[Talk-ca] Question about NIDs

Richard Degelder rtdegelder at gmail.com
Tue Jan 20 01:53:40 GMT 2009

So far Steve has demonstrated, with the Fort McMurray import that we can
detect the data that is currently available and only add the new
roadways to OSM.  Michel is, using different tools, going to do the
same.  So we are going to have two methods of augmenting the current
data within OSM to incorporate GeoBase.

This is all good, we are going to need to do this to improve the map for
Canada.  And we are doing it without damaging the current contents of
OSM that was put there by the different users, one of the goals when we
started the import.  There may be some issues but from the work I have
been doing to check the data I have found very little that would make me
uncomfortable in the least.  I approve of the way Steve is doing the

But my question is about another goal of the import process.  The value
of the GeoBase data lies in both augmenting the current contents of OSM,
which we are doing with the import of roadways that are not currently
within OSM, and the ability to update OSM with either new data available
from GeoBase or from the GeoBase updates to their current data.  For
this to work we need to incorporate the GeoBase NID for every one of
their segments.  Is there a process to include the NID, in an
appropriate manner, with the import data that was derived from GeoBase?
And has there been any considerations as to how to add the NID
attributes to those ways already within OSM?

Without the inclusion of the NID, either on the currently existing OSM
data or on the data we are currently importing from GeoBase, our ability
to update the information based on those NIDs is going to be severely
curtailed.  Without NIDs the GeoBase import becomes a one time event and
will not benefit from updates nor from additional information that will
become available from GeoBase in the future.

On the other hand having the NID available for all appropriate ways
means that when we want to import the roadways for any province that
does not have the complete data set, such as road names and address
ranges, or to incorporate any new data, such as Postal Codes, we can do
so with full confidence that when that data becomes available it will be
a simple step to run a script that adds the additional attributes based
on those NIDs.  But we are going to have to have a clear process on
adding the NID as an attribute to all existing ways as well as for those
we are adding now though the GeoBase import.

It is acceptable to worry about having the NID available later while we
are working on importing the roadways from GeoBase to make the road
network complete.  But there should also be some thought as to how we
are going to go about adding those attributes later.  To add a lot of
new data, that has the current GeoBase NID readily available, without
any idea of how we are eventually going to be able to add it in late
means that we are only compounding the problem later when there is
significantly more data, due to the import, than we are currently
facing.  The current contents of OSM does not have the GeoBase NID and
so it will have to be added at some point.  Compounding the problem with
even more ways within OSM, ways that within GeoBase had an NID, only
makes the issue more difficult.

Richard Degelder

More information about the Talk-ca mailing list