<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="">
<br>
</span>No tags on the shared nodes - just shared nodes.<br>
<span class=""><br></span></blockquote><div><br></div><div>What is IMHO a quite bad idea for two reasons:<br>– It’s unlikely that there will be software supporting features when there is no tag.<br></div><div>– You would introduce a concurrent solution to a node highway=traffic_signals. I do not think that it’s a good idea to have various ways to tag the same thing.<br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="">
</span><br>
<br>
Made a test to show you what I was thinking. <a href="https://www.openstreetmap.org/#map=18/36.32478/139.10396" target="_blank">https://www.openstreetmap.org/#map=18/36.32478/139.10396</a><br>
<span class=""></span><br></blockquote></div><br></div><div class="gmail_extra">And there, you see even more problems:<br></div><div class="gmail_extra"><br>– At <a href="https://www.openstreetmap.org/#map=19/36.32487/139.10370">https://www.openstreetmap.org/#map=19/36.32487/139.10370</a>, you do not have a shared node between the highway and the area. But this would be necessary to have a reliably hint for routing/turn-to-turn navigation software, otherwise it will be hard to know there the area ends. This would make a working routing solution quite unlikely.<br></div><div class="gmail_extra"><br>– At <a href="https://www.openstreetmap.org/#map=19/36.32492/139.10357">https://www.openstreetmap.org/#map=19/36.32492/139.10357</a> you have the area nearly in parallel to the footway. There will be other situations, where it will be exactly parallel. This is not comfortable to edit.<br></div><div class="gmail_extra"><br>– At <a href="https://www.openstreetmap.org/#map=19/36.32492/139.10347">https://www.openstreetmap.org/#map=19/36.32492/139.10347</a> you do not have a shared node between the footway and the area. But footways are not oneway. So a routing engine does not know when you enter the area respectively when you leave it.<br></div><div class="gmail_extra"><br>– At <a href="https://www.openstreetmap.org/#map=19/36.32440/139.10395">https://www.openstreetmap.org/#map=19/36.32440/139.10395</a> the footway overlaps only slightly the area. There will be cases where it will not overlap at all. How to decide reliably for software if this footway passes through the junction or not?<br><br>– At <a href="https://www.openstreetmap.org/#map=19/36.32440/139.10395">https://www.openstreetmap.org/#map=19/36.32440/139.10395</a> you have a shared node. But probably, when you are passing through the footway and go from south to est, you do not pass over the traffic signals. (You would to so only when you go from south to northwest, and the traffic signal node should be at the intersection between the footway and the highway: <a href="https://www.openstreetmap.org/#map=19/36.32445/139.10392">https://www.openstreetmap.org/#map=19/36.32445/139.10392</a> )<br><br></div><div class="gmail_extra">– The complicate roule when to share node and when not will in practice be prone to errors. It’s to difficult.<br><br></div><div class="gmail_extra">– And: I still not see what you gain with this.<br><br></div><div class="gmail_extra">– And overall: It would mean that you may not add any of these areas to OSM unless you know _exactly_ where the individual traffic signals are located. So, in practice, either the tagging will hardly be used, or (what I think is more likely) people will tag nevertheless the area, and just not comply with the rule of the shared nodes.<br><br></div><div class="gmail_extra">– All in all, I do neither see this practicable nor do I see a gain.<br> </div></div>