<div dir="ltr"><div>Hi Christoph,</div><div><br></div><div><div class="gmail_extra"><div class="gmail_quote">2018-07-03 10:25 GMT+02:00 Christoph Hormann <span dir="ltr"><<a href="mailto:osm@imagico.de" target="_blank">osm@imagico.de</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">* inventing new tags so far not used and documented is fine - but you <br>
should document them.<br>
* adding new uses to secondary tags (like using surface=* or usage=* on <br>
features it is so far not commonly used on) is also fine if it matches <br>
previous use in meaning.<br>
* adding new uses to existing primary tags is highly sensitive and <br>
should usually be discussed first. Creating a new tag is almost always <br>
a better idea.<br></blockquote><div> </div><div>I merely agree with all those points except there is no primary tags but only tags.</div><div>What is primary for a given mapper will be secondary for someone else, despite so called primary tags are more used than any others.</div><div>A tag may be primary for a specific render but secondary for routing. Then, saying a tag is primary is tagging for render or routing or any particular purpose.<br></div><div><br></div><div>Given example is waterways in tunnel</div><div>If I look for waterways, then waterway=* will be the main tag I'll look for</div><div>If I look for tunnels, then I'll look for tunnel=* and don't even look at waterway=*</div><div><br></div><div>Then it should be really simpler to extend values on existing tags when it's relevant (which is the only criteria)<br></div><div>Defining primary and secondary tags just put pointless barriers on refinement process and I don't see any benefit.</div><div><br></div><div>All the best</div><div><br></div><div>François<br></div></div><br></div></div></div>