<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>I foundation very recent boundaries but GML format. <br><br>Verstuurd vanaf mijn iPhone</div><div><br>Op 14-jun.-2013 om 15:44 heeft Glenn Plas <<a href="mailto:glenn@byte-consult.be">glenn@byte-consult.be</a>> het volgende geschreven:<br>
<br></div><blockquote type="cite"><div>
<meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
<br>
<blockquote cite="mid:201306141414.59477.benlaenen@gmail.com" type="cite">
<blockquote type="cite">
<pre>I was just looking that information up in the forum.
As far as I can find out now, a boundary is a way with a tag boundary=*
</pre>
</blockquote>
<pre>Yeah, I'm making some shortcuts: someone has to map the boundaries with
boundary=administrative + admin_level=9 for the deelgemeenten. Then the proper
boundaries have to added to the relation that defines the full area of
Werchter.
As you can see on
<a class="moz-txt-link-freetext" href="http://nominatim.openstreetmap.org/details.php?place_id=39355280">http://nominatim.openstreetmap.org/details.php?place_id=39355280</a>
nominatim doesn't have any knowledge about it belonging to Werchter. It knows
it's inside the Rotselaar boundaries. It can look at nearby place nodes, but
then can't do more than guessing if it does belong to it or not, and decides
on that by measuring the distance to that node. In the case of Leybos, it
relates it to Heikant, so searching for "Leybos, Heikant" does give you the
results.
If you look at another street more closely to the place node of Werchter, like
<a class="moz-txt-link-freetext" href="http://nominatim.openstreetmap.org/details.php?place_id=5990195272">http://nominatim.openstreetmap.org/details.php?place_id=5990195272</a>
you see that it does relate the street to Werchter, so "Hazepad, Werchter"
returns the expected results.
So what has to be done to make Leybos belong to Werchter: we have to map it as
being part of it, which we do by adding administrative boundary relations.
Which is something we're going to need a source for first, since they
generally don't mark these boundaries with big white painted lines in real
life... So if anyone has them, share them :-)
Ben
</pre>
</blockquote>
<br>
I think that will be the solution, all I can find dates back to
1600.<br>
<br>
<meta http-equiv="content-type" content="text/html;
charset=ISO-8859-1">
<a href="http://search.arch.be/eadsearch/inventaris/index/eadid/BE-A0510_004732_005707_DUT/inventarisnr/I47325707170/level/file">http://search.arch.be/eadsearch/inventaris/index/eadid/BE-A0510_004732_005707_DUT/inventarisnr/I47325707170/level/file</a><br>
<br>
It's an index , so no real map.<br>
<br>
Glenn<br>
</div></blockquote><blockquote type="cite"><div><span>_______________________________________________</span><br><span>Talk-be mailing list</span><br><span><a href="mailto:Talk-be@openstreetmap.org">Talk-be@openstreetmap.org</a></span><br>
<span><a href="http://lists.openstreetmap.org/listinfo/talk-be">http://lists.openstreetmap.org/listinfo/talk-be</a></span><br></div></blockquote></body></html>