[OSM-talk] mountain_pass=yes: combination with place=locality and rendering

Ulf Lamping ulf.lamping at googlemail.com
Wed Jul 28 21:22:09 BST 2010


Am 28.07.2010 17:24, schrieb Steve Bennett:
> On Wed, Jul 28, 2010 at 7:42 PM, Niccolo Rigacci<osm at rigacci.org>  wrote:
>> For bikers - at least in Italy Alps and Appennini mountains - a
>> mountain pass along a road is often used as a meeting point or as
>> a trip destination. So the use of place=locality is not so
>> misleading.
>
> Yep, I was thinking something similar. Sometimes a pass is a genuine
> location/township (eg, Charlotte's Pass ski resort). Sometimes it's
> just a mountain pass of no renown.

As a motorbike rider that visited ~400 passes, I know what you both mean 
- there are passes that are much more "important" in your trip planning 
than others.

But for me place=locality is not about the importance or renown of 
something. It's about a place that has a name but no significant other 
feature - which is not the case here.

But even more important: Even if you use that tag, it is so generic, 
that it provides no real information what you actually want to say to a 
map renderer or other data user.

BTW: The usage of place=locality together with mountain_pass=yes is 
currently very arbitrary and doesn't reflect the importance of some 
passes I saw very well ;-)

> Sometimes the situation is much
> muddier, like Arthur's Pass in New Zealand, which is both a mountain
> pass (of significant renown), a nearby township, and a national park.

So, there are three things to map, but no real problem here :-)

> Regarding ULFL's suggestions:
>> a) We might want to improve rendering of mountain_pass=yes to appear in lower zoomlevels (e.g. z14 on both default mapnik / osmarender maps)
>
> But we don't necessarily want every mountain pass to appear equally
> prominently.

Agreed. It doesn't make sense to render all passes at the same zoomlevel.

>> b) We might want to discourage the use of place=locality to "enforce" rendering of otherwise "well known tags" that has "low render priority"
>
> Yes, if there was a better way for the tagger to communicate that this
> mountain pass is particularly worthy of viewing at lower zoom levels
>
> So,
> c) continue the importance=* debate, and suggest people use
> importance=* rather than locality=* for some instances (and continue
> to use locality=* where appropriate)

Now the funny part of the discussion.

Wouldn't it be an idea to simply render a pass "importance" based on the 
type of highway it is connected to? On my personal map, I'm displaying 
all passes that are connected to a highway=track with tracktype=grade1 
(and all "higher") - works well for planning my street motorcycle trips.

Otherwise, a single importance tag will be important for which audience? 
A mountain_pass on an international hiking route might be important for 
hikers ("highest pass on this hiking trip"), but not even accessible by 
bike. So marking this one important is pretty misleading for someone not 
planning a hiking trip.

If we want to distinguish the importance for specific groups we would 
then need a hiking_importance, bicycle_importance, 
motorcycle_importance, ...

Regards, ULFL




More information about the talk mailing list