<div class="gmail_quote">On Wed, Jul 28, 2010 at 11:01 AM, McGuire, Matthew <span dir="ltr"><<a href="mailto:Matt.McGuire@metc.state.mn.us">Matt.McGuire@metc.state.mn.us</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="im">> There's no observation that will tell you whether a road is primary or<br>
> secondary.<br>
<br>
</div>Agreed. There is no observation that will tell you whether a road is more important than another road that is not where you are. But you can identify physical characteristics. A lot of these observations will lead to a coherent whole.<br>
<div class="im"><br clear="all"></div></blockquote></div><br>Yep that is the problem. And all the discussion and many wiki pages are about trying to come up with a consistent guideline for turning those observations into consistant and useful OSM tags.<br>
There are two extremes of OSM that are fighting each other because lack of consensus and organization. We say both tag as you want, and follow what has been agreed on in the wiki (which is constantly changing).<br><br>Existing DOT data is also an observation to add to the mix in deciding how to tag something. Don't dismiss it just because there are exceptions to the way someone else tags.<br>
<br>Ramblings and grandiose goals:<br>The goal of describing what is on the ground is great, but to what end. The tags will be plentiful but perhaps useless if not designed for specific uses! The primary use will be visual maps. Any secondary usage would also need to be considered. Layout a tagging plan for these uses and tag for that, but make sure it is still flexible for future changes.<br>
<br>Don't force using a tag because it displays nicely on a map that way, but do use a tag because it is the common (thought out and planned) tagging scheme that the render's happen to implement. Render's should be the visual representation of what the wiki describes.<br>
<br>-- <br>Dale Puch<br>