Broadband Access in Alameda County

The digital divide is a very real and very stable reality in communities like Oakland, California.  Knowing which neighborhoods have solid access to high speed internet is a critical aspect of planning for government and nonprofit provided online services- if we want low income folks from Oakland’s flatlands to use a new digital application, we’d damn sure better know how many households in the target areas likely have decent speed internet hookups at home!  Luckily for us the FCC collects reliable data on this and they publish it freely at a local level

http://open-oakland.cartodb.com/viz/1d8f9768-d7b9-11e3-b620-0e73339ffa50/embed_map?title=true&description=true&search=true&shareable=true&cartodb_logo=true&layer_selector=false&legends=true&scrollwheel=true&fullscreen=true&sublayer_options=1&sql=&sw_lat=37.655014078010716&sw_lon=-122.55455017089844&ne_lat=37.91224232115994&ne_lon=-121.67564392089844

Do yourself  a favor and view the fullscreen version: http://cdb.io/1jq8wDq

I took the raw tract level data and joined it to census tracts in QGIS, calculated a new string field called “res_fhsc_per_1000hhs” and calculated the real rate values to display in the map legend and popup- the raw data contains coded values that correspond to real numbers- so 5 means a rate of 800-1,000 per 1,000 households. The GeoJSON file was then loaded into a CartoDB mapping system. 

As with most social phenomena, Oakland’s east and western flatlands stand out as parts of the county with quite low home broadband. Those communities may have internet via very slow services that many modern web sites won’t run well over (these data include all services providing over 200kbps – try using the web on a 256k plan in 2014!).  The data are for 2010 Census tracts and were last collected and published for December 2012. Many households will have improved access since then and we also know from Pew research that many minority communities use mobile devices as a primary means of internet access.

Advertisements

Data Collection Community Style

I had the chance to see a preview today of the Detroit team’s app called LocalData today- part of the Code for America program.  Once again it’s very exciting to see what a team of outside designers, technologists, planners and data geeks can do in the right environment and with the right pretext. For me this is a liberating and exciting trend. Having come up with ways to do the exact same thing in my work I’m painfully aware of how clunky and onerous it is to set up the tools to conduct and manage and publish survey data- our city has been surveyed to death and the results are typically in the filing cabinets of nonprofits and government departments. useless.

Take a quick look at the video showing what LocalData is and does and read on..

LocalData Demo by Code for America from CfA Detroit on Vimeo.

These kinds of newly developed tools would have made a huge difference in the development and implementation of our last major community survey project with a team of high schoolers from Youth UpRising in East Oakland (full article here).

Typically if a community group, organizing group or public agency want’s to collect data they need to call some experts. And we don’t come cheap typically. When we started this project the ask was to help their youth survey their community, so we helped them to identify the focus- park conditions and safety, property conditions and healthy food availability.

Paper sucks, in most instances. We ended up choosing to purchase some discounted Trimble Juno GPS units and got donated ArcPad software from ESRi along with ArcGIS Mobile that came with our server licenses. At the time (mid 2010) the new Mobile software was terrible, super unstable on any setup we tried, bugs the development team couldn’t solve, so out went the more elegant interface and in went an ArcPad project. Trouble is no-one else on our team had had the joy of working with this product before, so fun for me… The devices were loaded with all the parcel data for their community and a bunch of survey screens to gather condition data- we copied a property survey done in another major city to have some form of data standard in the end. Let me just add that even for a seasoned GISP the learning curve and setup time for this app is not small or pretty.

The units froze in the field, crashed, lost data and all kinds of fun that frustrated the students and added time and cost to managing the project. I don’t think we had more than three students actually like using the device/app. But we got out data for the parks and for every property in the tract. 1,000 parcels surveyed. Every park in East Oakland surveyed (there we used a survey template from a HEAL group in Richmond to again try to get some comparable data in our region). But all the data had to be managed, processed and mapped by our team of researchers.

The finished poster of the project is below. The photos we had to take later with a real camera as the Juno devices would only provide ArcPad with a 160×120 image, awesome, that’s like 0.01 megapixels?? Another reason why a smartphone solution just kills it. Once we finished a friend of ours who is a java guru suggested next time we get a Java app built for a smartphone, faster and cheaper. This is where the Code for America approach is powerful. Most small nonprofits don’t have a budget for app developers and we have a hard time convincing managers to fund app/automation development up front to produce a reusable product that saves serious time later and can be reused elsewhere freely.  So we’re looking forward to implementing this new tool in Oakland and the east bay, especially on a full open stack with PostGIS! But we’ll still need funding to help make this happen 😉

Housing conditions

Oh, and as we believe strongly in open data this project’s data IS available in shapefile, excel or csv.

To see more about this great CfA project, to copy their code and more, hit up:

http://golocaldata.com/