Manual conflation
Conflating data, even manually, is a very challenging process and such edits should only be carried out by those with experience and understanding of the way the OpenStreetMap community creates maps, and only with careful planning and consultation. See Import/Guidelines, and specifically consult with your local community and/or imports@ mailing list before proceeding with any conflation tasks. |
This page is intended to share tips for manually conflating external data with OpenStreetMap. Please add your suggestions.
Replacing roads
This should happen rarely, but in the United States there is a great deal of road data that is poorly aligned, out of date, or has other characteristics that make it a candidate for replacing with more recent, higher quality data. This is particularly true in parts of the US where TIGER2005 data was imported (in 2007/2008), but which has been edited very little or not at all. In this case, it may be worthwhile to replace large sections of road data with more recent TIGER (currently TIGER2011) data and/or data from the state or local government that may have more details such as speed limits, route numbers, address data, height/weight restrictions, and other information.
JOSM is the best tool to use for this task, but it is still very challenging and should only be done by experienced users. You must first convert your Shapefile or other external data source to OSM XML files; be sure to consult the imports@ mailing list during this process before uploading changes to the OSM database. Only replace data that has licensing problems which can't be readily repaired, or that hasn't been edited by any human user since import. The latter is found in JOSM by searching for:
type:way ((version:1 user:DaveHansenTiger) | (version:2 user:balrog-kun))
The following are a collection of tips:
- Open the newly-converted data in JOSM, then download a relevant portion from the live OSM database in another layer, naming this WIP (work-in-progress) or something similar.
- Change the inactive layer color to something that provides a stark contrast, such as a bright red or green. This will make seeing missing or badly aligned roads easier.
- If there are large sections of untouched TIGER data that only touch major (usually edited) roads at a few points, such as residential subdivisions, they are candidates for a particularly speedy form of replacement. Unglue the intersections (G shortcut), then use "All connected ways" (Ctrl+Shift+E) from utilsplugin2 to select the subdivision, then delete after checking that no other users have edited (using above search expression). Use the same method to select the corresponding subdivision from the converted data, and copy over (Ctrl-C, switch layer, Ctrl-V) or merge (Ctrl-M) to the OSM layer. Go to each intersection with the OSM data and merge the nodes (select both nodes and use M).
- An alternative way to remove the old data in the WIP layer is
- Draw a rectangle around the area of interest with the selection tool. This will select any ways that are wholly-contained within the rectangle.
- Search for the expression above, but click the Find in selection radio button in the search dialog to tell it to only find ways that you've already selected.
- Delete the selected ways.
- Any remaining ways in the area will need to be handled manually. Run the search again (check Replace Selection) to highlight those that are untouched. You can then break those roads at an appropriate place near the borders of your target area (select or create the point and press P) and delete the road segments within.
See also
External links
- Efficiently merging government park boundaries in JOSM - OSM diary entry
- Remapping using TIGER 2011 - blogpost from Toby Murray