On 8/3/06, <b class="gmail_sendername">Laurence Penney</b> <<a href="mailto:lorp@lorp.org">lorp@lorp.org</a>> wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Wollschaf and Mike Collinson makes some good points on separating physical<br>descriptions and administrative classes, but I agree with Etienne that the<br>physical descriptions need to be further split up. (Terminology is also
<br>problematic: 'waytype' and 'wayclass' are way too close!)<br><br>We need shorthand methods of referring to whole physical descriptions of ways,<br>and defining them would ideally be possible within apps. Neighbourhoods often
<br>have many similar streets. E.g. once you'd set up a "typical Manhattan street"<br>(width=2 trucks, parking=both, pavement=both, pavement_width=5 people), then you<br>could use that for all streets that matched the initial physical definition.
<br><br>I am in favour of human and vehicular units of measurement. We'll get fewer<br>errors that way IMO (although perhaps North Americans will use "1 car" to mean 4<br>or even 5m, considering how difficult they find driving in Europe...). Width in
<br>metres might be added automatically: (width_derived=5.3m)</blockquote><div><br>To me, width=3m implies an accuracy of +/-1m. Since our GPS surveying methods do not measure the width of anything to that kind of accuracy we could be in danger of creating data that is open to misinterpretation.
<br><br>By all means have a width tag in metres, but only use it for cases where the width has actually been measured to at least +/-1m accuracy. For other cases approximate_width=3m or subjective_width=narrow seems to me to be a better record what we are able to observe.
<br> <br>Etienne<br><br></div><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">-- Laurence<br><br>Joerg Ostertag (OSM Munich/Germany) wrote:
<br>> ...<br>><br>>> If we are seeking a purely physical description of the roads then shouldn't<br>>> there be a separate key for each aspect of the road. The following would<br>>> describe one carriageway of a motorway:
<br>>> lanes=3<br>>> width=18m<br>>> breakdown_lane=yes<br>>> oneway=yes<br>>><br>>> And this would describe a Sydney lane:<br>>> lane=1<br>>> width=3m<br>>><br>>> One problem with this is that there is no easy way to accurately measure
<br>>> the width of a road with a GPS unit. Maybe the width attribute should<br>>> allow units other than metres (1 car, 2 trucks, 1 person), or approximate<br>>> values 0-5m, 5-10m, or even comparative values: very narrow, narrow,
<br>>> normal, wide, very wide.<br>><br>> I would restrict the field to meters. Since this way we have a chance of<br>> making use of it in automatic processing.<br>> And give the user some hints to enter data:
<br>> 1m for bikes and Foot only ;-)<br>> 3m one car(narrow)<br>> 3,5m one car easily<br>> 4 one truck<br>> 7m two cars narrow<br>> ...<br>><br>> -<br>> Joerg<br>><br>> _______________________________________________
<br>> talk mailing list<br>> <a href="mailto:talk@openstreetmap.org">talk@openstreetmap.org</a><br>> <a href="http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk">http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
</a><br><br><br><br>_______________________________________________<br>talk mailing list<br><a href="mailto:talk@openstreetmap.org">talk@openstreetmap.org</a><br><a href="http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk">
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk</a><br></blockquote></div><br>