<div dir="auto">Ok, two things.<div dir="auto"><br></div><div dir="auto"><b>Function vs physical characteristics</b> </div><div dir="auto">First, I've increasingly realized what's probably at the heart of this 12+ years discussion, the enormous problem of interpreting highway=path|footway|cycleway (just like is currently being discussed about highway=track) in two entirely conflicting ways, as someone has mentioned, <i style="">function VS physical characteristics</i>. </div><div dir="auto"><br></div><div dir="auto">For some, it seems obvious that all highway tags, from the biggest roads down to forest tracks and wilderness paths, are supposed to be used according to their function, not at all their physical characteristics. But words like path & footway is telling a different story and confusing most mappers. </div><div dir="auto"><br></div><div dir="auto">And some say that highway=path either can mean a wilderness path or, if used with foot/bicycle=designated, a combined, urban foot- and cycleway. No, it can't, because often the latter case is tagged without access tags and therefore impossible to interpret based on the highway tag alone. </div><div dir="auto"><br></div><div dir="auto">And herein probably lies the fundamental error of</div><div dir="auto">1. using words that people normally would associate with physical characteristics, but to only mean function</div><div dir="auto">2. the default OSM rendering not considering physical characteristics (particularly for non-urban ways) together with underestimating the extent of tagging for the renderer (obviously people want their tagging to be confirmed) </div><div dir="auto"><br></div><div dir="auto">This makes these lower-end highway tags extremely ambiguous. And STILL I see people emphasize that path can have these two meanings at the same time. But if we are to make anything meaningful of the data they really can't. I can't believe this anomaly has been allowed to remain unsolved for this many years. </div><div dir="auto"><br></div><div dir="auto">1. We must decide whether Highway=path|footway|cycleway is to tell a ways function OR its physical characteristics, not both. Other tags will then have to be used (and should be strongly encouraged in the case of these lower-end highways) to denote other characteristics to make them useful for data consumers.</div><div dir="auto"><br></div><div dir="auto">Unless we clarify this, other tags will just continue to confuse the original meaning of highway=path|footway|cycleway and thus basically become troll tags. And there will always be a problem when no additional tags are being used. </div><div dir="auto"><br></div><div dir="auto">2. Once the single deciding factor behind <span style="font-family:sans-serif">highway=path|footway|cycle way has been clarified (likely function only) we can deal with ways to describe physical characteristics in a KISS way (so it's actually being used). That's what Kevin is asking for. </span></div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto"><b>Documenting the process</b></div><div dir="auto">Second thing I wanted to talk about is about the suggestion of the importance of documenting this discussion. I'm not sure what has to be documemted other than final decisions and significant opposing views? </div><div dir="auto"><br></div><div dir="auto">I thought that many people on this list would rather not be bombarded by 100+ emails about the same topic, which is why I suggested keeping the discussion among those who are interested in the actual discussion and not just its conclusions.</div><div dir="auto"><br></div><div dir="auto">Should we close the discussion in this mailing list, continue in a smaller format and then report back the concluding suggestions for confirmation before implementing? Or is there still enough interest to keep the entire discussion here? </div><div dir="auto"><br></div><div dir="auto">/Daniel </div><div dir="auto"><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Den sön 31 maj 2020 02:48Jarek Piórkowski <<a href="mailto:jarek@piorkowski.ca" rel="noreferrer noreferrer" target="_blank">jarek@piorkowski.ca</a>> skrev:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Sat, 30 May 2020 at 20:13, Tod Fitch <<a href="mailto:tod@fitchfamily.org" rel="noreferrer noreferrer noreferrer" target="_blank">tod@fitchfamily.org</a>> wrote:<br>
> I’ve spent too much time recently trying to figure out how to better determine whether the ways I am rendering should be shown as an urban/suburban walkway versus a non-urban hiking trail (intentionally not using “footway” and “path” as words for this).<br>
<br>
I realize this might not apply to your map, but just to give people<br>
discussing path/trail semantics another data point:<br>
urban ravine/hillside areas like<br>
<a href="https://www.mapillary.com/map/im/pcIl9nspFIi38uEDY5q_OA" rel="noreferrer noreferrer noreferrer noreferrer" target="_blank">https://www.mapillary.com/map/im/pcIl9nspFIi38uEDY5q_OA</a> (this one is<br>
300 m from a normal low-density neighbourhood) or<br>
<a href="https://www.mapillary.com/map/im/hHkS__YTVtqWYcg-l4kMGQ" rel="noreferrer noreferrer noreferrer noreferrer" target="_blank">https://www.mapillary.com/map/im/hHkS__YTVtqWYcg-l4kMGQ</a> (this is up to<br>
1 km walk in any direction from a "normal" urban street with a<br>
sidewalk) or <a href="https://www.mapillary.com/map/im/rjjuiRG0giyX_kufNrX_mA" rel="noreferrer noreferrer noreferrer noreferrer" target="_blank">https://www.mapillary.com/map/im/rjjuiRG0giyX_kufNrX_mA</a><br>
(300 m from a normal mid-density neighbourhood)<br>
<br>
--Jarek<br>
<br>
_______________________________________________<br>
Tagging mailing list<br>
<a href="mailto:Tagging@openstreetmap.org" rel="noreferrer noreferrer noreferrer" target="_blank">Tagging@openstreetmap.org</a><br>
<a href="https://lists.openstreetmap.org/listinfo/tagging" rel="noreferrer noreferrer noreferrer noreferrer" target="_blank">https://lists.openstreetmap.org/listinfo/tagging</a><br>
</blockquote></div>