I’ve never done any interactive mapping before, and wanted to test the waters using Processing. After the positive reactions to my post office expansion video, I thought some people might find an interactive version interesting or useful.

I used the code that produced the video as a foundation to build interactivity into the map. The most significant visual change was the move to a dark background with bright data points. While I’m not entirely happy with the “civilization illuminating a dark continent” effect that might be communicated by this choice, I wanted to provide some variation in how opening dates were represented, and on monitors I’ve always found bright colors easier to distinguish from one another than dark.

Interaction allows a more in-depth exploration of the data: this map includes the ability to set the range of dates displayed, zooming and panning, clickable points, and user-controlled playback for a given date range. I tried to make things relatively straightforward: use the buttons at the top or drag and scroll with the mouse to pan and zoom. Click on a highlighted point to show more information. High and low dates can be changed by dragging them along the timeline, while dragging the selected range will move it while maintaining the date gap. Use the buttons to the left of the timeline to play, stop, or change the speed of automated playback.

I don’t have much to add in the way of analysis that wasn’t mentioned in the thread of my previous post. I’d encourage you to explore different date ranges, though – you can see some interesting patterns of activity that were lost in the previous video by zooming and selecting specific dates.

I’m happy with the final result, I learned a lot during this process, and I’m excited to apply what I’ve learned to some other projects I have in mind. I wish the performance were a little better, and in the future I’d like to try my hand at some more browser-friendly interactive projects using Processing.js. I’d welcome any criticisms or suggestions on the design – particularly regarding the user interface and implementation of interactivity.

(HD and 1080p download here. It’s much prettier!)

A few months ago, I scraped post office location information from the USPS Postmaster Finder, and then extracted lat/long coordinates  by correlating placenames to the USGS GNIS. Recently I remembered I had this data sitting around. I’ve been experimenting with Processing a lot lately as a tool for geographic visualization, and decided this would be an interesting dataset to use as a first stab at animation/dynamic mapping. I used parts of zipdecode from Ben Fry’s excellent book Visualizing Data as a foundation for my code. After some frustrations, this visualization is the result.

Of course, there are a few caveats: the USPS admits that their post office data is a constant work in progress, so there are likely many offices that aren’t shown on this map. About 10% of the placenames in the USPS data failed to find any coordinate matches in the GNIS data, so those  don’t show up either. Finally, the USPS data includes closing dates for many offices, but these aren’t represented on the map, which might give a false sense of density in some areas.

Still, I think the results are pretty interesting. I’m no historian, but here are a few interesting patterns I’ve noticed aside from  general westward expansion:

• 1776 – Several new post offices crop up along the east coast after the Revolution.
• 1846 – Rash of openings in Texas after statehood and the end of the Mexican-American War.
• 1848 – First offices established on the west coast, with lots of activity afterwards, likely due to gold rushes and CA statehood.
• 1851 – New Mexico and Utah start to see some activity as they become territories. I especially like the lines extending from Santa Fe along the Rio Grande / El Camino Real.
• 1860s – No activity in the South during the Civil War; also an interesting sweep across the Great Plains, with Oklahoma remaining conspicuously quiet.
• 1870s – Distinct traces along railroads in Nebraska and Kansas.
• 1890s – Oklahoma lights up due to several land rushes.

Anyone notice anything else interesting? Let me know.

Update: Very cool parallels to this visualization of newspapers in the U.S. 1690 to 2011, from Stanford’s Rural West Initiative!

I’m fascinated by how the meanings of places change from one person to the next. A given place could be sublime or mundane: it depends on who you ask. One example springs to mind. Much has been written (and photographed) of Detroit’s decline in recent years, so much so that although I’ve never visited the city it’s become linked in my mind to images of stagnation and urban decay (even though I recognize that isn’t necessarily the case). In some sense I have a feel for the place, even though I’ve never even come close to visiting it. I think there are some implications to my feeling as if I understand a place that I’ve never been to. You could certainly argue that anyone’s feelings about a place are always filtered and biased, and I would agree. But I think this is different.

How do I construct my feelings about a place that I have never visited? Increasingly, the internet acts as a significant source of our information about distant locations. With the emergence of crowd-sourcing and social media, we often tacitly assume that the content we find online is representative of a variety of viewpoints. As descriptions of places online become more and more specifically geolocated, they become vitally associated with the places they describe. I think, though, that we often fail to question whether what we see online is really representative of a place.

The internet isn’t nearly as open as many give it credit for being, and this extends to online portrayals of real places. The digital divide – differences in education, age, economic position, culture, location – affects who wants to/can/does author the descriptions of places we find online. I’m interested at getting underneath some of these issues, which was the motivation behind my making this map, showing concentrations of urban decay in the US as defined by the tags on photos uploaded by flickr users. Data-wise, I used the fantastic flickr API to mine lat/long coordinates of any photos in the US with the tag “urban decay”. I filtered the results to remove mass uploads by individuals, and (after some confusion) removed results that referenced an inconveniently named brand of makeup.

I don’t argue that this map provides an accurate portrayal of people’s opinions in broad terms. Instead, I wanted to embrace the biases inherent in the flickr user base to try to map how the photos there might represent (and perhaps distort) the world. You may be right to argue that this verges on triviality, but at the same time, maybe not. Flickr is a pretty popular website. I’d say that alone confers some significance to any content uploaded to it. To what extent could the patterns on this map affect how people who have never been to those cities think of them? Other types of media change how we think of places, but I think the faux-democratic ubiquity and dynamism of online social media makes it somewhat unique. I’m in the early stages of doing some similar work for my thesis, diving deeper into data from flickr to ask some questions about how the US/Mexico border is represented online. A lot of the inspiration for this map came from the floatingsheep folks as well as Eric Fischer’s flickr stream, both of which offer some interesting explorations of crowdsourced data.

Update: Interesting comparisons with the population change maps recently posted at Data Pointed…

Generic place names (or toponyms) such as Cumberland Gap or Mount Rainier provide general categorical descriptions of a geographic feature, in contrast to specific toponyms, which provide a unique identifier: Lake Huron. This map taps into the place names contained in the USGS National Hydrography Dataset to show how the generic names of streams vary across the lower 48. Creeks and rivers are symbolized in gray due to their ubiquity (although the etymology behind the American use of creek is interesting), while bright colors symbolize other popular toponyms.

Lite-Brite aesthetic notwithstanding, I like this map because it illustrates the range of cultural and environmental factors that affect how we label and interact with the world. Lime green bayous follow historical French settlement patterns along the Gulf Coast and up Louisiana streams. The distribution of the Dutch-derived term kill (dark blue) in New York echoes the colonial settlement of “New Netherland” (as well as furnishing half of a specific toponym to the Catskill Mountains). Similarly, the spanish-derived terms rio, arroyo, and cañada (orange hues) trace the early advances of conquistadors into present-day northern New Mexico, an area that still retains some unique cultural traits. Washes in the southwest reflect the intermittent rainfall of the region, while streams named swamps (desaturated green) along the Atlantic seaboard highlight where the coastal plain meets the Appalachian Piedmont at the fall line.

There are a few patterns on the map that I haven’t been able to figure out. West Virginia shows a sharp north to south division between runs and branches that continues to puzzle me. Some other geographic patterns I’ve noticed in WV largely run parallel to the Appalachians, from the SE to NW. I don’t know much about the area, though, and I have no idea what could be behind such a distinct division. Any West Virginia-ites willing to take a stab? I’m also intrigued by the patch of branches in southwestern Wisconsin, which I suspect may have something to do with the diffusion of naming practice by way of branch-loving Appalachian miners during a regional lead mining boom in the early 19th century.

This map originally came from a late 2009 project in a class by Joby Bass. If I remade it now, I’d probably try to negotiate some of the overlaps in symbology that happens in very crowded areas, but I still think it’s interesting as-is. If you are interested in learning more about toponyms, George Stewart’s Names on the Land is an engaging classic on naming practices in the US, and there are more specific articles about stream names from Wilbur Zelinsky and Robert West.

Edit: It inexcusably slipped my mind, but a tip o’ the hat to Bill Rankin  for design inspiration!

Update: James Cheshire over at Spatial Analysis has posted a version for the UK. Very cool to see which names did – or more interestingly, the ones that didn’t – make the jump to the US. For example, why did brook have such influence in New England, when it doesn’t seem to be very pervasive on James’ map? Was brook more popular in the UK in the past? Interesting stuff!

Update 2: Pfly posted this link to his flickr page in the comments below, with some interesting maps of toponyms pulled  from the Geographic Names Information System (the underlying dataset for the names I queried from the NHD). Some stream patterns are much easier to see when only one or two names are mapped at a time, and the set also includes some nice maps of non-watery toponyms, as well.