: Are you trying to tell a specific story? Or are you aiming to help others dig and explore to craft their own narratives?

Complexity is, well, confusing.

This relates directly to the previous point. With drug-smuggling routes, seven years of homicide counts (both drug-related and total), drug-eradication zones and more, this map's tooltip/layer-switcher starts to resemble the screen of a Bloomberg terminal. It was my fourth visit before I realized that the checkboxes controlled additional layers on the map, not (as I'd assumed) filters on the homicide data. And with the legend relegated to an overlay, there's no way to look at both it and the map it explains at the same time.

Rich visualizations should spark deeper discussions.

The world is awash in infographics that string together some provocative stats, get shared and retweeted widely, but never lead people to the underlying data and issues. This Mexico map is different -- not least because Valles-Jones seems to know that he can't tell the full story in this form alone. So he lets you download the data in full. Points to the original sources, as well as other, complementary, maps and essays on these issues. And he adds his own supplementary data tables and charts to flesh out the questions -- both policy- and data-related -- that seem especially ripe for further exploration.

It's really fantastic, and I hope we see more efforts like it.

Cross-posted on data.newamerica.net

Hear Us Now? A California Survey of Digital Technology's Role in Civic Engagement and Local Government

Troy K. Schneider — Sun, 23 Oct 2011 18:30:49 +0000

Policy Paper, New America Foundation

Hidden in all the bad news about California’s troubles is this delightful paradox: Californians, while living in a state that experts say is ungovernable, have within their reach new tools that give them greater power to govern themselves than ever before.

Technology is the reason. Often with little public notice or scrutiny, most of California’s 5,000-some local governments are experimenting with technologies to engage the public and improve services. The sophistication of this use of digital technologies for citizen interaction — referred to as eGovernment, digital government, or Government 2.0 — varies. The benefits are wide-ranging.

You can go on-line to have the city police in Santa Clarita check on your home while you’re on vacation. In Pebble Beach, you can add yourself to the Community Services District’s database of local people that need special assistance in the event of an emergency evacuation. You can schedule a visit to your cousin in jail via the Santa Clara County web site or public kiosks. If you need to appear in court or qualify yourself for social services in Nevada County, you can avoid long drives over windy, snowy roads by finding one of the 60 county video cameras set up for direct conferencing with local government. And if you’re a truant in Anaheim, you can avoid school reassignment or prosecution by carrying a hand-held tracking device, provided by your school district and the city police, that monitors your location throughout the day.

The timing of eGovernment’s rise is at once problematic — and fortunate. Public frustration with government and cuts in public spending are natural obstacles to launching new programs. But the same factors also create an opportunity to redesign how government interacts with, and services, the public. Technology, if deployed wisely and efficiently, may provide better engagement, better information and better service delivery, at less of a price. How is California doing so far at this task? The early results are uneven. California’s powerful culture of innovation has produced clear progress from the days of simple government web sites. But the progress has been unevenly distributed. And success stories have yet to be identified, much less encouraged and disseminated. When it comes to eGovernment, Californians don’t know what other California are doing, don’t know what works, and don’t know how to measure success.

Californians deserve — and should demand — a basic level of technology-driven service and engagement, just as they do with analog government services such as emergency response and sanitation.

This report provides a starting point for moving the state in that direction, by documenting some of many innovations already underway, and charting where California could go in the future.

It focuses on California’s local governments, because they provide most services and are the level of government with which most Californians most often engage. As importantly, local governments are responding to that heightened engagement, and heightened expectations, with experiments in technology that are both more expansive and citizen- focused than those that states and countries generally have undertaken. Thus, the search for promising practices that have the potential to transform the citizen-government relationship starts locally, and the innovations we find there can be expanded to serve broader populations tomorrow.

Click here to download the full report.

Attachment	Size
PDF of Full Report	875.38 KB

Cool New Mapping at New America

Troy K. Schneider — Wed, 22 Jun 2011 01:11:51 +0000

We've been playing with TileMill at New America as we do more with mapping and data visualization. Much of this work has been spearheaded by the good folks at DevelopmentSeed, but this sample below is entirely home-grown. (We're learning!)

This map shows the 300+ participants in this weekend's Deliberative Poll in California. You can mouse over any point to learn a bit about the individual in question, as well as some basics about the part of California in which he or she lives. You can also zoom and a drag the map to zero in on a particular region.

If you'd like to embed this map on your own blog or website -- whether for the whole state or to zoom in on particular region -- you can customize and then grab the embed code here.

A Better Way to Map Arlington Cemetary

Troy K. Schneider — Fri, 06 Aug 2010 02:45:01 +0000

As Thursday’s congressional hearing and the Army’s own investigation have made clear, the dysfunctions at Arlington Cemetery go much deeper than the antiquated record-keeping. Yet that lack of a real tracking and mapping system clearly contributed to the errors detailed by Washington Post reporters -- and the fact that 11 years, 35 contracts and $5.5 million have failed to produce a usable database is appalling in its own right.

It doesn’t have to be this way. Yes, technology projects are complicated, and government contracts have challenges all their own. But Arlington Cemetery -- with its 300,000 graves and few hundred acres -- is not *that* big of a data project. And there are countless current examples, using open-source tools and widely-used software, that show just how much can be done for a fraction of the funds wasted so far.

Take, for example, earthquake-ravaged Haiti -- a place most would agree faces resource and governance challenges that dwarf Arlington’s. Today, the definitive map of Port-au-Prince and the surrounding areas is on Open Street Map -- a free, global, volunteer-driven project that aims to provide “a free editable map of the whole world.” In the hours after the Jan. 12 quake, volunteers on the ground began updating and improving the maps to facilitate relief efforts -- giving aid workers a way to navigate the confusing and often-blocked streets. Then on Jan. 14, high-resolution satellite images of Haiti were released, and volunteers worldwide began “tracing” and annotating -- not only roads, but buildings, obstacles and even communities of displaced Haitians. Within days, virtually every corner of the region had been accurately mapped -- no contract required.

More recently, InterAction -- a D.C.-based non-profit that focuses on aid and development -- launched HaitiAidMap.org. This site inventories, at last count, 1,045 aid projects in Haiti involving 87 different organizations. This project did involve a contractor -- which billed all of $40,000, and took less than a month to deliver the final project.

HaitiAidMap.org, of course, started with much cleaner data than the cemetery, with its index cards and “hand-drawn maps on six-foot-wide sheets of paper,” can provide. But as the Open Street Map project has proven, in Haiti and elsewhere, amazing things are possible with GPS, satellite images, and careful note-taking.

The basic mapping of Arlington Cemetery, in fact, is already done in Open Street Map. One woman, contributing to the project in her spare time, has added cemetery roads, monuments, several sections and a handful of notable burial plots. And as part of Open Street Map, that data is freely available for the Army to use and extend as it pleases.

Surveying and cross-checking every single burial plot would take significantly more effort, of course, and a functioning records system for our most hallowed cemetery requires more than just accurate and attractive electronic maps. But surely we can get more for our time and money than cemetery officials have managed over the past decade.

So next Memorial Day, when soldiers from the 3rd U.S. Infantry place flags on the graves of service members (a ritual that, according to the Arlington National Cemetery website, takes roughly three hours), let’s equip them with some GPS units and make a day of it. Or better yet, let’s round up volunteers and do it now. The heroes buried at Arlington deserve a system that works -- and so do the American taxpayers.

Hitting the Reset Button

Troy K. Schneider — Mon, 18 Jan 2010 22:20:00 +0000

OK, this is ridiculous. In the interim since my last post, I could have had a baby -- except, y'know, for that whole being-male thing.

So this post is to officially shame myself into re-thinking and re-tooling the site. Either start posting again, or change things around to pull in my updates from elsewhere. Or maybe both.

Google Gets That Much Closer to Indexing F-ing Everything

Troy K. Schneider — Tue, 08 Sep 2009 18:22:52 +0000

Looks like Google is serious about that whole "organize the world's information and make it universally accessible" thing.

The official Google Blog announced today that the company (which, full disclosure, has a connection or two with my employer, the New America Foundation) is "partnering with newspaper publishers to digitize millions of pages of news archives."

That's big. Big for democratizing access to information, which I think is great. Big for newspapers themselves, which might be able to make more by splitting the revenues with Google than by muddling through attempts to monetize their archives alone. Big -- and bad -- for the Westlaws and Nexises (Nexiae?) of the world, which charge a decent chunk of change for combined newspaper archives. And probably big for IP and copyright attorneys, as former stringers, press guilds and others inevitably raise new questions about just who owns the words and pictures on those yellowed old pages and pieces of microfiche.

Most interestingly, though, it's potentially huge for Google -- or more precisely, for Google's reputation.

Over the weekend, John Battelle wrote that Google's recent foray into satellite imagery might be dangerously close to "the company's Waterloo - a viral meme that Google is sensing too much, knows too much, and is too powerful." Somehow, I think that digitizing "billions of pages of newsprint" creeps a heckuva lot closer to that line.

But then again, Google's already done it to the book business...

Talking Feed-in Tariffs on Etopia News Now

Troy K. Schneider — Fri, 10 Apr 2009 02:07:52 +0000

I did a quick interview yesterday with Etopia News Now on feed-in tariffs and the event New America is convening Friday in partnership with the Washington Monthly:

It's an interesting topic -- but if you really want to understand the idea and its potential, read Mariah Blake's Washington Monthly article.