OpenStreetMap

In the light of two very interesting blog posts, one of Erica Hagen from Ground Truth and one of [Gwilym Eades] (http://place-memes.blogspot.de/2015/05/the-hubris-of-proactive-disaster-mapping.html) lecturer geography in the Royal Holloway University of London, in the aftermath of the Nepal earthquake; I would like to share with you some of my thoughts.

Types of mapping activities

First of all I would like to share with you a table which I created and showed last year during GeOng in France, at the Missing Maps launch in London and at the HOT summit in Washington. Every time this table lead to interesting discussions afterwards.

The table shows the existence of four different ways of OpenStreetMap-mapping:

four different ways of mapping

The first way of mapping is remote mapping without any local knowledge. Remote mappers will be able to map geometrics of buildings, roads and land use with the help of aerial imagery. This, but nothing more. A next way of OpenStreetMap-mapping is field mapping by ‘outsiders’. ‘Outsiders’, like for example tourists, development workers or humanitarians going outside to collect geographical data. This is ground truthing, but in a limited way. As outsider you can after all only collect the best known points of interest or visible things like for example the condition of roads. Field mapping done by locals is much more interesting. Locals know their environment! So they move faster, they know for example how the address system is working, they can easily ask street names in the local language, they know where to go for information on the borders of neighbourhoods, etc. But imagine what can happen when it comes to a real local OpenStreetMap-community. A maintained database of local knowledge will appear on the OpenStreetMap website and a permanent base of mappers will be present and keeps the map up to date. Isn’t it this, which is in particular interesting in prevention and preparation of crises?

Looking through the eyes of a humanitarian

Of course humanitarian organisations do what they want to do, that’s the freedom and openness the OpenStreetMap platform offers. And yes if I look through the eyes of a humanitarian organisation, I sometimes would take this freedom to map and take control of the situation as soon as possible. Because do we want to let Ebola spread in West Africa if we can help stop it? Or what do we do when a typhoon is coming towards Vietnam?

In this way you indeed can see the mapping we are doing with HOT out of a very colonial viewpoint: we connect maps with power and control again. Although me myself I’m convinced we should intervene and map from a distance in some cases, I’m also convinced we should not stop there… We should always connect crisis mapping with building and supporting local OpenStreetMap communities, and even better: if we want to do good, humanitarian mapping we should help develop and support them even before a crisis starts!

Can everybody map?

But who are the mappers of a local OpenStreetMap community exactly? Is this community a collection of the more technology-oriented people of a country? Are they youngsters out of one particular neighbourhood? Are they volunteers who are getting a per diem to collect data around the town? …

In OpenStreetMap there is one thing we cannot ignore; it will always be a project that depends on some level of technology. By this I don’t mean you need a degree in IT, but just the fact that for uploading data you need to have basic computer skills and the opportunity to use a computer with internet connection. Where this is not a problem in Europe, in a lot of other parts in the world it is. How should people in the heart of Africa discover OpenStreetMap by themselves when there is no phone network, an internet connection of 384Kbit/s costs more than an avarage monthly salary and when even the idea of mapping is strange for a lot of people.

But does this mean we only have to support and build OpenStreetMap communities with people who can handle a computer and can pay an internet connection? I think we shouldn’t: every person in the world can contribute to OpenStreetMap, with his or her very local knowledge (places and people) even if he or she never touched a computer before and doesn’t speak a word of English.

In January I was for example two weeks in Bangladesh to map Hazaribagh and Kamrangirchar, two neigbourhoods in the capital Dhaka. Besides our more technical mapping heroes, we worked together with some local people who never had anything to do with mapping before, but who saw the advantage for their own neighbourhood. Mister Babul, Robin, Sharmeen and our Kam-boys; they learned about satellites in the sky, how to use Field Papers a smartphone and most of them edited in the end OpenStreetMap by themselves. Some of them are now even teaching others to map.

Same same, but different

A mapper in Bangladesh, in Mali or in Kenya is in fact not that different from a regular OpenStreetMap-mapper in Europe or somewhere else in the world. The mapper with whom he differs the most, is maybe even the humanitarian mapper… Why is somebody in Europe putting all the cycle ways on OpenStreetMap? Because he and his fellow bikers can make use of it! Why is a geography student in Cameroon mapping roads in Cameroon? Because in this way he doesn’t have to use some old geographical dataset from France anymore. Why will a slum dweller start to map his living environment? Only because he can get advantages out of it: maybe he sees possibilities in finally having an address for his business, or he sees possible improvements in the sanitation situation for him and his family.

In general, if people don’t see a direct personal advantage, it will be more difficult to motivate them to map in a voluntary way. And it might even be that people see a threat in all this mapmaking: some slum dwellers of small slums maybe prefer to stay of the map, because then it will be less likely to be discovered and evicted by the government. And it are these decisions I think we definitely should respect within our Humanitarian OpenStreetMap Team. Especially with vulnerable communities, like slum communities, or people in very remote places we have to be careful. In a perfect world, we should work together with existing communities in villages, neighbourhoods … at the smallest level possible, we should explain what mapping is, what OpenStreetMap is, what the possibilities are and let them discuss, discover and decide.

Comment from Warin61 on 14 June 2015 at 23:52

You say “Remote mappers will be able to map geometrics of buildings, roads and land use with the help of aerial imagery. “

Remote mapping buildings .. I don’t usually map those. I’ll map them where they are unusual (shape) or a long way from anywhere else. But there it typically much to map, so I concentrate on the roads, large villages, towns. The rest … local knowledge should be used.

I take a cautious approach to remote mapping roads. I prefer to down grade a road classification rather than make an assumption that upgrades it … possibly beyond its’ actual class.

I find remote mapping land use is not precise. And orchard looks much like a vineyard … a group of buildings could be a village … or a collection of farm buildings … There can be a lot of assumptions made .. best to leave things off the map if uncertain?

Comment from pedrito1414 on 15 June 2015 at 09:20

Super interesting, Jorieke.

Some thoughts…

When people prefer to stay off the map, for fear of eviction for example… Do they know in what ways a map would be beneficial? What if there is repeated cholera outbreaks in that slum? What if there is a high risk of fire? How well can/do NGOs explain the positive impact that mapping might have? Should organisations in this case just map privately for these purposes and keep the data closed?

On the excellent table you have made, for me there is a column missing… This one says ‘Local OSM community in collaboration with a global, remote community’. If these two can work in tandem, with each group playing to their strengths and feeding back to one another, they would surely be greater than the sum of their parts. And, I mean this not just at the time of a major emergency but as something sustained. With Missing Maps, we are starting to seed some collaborations like this at the moment and it is a very interesting process! Nothing to report yet, but hopefully not too long…..

Great piece ;)

Comment from Jorieke V on 16 July 2015 at 16:19

Hey guys,

I’m very sorry I did not answer before, but thanks a lot for your valuable reactions!

Indeed I think people sometimes should not be put on the map openly. A professor of architecture in Bangladesh told me a few years ago: ‘Putting slums on the map is exactly the same as shouting: “Evict them!”’. And not only in the case of slums we have to be careful, also mapping in contexts of complex political emergencies might be very sensitive. What do we for example do with informal refugee camps? I think we should always have in our heart the so called ‘do no harm’ principle as highest value.

But does ‘do no harm’ mean by definition that we do not have to intervene? In some cases yes, but in other cases you can adapt the methods you are using. In the case of mapping I believe we could indeed keep the data private like you say Pete. It might sound contradictory to say this as OpenStreetMapper, but I believe that we with HOT cannot afford ourselves to act solely as cartographers or open data advocates. Even not, if we get millions of funding or huge media attention. We should have our principles, and we should act as humanitarian mappers.

Further I very much like the column ‘Local OSM community in collaboration with a global, remote community’ you would like to add into my table! I definitely agree with you in here. The feeling of being part of a worldwide network of mappers is in my experience incredibly important to keep people mapping. Also, I especially like it when there is a real exchange: not only from north to south which is most likely to happen at this moment, but an exchange from south to south, north to north and south to north. This means for example Senegalese mappers mapping for their ebola affected neighbour, Bangladeshi mappers mapping for Nepal, or a Burkinabé answering your questions on the HOT mailing list. So mappers everywhere in the world, don’t be afraid to show what you are doing, ask questions if you need help and offer help if you can help somebody out!

And to Warin, I think you must be an excellent (humanitarian) mapper! You reflect on what and how you are mapping, and this is the most important we need. So I also have no ready-made answer for you on your question if we better leave things off the map if we are uncertain… I suppose in some cases yes, and in others not.


Login to leave a comment