2013/3/14 Steve Doerr <span dir="ltr"><<a href="mailto:doerr.stephen@gmail.com" target="_blank">doerr.stephen@gmail.com</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<div>I suppose the main downside is that it
requires a relation. I've not mapped give-way relationships
myself, but it would be good to map them, and the node method
seems simpler and would involve less database bloat than adding a
relation at, basically, every junction. I would think the node
method would be sufficient for most junctions, while the relation
method could be available for any more complex cases. As far as I
can see, with the node method, the important thing to remember is
that the give-way node needs to be closer to the intersection node
to which it applies than to any other intersection node on the
way, which doesn't seem too difficult to achieve. It should
perhaps be made clear in the wiki that there is not necessarily an
actual Give Way sign: it can be used to represent a give-way line
as well.<span class="HOEnZb"><font color="#888888"><br>
</font></span></div></div></blockquote><div><br></div><div>I see your point, and I've tagged a few highway=give_way and highway=stop nodes myself. However, since we are already mapping turn restrictions as relations, I think it wouldn't be so absurd to map give-way's and stops that way too. Granted, there are much more stops than the turn restrictions that need to be described explicitly. <br>
</div><div><br></div><div>I think the two ways may coexist: the node method being easier on the mappers, and the relation being easier on the consumers. </div><div><br></div><div>Regards,</div><div><br></div><div>Simone</div>