OpenStreetMap
I am writing this because i am not sure who would be best to talk to. I map heavily in latin America and while looking around recently using the ITO maps I realized that a lot of edits had been made in a short amount of time in Guatemala and Honduras. (http://product.itoworld.com/map/129?lon=-88.44987&lat=15.72091&zoom=7)
It seemed like an import which I thought was unusual in Central America as there is usually not that much data to rely upon. Upon Closer look I realized that it was actually a huge mapping project with HOT OSM Like this one (http://tasks.hotosm.org/project/2461#).
But when I looked at the data I realized that the data is incredibly rough, incomplete with often times incorrect geometry and attributes. This area here (http://www.openstreetmap.org/#map=12/15.6948/-90.3553) and here (http://www.openstreetmap.org/#map=12/15.8144/-85.4953) help to illustrate the problem. Random sections of road are marked as high priority. Upon closer examination you will see that often times the geometry of these roads is also extremely rough, with many roads missing and the ones that are mapped seem to be have done at random.
On HOT osm the areas are almost completely marked finished with no reviews. as can be seen her on project 9, 10, 11, 12, 13.
I am wondering what can be done to rectify this situation, and what can be done to prevent this in the future. It is very confusing to have this much low quality data being imputed into the map in mass like this. And to have this much low quality data dumped into the map lowers the quality of the data overall.
What can we do to fix this?
Discussion
Comment from russdeffner on 7 February 2017 at 19:19
Hi Sunfishtommy,
I’m the project manager for the HOT Malaria Elimination mapping. We are working on the validation and clean-up of the mapping. We have a lot of student mappers so hopefully their work will improve as we progress. Thanks for the patience and feel free to jump in to help!
=Russ
Russell.Deffner@hotosm.org
Humanitarian OpenStreetMap Team
Comment from pnorman on 7 February 2017 at 19:26
As a first step, I always recommend starting a changeset discussion. If the users do not reply, two options to consider are reverting their changesets or contacting the Data Working Group. In the case of a user ignoring discussions or messages about their mapping the DWG can require a user to respond to the discussion before continuing mapping.
Comment from Adityo on 8 February 2017 at 02:54
Hi Sunfishtommy,
I think a lot of building have incorrect geometry because in those tasking manager, we using aerial imagery from tasking manager align with bing imagery. So maybe you will see some building come out of nowhere where you compared with bing imagery.
We still validating those tasks up until now and thank you for your concern about mapping error in Guatemala and Honduras. We will try to validate more better to bring good data in guatemala and honduras.
Cheers,
adityo.dwijananto@hotosm.org
Comment from BushmanK on 9 February 2017 at 20:13
Does it really make sense to allow completely untrained people to do so much before performing any validation? It seems like fixing so many mistakes will take way more time that it took to add that kind of data. Doesn’t look like an effective way of managing human resources and teaching people.
Fixing a lot of own mistakes you could have avoided if given a proper feedback earlier is super-frustrating.
Comment from Sunfishtommy on 9 February 2017 at 21:17
I agree with BushmanK I am super excited to see Hot OSM introducing so many new mappers, but I think more feedback should be given before they go and map such large areas. Fixing the kind of mistakes I am seeing is tedious and time consuming. Roads that are not connected at intersections, roads that don’t exist, and the most obvious one, roads that have the wrong attributes and priority.