[Talk-ca] Woods along Canvec Borders

[Sam Vekemans]

Hi Daniel
Cool thanks, this also helps

This could also explain for some of 'holes' that were seeing on CommonMap
http://api.development.i386.commonmap.org/?lat=49.7172&lon=-121.122&zoom=12&layers=B0000FTF
<http://api.development.i386.commonmap.org/?lat=49.7172&lon=-121.122&zoom=12&layers=B0000FTF>as
it could be the actual data it's self, taken from a different year over year
along nts tiles
This data does origionate from GeoBase
http://www.geobase.ca/geobase/en/data/landcover/csc2000v/description.html

So for the new version of CanVec, will the map features it be directly
derrived from this LandCover data?
If so, we (Primarly just Brendan) has done the osm-tag matching
https://github.com/morb-au/CommonMap-importers---experimental/blob/master/schema/ca.lc/ca.lc.schema.pl

If needed, I can convert this into a 'proper' spreadsheet showing each of
the osm-tags.

I'm currently going through all of these tags again, and matching it up with
the existing canvec features, (as well as Natural Earth, GeoBase NHN and
GeoBase NRN) for the Formal 'CommonMap Feature catelogue'

Let me know if their is interest in this, and i can speed up the process :-)

Cheers,
Sam

On Thu, Mar 24, 2011 at 7:50 AM, Bégin, Daniel <
Daniel.Begin at rncan-nrcan.gc.ca> wrote:

> Bonjour Adam,
> You have just learn what is an edgematch problem!
>
> This problem of mismatched lines/polygons at the edge of NTS mapsheets is
> usually caused by a difference in imagery/photography acquisition dates.
>
> Each year, we were contracting out imagery/photography acquisition over
> areas convering many NTS mapsheets. The problem arise at the edge of two
> areas when many years passed between each contract.
>
> Things are changing over time out there !-)
>
> Daniel
>
>
> -----Original Message-----
> From: Adam Dunn [mailto:dunnadam at gmail.com]
> Sent: March 18, 2011 14:19
> To: Bégin, Daniel
> Subject: Woods along Canvec Borders
>
> When importing Canvec data within a single NTS tile (092H05.x.x), the
> border between sub-tiles lines up and matches very well. It's a simple task
> to merge nodes along the border and join ways/areas if necessary.
>
> Doing a border between NTS tiles (092Hx) doesn't tend to work out as well.
> Sometimes it seems to be okay, but sometimes it looks like there's a serious
> mismatch.
>
> In attached screenshot, you can see how natural=wood appears to mismatch
> from 092H05.3.3 (left of screen) to 092H06.0.0 (right). I don't think it's a
> bug in how JOSM drawing fills polygons either.
>
> Both of those tiles are Canvec 7, downloaded within the past month.
>
> Ideas?
>
> Adam
>
> _______________________________________________
> Talk-ca mailing list
> Talk-ca at openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-ca
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-ca/attachments/20110324/7cead096/attachment.html>