<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hi all,<br><br>thanks for the input.<br>Please don't mix up the routing part and the address search.<br>I don't want to discuss here if you are allowed to get to that point,<br>but yes, I'd prefer that address search returns a point on a public road<br>if that point is close. <br><br>We recently found out that we can use address search in Garmin software<br>without even storing any routing information in the map, because all<br>address information is stored in the NET sub file, while routing requires the NOD<br>sub file.<br>The big limit in the Garmin format is that address data is stored with roads.<br>In other words, the result of Garmins address search is a point next to a road,<br>and can only be next to a road, this also means that it requires quite a lot <br>of bytes to store a single road for each address which would allow <br>very precise address search (presuming that the OSM data is precise<br>and not itself an approximation like the addr:interpolation ways).<br><br>I think I have made up my mind now how to handle the special cases.<br><br>I think mkgmap should only use the closest road if that is much closer than the road <br>with the matching name. <br>I have to find out how to calculate what "much closer" is ;-)<br><br>Gerd<br><br><br><div>From: gdt@ir.bbn.com<br>To: marko.makela@iki.fi<br>Date: Thu, 30 Apr 2015 09:57:04 -0400<br>CC: mkgmap-dev@lists.mkgmap.org.uk<br>Subject: Re: [mkgmap-dev] RFC: naming unnamed roads<br><br><pre> <br>Marko Mäkelä <marko.makela@iki.fi> writes:<br> <br>> The general assumption would seem to be that the street names attached<br>> to house addresses belong to roads that are reachable by car, or that<br>> each residence is reachable by car. Maybe in some rare case there is<br>> some access restriction on the road associated with the address, such<br>> as access=destination. There could be named cycleways or footways<br>> between the road and the address node, but no named public roads with<br>> a different name, unless there is an error in the map data.<br> <br>That's an interesting point. In the US, around me, there really aren't<br>such assumptions. Instead, a lot (area of land that can be bought and<br>sold as a unit) has an address, generally taken from a public or private<br>way that borders the lot. Some lots don't really have addresses that<br>are useful, if they aren't near roads. Then a building on the lot,<br>certainly if there's only one, inherits the address of the lot. ANd if<br>there's going to be a building, then the lot needs to have a proper<br>address (for emergency services purposes) and one will get assigned. So<br>it definitely tends to work out 99.99% of the time that a building's<br>address is near the named road, and that one drives to that road to<br>access the building, but it's not strictly by design. In confusing<br>cases new addresses tend to get made up, usually by granting the (new)<br>access road a proper legal name, so it that sense what you said<br>describes how we do it. That's a long way of saying that it's messy<br>and that general rules don't always hold (in mass.us; not saying that<br>applies to .fi).<br> <br>This is quite separate from access. There are addresses on military<br>cases, and very often in residential complexes with gates that you need<br>a code/etc. to get through. So it's not just access=destination but<br>access=private, and yet they are real addresses on named roads inside<br>the gate.<br></pre><br>_______________________________________________
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev</div> </div></body>
</html>