What do maps mean to us?

Posted by athaikdin on 7 September 2015 in English (English)

It sometimes still comes as a surprise to me that large swathes of the planet are completely unmapped. With the advent of easily accessible technology (i.e. Google and Bing maps) and satellites, I’d have thought that the entire planet would’ve been mapped years ago. I mean, look at Google Maps. It gives you the option to view almost the entirety of the UK in either satellite or traffic view. It’s this high level of accessibility to maps and the illusion that our surroundings have been fully mapped, I think, that prevent most of us in developed, urban environments from becoming fully aware of the lack of maps in underprivileged areas.

Before I became fully involved in the Missing Maps project, I always thought that satellite images equalled maps. If there’s a satellite image of an area, went my thought process, how difficult could it be to trace those buildings, roads, and the natural landscape? From my experience with Missing Maps, I’ve come to learn that the answer is ‘very difficult’. The world is a very big place, and it takes a lot of effort, concentration, and experience to remotely map an area accurately. Anyone who’s done a small grid of an urban area through the OSM Tasking Manager can probably testify to the complexity of remote mapping. Buildings in other countries are often completely different to what you are used to in your own country. Sometimes what you think is a large apartment block is actually an hospital. And the problems don’t stop there. Satellite images don’t tell you road names or where the boundaries of a district end. You need local knowledge to fill in these gaps, and that’s where the Field Papers stage of the mapping cycle kicks in. A satellite image can visually show you an area, but it’s how you interpret what’s on the ground that matters most because it’s what gives the map its meaning.

A point similar to this was made in a podcast episode I was listening to a few days ago about North American buckskin maps from the late 1700s. Buckskin maps were originally created by Native Americans to help them navigate around their surroundings and plot concentrations of forestry, animals, and water. Essentially, it reflected how they saw life around them - as a source of nourishment. But when American land companies began moving into the very same areas, they hired local Native American guides to create maps for them to plot the locations of profitable land that could be developed industrially. Unlike the Native Americans who were one with their surroundings, the American businessmen mostly sought to exploit the land for profit. The same maps of the area were being used for two, completely different purposes because of how each group interpreted the land.

I’d like to wait and see how the local populations of the places we’ve mapped remotely will use maps created through Missing Maps. What data will they input in themselves? What will they use the maps for? Will it be for navigation around their area or for negotiation between communities? At the moment, NGOs are using ‘missing maps’ mainly for disaster preparedness, assessments, and epidemiological surveys. Hopefully someday soon, we’ll see the full potential of this project of ours.

Comment from redsteakraw on 7 September 2015 at 15:30

I mapped a section of Dar es Salaam, Tanzania earlier this summer as part of Missing maps. I added road networks and building footprints to the area. Basic stuff but building footprints can get a bit dense in some residential areas. I checked back to see what happened to the area I mapped and to my surprise it all ready had shop POIs, building addresses, footpaths and other on the ground information added. I have employed similar techniques where I trace buildings and do road fixups then survey an area to add POIs. I call this method the “Double Tap”. The base information added is a nice platform to add the rest of the features people may care about.

Even incomplete data helps, the more information that is added the more use-cases the data can provide. A road network with no road names can allow you to navigate to other known geographic regions. Place names allow you to find where villages, towns and cities are and can aid navigation to those areas. Having a road network mapped allows you to navigate between towns and cities and keep you on the right road even when forks make that ambiguous. Having road names then allows people to navigate to even more specific places in a town / city knowing the closest cross street aids with the navigation to the closest section in that road to the destination. Then addresses allow you to find the exact destination without the need of cross street information. POIs allow you to semantically find places you may want to go to or facilities that satisfy your needs. Where is the closest cinema? Where can I get water? Where can I get Chinese food? These are all questions that can be answered with POIs.

POIs allow people to add the features they care about into the map. In the end POIs are the last thing to fully complete as it is a moving target and constantly changing. It requires the most work and diligence to keep it reliable and in the end provides locals with the most tangible benefits. In the end locals need to know how to edit, and how to make use of the data so that it fits their use cases and contains the information they care about. This last part is the problem that needs to be solved as it is the endgame the problem that if solved keeps the data fresh useful and relevant and impact in the daily lives of people. I don’t have a solution for that yet but at least there is a target.

Comment from woodpeck on 8 September 2015 at 23:23

The armchair mapping mostly advocated and organized within the Missing Maps project can only go so far. “Missing Maps” could be better than nothing - but the jury is still out on that one; they could, in the long run, also be worse than nothing. It remains to be seen whether communities that have been “primed” by Missing Maps will take ownership of their maps in the same way as communities that have created their maps from square one. One way or another, we can only hope that in the end this is not “this project of ours” making maps for “the local population” (your terms), and instead everyone making their very own map!

Comment from redsteakraw on 9 September 2015 at 01:05

@woodpeck, Missing Maps isn’t just armchair mapping it is armchair mapping paired with a one the ground local mappers adding more information to the map. Furthermore tools like OpenMapKit have been developed to aid local collection of data and it relies on previous arm chair mapped objects that can have further on the ground information added to it. OSM is open and free for people to make their own tags or add the map features they care about. OSM is an international project and can be molded to the interests of the users. If you like movie thearters add them. if you like water pumps / fountains add them If you like brothels add them. NGO’s can push the map features they care about but that doesn’t come at the expense of map features locals may care about. In many times it is complementary, address data and buildings give geographic references for routing and can be used to add POIs and other information you care about. It serves as a base-map to add the layers you care about to. Like a ratchet it helps push things forward progressing the map forward. In the end locals don’t need maps for their immediate neighborhood, maps are used for and by people outside of where they immediately live to gain the knowledge of that area that locals all ready process.

Comment from pnorman on 9 September 2015 at 20:50

To see what people would map themselves, we can look at how mapping has progressed in OSM, and how mapping progresses as a new area puts itself on the map without distortions from imports or organized non-local mapping.

From this, we can see residential areas (landuse=residential) and comprehensive building coverage are some of the lowest priorities, but these are regularly the main focus of humanitarian mapping.

Comment from Jorieke V on 30 September 2015 at 20:22

Just discovered your post and discussion, interesting! Maybe the blogpost I wrote a few months ago is also interesting for you…

Login to leave a comment