Michael Reichert
2015-03-27 20:43:29 UTC
*TL;DR* I myself (one of the OpenRailwayMap developers) absolutely
oppose a proposed railway import in the Netherlands.
Dear Jeroen, dear readers of Imports and OpenRailwayMap mailing list,
I have heard and read that you tried to import railway data in the
Netherlands.
I myself (one of the developers of OpenRailwayMap) am against importing
railway data. OpenStreetMap is not a 0-Euro data center for geodata. OSM
is a project where /people/ collect data and care for it.
Because the Netherlands already have mapped railway lines, you have to
preserve the objects' history if you import railway data. This makes an
mechanical import is nearly impossible.
That's why I suggest you to do it differently. Set up a WMS or TMS
offering some of the data. The WMS/TMS can be used at JOSM. The mappers
can use this service if they improve a railway line or station. They
correct/improve existing data and preserve the objects' history.
We have several of these "WMS imports" in Germany and Austria. Usually a
state surveying authority allowed us to use their WMS service to derive
data. Some examples are City of Berlin, City of Hamburg, State of North
Rhine-Westphalia or basemap.at in Austria. Mappers usually derive
geometry of objects from these services while adding additional data
which they surveyed in field.
This type of import would be much slower but better in the long term.
People only start mapping if they see a (nearly) blank map or missing
data. This effect has worked perfectly in Russia (where there seems
supposedly be no better map) and Germany. At the moment German railway
mappers experience the same effect at detailed railway mapping (speed
limits, signals, milestones, âŠ). Deutsche Bahn (Germany's national
railway company) refuse to publish any piece of data. I have found some
newbies, who started mapping a few months ago and did railway edits from
day 1, for the last months while mapping railway stuff. This happens if
people miss data.
If you import all data, you won't get such a large community. People
only care for the data they mapped themself ("their own things"). They
will keep these data up-to-data. If you imported these data, they
cannot/won't do it because they did not map it and did not join OSM.
Railway infrastructure is changing day by day (like highway
infrastructure does this, too). You need a large number of mappers if
your data shall be up-to-data.
My suggestion: Set up an WMS service which offers a rendering of all
tracks and milestones of ProRail (maybe different colours for different
track types).
You should think about a tagging scheme before starting signal mapping
in the Netherlands. Every country has a different railway signalling system.
Best regards
Michael
(Nakaner)
PS Sorry for starting a new thread. I joined the list after the initial
discussion about this import.
oppose a proposed railway import in the Netherlands.
Dear Jeroen, dear readers of Imports and OpenRailwayMap mailing list,
I have heard and read that you tried to import railway data in the
Netherlands.
I myself (one of the developers of OpenRailwayMap) am against importing
railway data. OpenStreetMap is not a 0-Euro data center for geodata. OSM
is a project where /people/ collect data and care for it.
Because the Netherlands already have mapped railway lines, you have to
preserve the objects' history if you import railway data. This makes an
mechanical import is nearly impossible.
That's why I suggest you to do it differently. Set up a WMS or TMS
offering some of the data. The WMS/TMS can be used at JOSM. The mappers
can use this service if they improve a railway line or station. They
correct/improve existing data and preserve the objects' history.
We have several of these "WMS imports" in Germany and Austria. Usually a
state surveying authority allowed us to use their WMS service to derive
data. Some examples are City of Berlin, City of Hamburg, State of North
Rhine-Westphalia or basemap.at in Austria. Mappers usually derive
geometry of objects from these services while adding additional data
which they surveyed in field.
This type of import would be much slower but better in the long term.
People only start mapping if they see a (nearly) blank map or missing
data. This effect has worked perfectly in Russia (where there seems
supposedly be no better map) and Germany. At the moment German railway
mappers experience the same effect at detailed railway mapping (speed
limits, signals, milestones, âŠ). Deutsche Bahn (Germany's national
railway company) refuse to publish any piece of data. I have found some
newbies, who started mapping a few months ago and did railway edits from
day 1, for the last months while mapping railway stuff. This happens if
people miss data.
If you import all data, you won't get such a large community. People
only care for the data they mapped themself ("their own things"). They
will keep these data up-to-data. If you imported these data, they
cannot/won't do it because they did not map it and did not join OSM.
Railway infrastructure is changing day by day (like highway
infrastructure does this, too). You need a large number of mappers if
your data shall be up-to-data.
My suggestion: Set up an WMS service which offers a rendering of all
tracks and milestones of ProRail (maybe different colours for different
track types).
You should think about a tagging scheme before starting signal mapping
in the Netherlands. Every country has a different railway signalling system.
Best regards
Michael
(Nakaner)
PS Sorry for starting a new thread. I joined the list after the initial
discussion about this import.
--
Per E-Mail kommuniziere ich bevorzugt GPG-verschlÃŒsselt. (Mailinglisten
ausgenommen)
I prefer GPG encryption of emails. (does not apply on mailing lists)
Per E-Mail kommuniziere ich bevorzugt GPG-verschlÃŒsselt. (Mailinglisten
ausgenommen)
I prefer GPG encryption of emails. (does not apply on mailing lists)