Downloading the OpenStreetMap Node Database

I apologize in advance if this question doesn’t make sense or if I use any unclear wording or vernacular; I’m very new to OpenStreetMap and GIS in general.

To briefly describe my purpose: I’m conducting research on the routes that internally displaced persons take as they travel. I have a database that contains the names of their origin towns/villages and then the names of all of the other points of interest they spend time at afterward. I would like to be able to associate a coordinate with each point of interest, and then trace the overall trajectories each person takes on to one map.

So, I’m wondering if there exists a database of points of interest that can be downloaded separately from any section of map. I understand how to use OSM to download a section of map as an XML file, which I can then load into JOSM with any number of GPX files. By using the OSM Routing Service tool, www.yournavigation.org, compiled and described by Lambertus in a post from years ago, I’d like to produce GPX files for each trajectory. Though I can do this by hand, I would of course prefer to automate this process, and so having direct access to a database of coordinates for each point of interest would be a big help.

I hope that wasn’t too unclear. Thanks so much for any help you can provide.

Hi,

Would it be enough to download only nodes for example with XAPI? http://wiki.openstreetmap.org/wiki/Xapi#Nodes
However, I must warn you that in OSM POIs are not always points, because of principle “One physical feature, one object in OSM”. Therefore lots of POI like features are stored into OSM as tags in the building polygons. So actually it might be better to make a list of all the tags which suit for the features you filter data by those. Converting polygons into points afterwards is not a big deal, you can do it with centroid function in PostGIS, Spatialite etc.

Thanks so much for your reply, JRA. I’m sorry I took so long to respond; other parts of the project needed some work and this was put on the back burner.

Your solution makes a lot of sense, and I do think it will work out. It does seem like I should be able to filter out duplicate nodes, so that I have only one database entry for each POI, and I’m still trying to work that all out. I’ll post again if I’m actually successful, or if there are any other issues.

Thanks again for your help.