During my undergraduate years, I have learnt about Open Street Map as a prominent example of volunteered geographic information. However, there is always great inertia adding new data into the map and there are two main reasons: First, I am afraid that my input would be inaccurate, which might compromise data quality. Second, many areas in Singapore have already been quite well-mapped - including my neighbourhood. Given that I do not have updated local knowledge of other parts of the world, my concern is that satellite imagery alone may not help me to map areas well.

My concerns were finally allayed after attending my first workshop as part of the Grab GeoStars 2.0 programme. This programme features a series of workshops and trainings conducted by Grab’s very own Geo team. Following which, the majority of the programme is student-led, but guided and supervised by Grab and the partner University. Through the programme, participants will get the opportunity to go on OpenStreetMap (OSM) mapping workshops for disaster risk areas, initiate exciting OSM mapping projects, showcase their work on Grab’s social media platforms, be a part of the global OpenStreetMap mapping community, and eventually conduct workshops to share their new-found knowledge and experiences with their peers in their communities. An exciting Mapathon event marks the end of the GeoStars programme.

Jinal conducting the workshop at the National University of Singapore

During the three-hour-long workshop, Jinal delivered a lecture about OSM and outlined the basic functions and editing tools on OSM. We had a hands-on-session mapping some areas in Singapore that were not well-mapped. It helped me to understand the overall interface of OSM and gave me the confidence to upload changesets into OSM.

The second part of the workshop was about using HOT Tasking Manager, which is a mapping tool designed and built for the Humanitarian OpenStreetMap Team’s collaborative mapping process in OpenStreetMap. The purpose of the tool is to divide up a mapping project into smaller tasks that can be completed rapidly with many people working on the same overall area. It shows which areas need to be mapped and which areas need the mapping validated. I really like the concept of this breaking each map into small grid-sizes. Therefore, instead of working on an entire map project alone, this tasking manager ensures that every task is bite-size and it gives users a sense of satisfaction upon completion. It is also collaborative in nature, so multiple users can access the map project at a single point of time and add new features. This helps to speed up the completion of each mapping assignment.

My experience using Tasking Manager

Our NUS Geography Team of Geo*Stars mappers were assigned to Project #6477, India Floods 2019 : Madhubani Roads, Bihar. We were tasked to map the road network in this area. As Jinal mentioned, it is important to understand the conventions of each task before mapping. Therefore, I spent about 1 hour looking through the tasks which have already been mapped by previous users and understood the types of road networks that were used in each changesets. I found that users mainly categorised road networks as a) Secondary Road, b) Minor Road and c) Residential Road. This information helped me to look at my own task assignment and make decisions better.

I could not tell if it is a road nework

However, I encountered some issues reading the satellite imagery on Tasking Manager. For example, the satellite imagery showed a yellow line that resembles a pipe. It does not look like other road networks on the map and it also does not connect to any existing road network. I decided that it could probably be a pipe for people to irrigate their crops across large tracts of agricultural land. Still, I was unsure about my decision and needed second advice. Therefore, before I uploaded my changeset, I tagged ‘Request Validation’. Since I am also new to mapping, I also tagged ‘New Mapper’. I learnt that this can alert other mappers and ensure that the reviewer can follow-up on this issue.

Choosing the correct type of road for each changeset

The first workshop has been particularly gainful for me. With new users (like me) signing up on OSM every day, it is probable that some users might not be familiar with general practices (e.g. tagging scheme), and this might contribute to accidental edits, and even intentional vandalism. I gained during the workshop that while there might inaccuracies in mapping, there are systems and tools in place to validate changesets (such as OSMCha). I also walked away with a deeper understanding of the social value of Open Street Map, and its role in humanitarian aid. For example, during the 2010 Haiti Earthquake, OpenStreetMap and Crisis Commons volunteers used available satellite imagery to map the roads, buildings and refugee camps of Port-au-Prince. This information has helped several organisations (such as World Bank) to think of ways to provide timely relief aid in the shortest time and distance possible. While Google Maps might provide some information about directions, it might not be the most up-to-date information, particularly in less developed countries like Haiti. Furthermore, unlike Open Street Map, Google Maps database is not open-sourced. Therefore, organisations or users would not be able to independently run API queries and analyse the resulting data for decision-making. This makes Open Street Map a useful alternative for crisis management and humanitarian work.

I am looking forward to learning more!