Create a gist now

Instantly share code, notes, and snippets.

What would you like to do?

OSM Tracing Guide for PEPFAR

PEPFAR has developed a series of mapping tasks to support HIV/AIDS prevention and treatment programs. The data that emerge from these mapping tasks will support our understanding of program coverage, the optimization of supply chain logistics, and the analysis of clinical site-level data. In many places where the epidemic is most severe, there is little spatial data on basic features such as buildings or roads. As a starting point, this tracing guide focuses on creating baseline data on buildings and roads

Getting started

We've done our best to supply the best possible imagery for tracing. The current bing imagery is often quite usable. We will focus on the iD editor, which is ideal for beginners. If imagery looks blurry or hard to see, double-check that imagery is set to Custom

You can change the imagery like this:

custom_imagery

If the imagery seems dark, you may want to brighten it a bit:

brighten_imagery2

Buildings

Please label all buildings as general buildings. We will seek additional community engagement and add detailed local knowledge to the map after we complete the task. Square the corners of all buildings by pressing "S" on your keyboard after labeling the building. If you see a round structure, such as a Manyatas make it perfectly round by pressing "O".

Tracing buildings

To trace, select the Area tool (Shortcut: 3) for circled buildings (huts), make a circle instead of an area, then trace the building/wall. Double-click to finish. The process looks like this:

building_trace3

Next, we classify (or "tag") the building, using the buttons on the left side of the screen:

building_classify

Note that once you've traced a building, you can square the corners easily by hovering and clicking the icon (Shortcut: s)

square

Roads

First we trace the roads, then we classify them. The whole process looks like this:

Tracing roads

To trace a road, use the Line tool in iD Editor (Shortcut: 2) and trace along the middle of the road. See below for an example of too few nodes (left), too many nodes (center), and the right amount of nodes (right).

Make sure the roads connect to each other where they cross. You should see a node appear at each intersection (see below).

Classifying roads

We're going to tag roads according to the typology described in this OpenStreetMap Wiki: [Highway_Tag_Africa] (http://wiki.openstreetmap.org/wiki/Highway_Tag_Africa).

road5

For more resources, see:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment