[OSM-dev] Review of proposed spanish castrade import (Avila.osm)

Matt Williams lists at milliams.com
Mon Mar 26 10:46:45 BST 2012


On 25 March 2012 09:42, Frederik Ramm <frederik at remote.org> wrote:
> Hi,
>
>   I'm forwarding this to dev because it begins to sound as if there
> might be a broader interest in this than just those reading the imports
> list.
>
> A group of people in Spain plan to import cadastral data, including land
> parcels, and have prepared a sample. There has been some discussion on the
> imports list; below, Paul Norman posts an analysis.
>
> Personally, I find the idea of importing a number of relations equal to
> the number of people living in a place (along with other issues Paul
> mentions) rather daunting, and this could have noticeable consequences for
> renderers and editors.
>
> Discussion should ideally take place in the relevant thread on the
> imports list.
>
> I have until now been a proponent of letting people do more or less whatever
> they want locally, but in this case it is claimed that the local community
> have reviewed the import and are happy with it, whereas I think that it is
> not suitable for OSM.
>
> When advised that such an import would make editing data much more
> difficult, the proponents of the import said that "then the editors need to
> be fixed".
>
> Bye
> Frederik
>
>
> -------- Original Message --------
> Subject: [Imports] Review of proposed spanish castrade import (Avila.osm)
> Date: Sat, 24 Mar 2012 23:18:35 -0700
> From: Paul Norman <penorman at mac.com>
> To: imports at openstreetmap.org
>
> As this is such a large file, I will first be listing some general concerns,
> then going through specific concerns for each object type. I will be
> referencing specific examples by either lat,lon pairs or by IDs from the
> .osm file.
>
> Just for reference the file has approximately 60k relations, 148k ways and
> 406k nodes. The database currently has about 80 relations, 2.4k ways and 31k
> nodes.
>
> Wikipedia indicates that the population of Avila is 58k.
>
> As this one file would increase the number of relations in the database by
> 5% (and the number of multipolygons by 10%), have you spoken with the
> sysadmins to see what the impact would be on the OSM servers?
> (http://wiki.openstreetmap.org/wiki/Import/Guidelines#Keep_server_resources_
> in_mind)
>
> Relations take longer to process by most data consumers than ways or nodes.
>
> If one town of 58k were to increase the number of multipolygons by 10%, I
> could see the import easily doubling the total number. What impact would
> this have on processing times for the various tools?
>
> Overlap:
>
> Frequently one multipolygon is overlapping another. An example of this is
> found at (40.6615, -4.6895). Here one MP (-1289355) covers the entire lot
> with landuse=farmyard while a smaller one (-1269673) covers the non-building
> part of the lot, overlapping with the first. The top four are tagged
> indentically.
>
> Breaking lots into too small areas:
>
> http://maps.paulnorman.ca/imports/review/avila1.png illustrates one example
> of this problem. Selected is one MP for a lot with 5 MPs inside with
> landuse=residential overlapping with the big one for the lot. Why so many?

If the data in this image
(http://maps.paulnorman.ca/imports/review/avila1.png) is indicative of
the quality of the data in the rest of the import then that doesn't
fill me with confidence. People not orthogonalising buildings which
are square in reality is one of my pet peeves and it makes the map
look so much messier -- just press 'q'! Importing a whole city (?) of
this data will just make it look messy and is unlikely to ever be
fixed.

Matt



More information about the dev mailing list