[Aust-NZ] NZ Geo Placenames

Francis Markham fmarkham at gmail.com
Thu Apr 7 18:12:32 PDT 2011


Is "linz geonames" the same as the NZ Gazetteer?  They have data available
at:
http://www.linz.govt.nz/placenames/find-names/nz-gazetteer-official-names/index.aspx

Cheers,

-F

On 6 April 2011 13:07, Peter <webwiz at pl.net> wrote:

> Some might remember my last attempt to produce a towns list for NZ.
>
> Im about ready to start a mark 2 version based off of the LINZ placenames
> shapefile layer.
>
> http://www.ew.govt.nz/Environmental-information/REDI/881334/
> Best outlines the issues with this layer, namely that the coords are not
> accurate.
>
> Ive made a preliminary review and comparison of the accuracy of the
> following datasets.
>
> geonames
> linz geonames
> opengps
> google
> osm
>
> See here: http://www.zoneblue.org/files/coord-comparison.pdf
>
> And conclude that most are poor and none really accurate. My proposal as it
> stands then is to take a subset of linz GEOGRAPHIC_NAMES namely DESC_CODEs:
>
> METR (4)
> TOWN(51)
> POPL  (261)
> SBRB (570)
> LOC (3446)
>
> and geocode them manually using some combination of the TOPO50 rasters,
> google satmap, and zenbu POIs to determine the best present geographic
> center of each place. Im hoping/assuming that by using these sources as a
> visual references only, that an accurate and liberally licensed dataset can
> be released, without recourse to google geocoder which i used previously.
>
> Im expecting the first 800 odd to be easy enough, then harder after that as
> the places become more amorphous. We shall probably consider that a bonus if
> the last two groups happen anywhere near 100%.
>
> As far as attributes go, the plan is to include all the present LINZ
> attributes, plus District and Regional council, and updated coordinates.
> Seeing as i will literally be looking at each one, some sort of area figure
> might be doable by drawing a rough polygon on each place and/or in
> conjunction with LINZ res_area_poly, or oliver and co's urban areas layer.
> Ive tried without success a few times to get QGIS to open the large StatsNZ
> datasets, and QGIS crashes, but from the online version, their boundarys are
> not always urban/rural. hence, that useful?. However if it could be tapped
> into population would be the prize.
>
> If anyone has any thoughts before i begin let me know. Im still half
> expecting a viable source to materialise but, unbelievably, after all these
> years, it hasnt.
>
> However its clear that better data does exist, as evidenced for example by
> zenbu. Google geolocation incorrectly places the small town of winchestor,
> canterbury by about 1km to the east. Zenbu places it pretty much right.
>
>
> Regards
>
> peter
>
>
> _______________________________________________
> Aust-NZ mailing list
> Aust-NZ at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/aust-nz
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/oceania/attachments/20110408/a0857f1e/attachment.html>


More information about the Oceania mailing list