[Tagging] Feature Proposal - RFC - natural=bare_rock

M∡rtin Koppenhoefer dieterdreist at gmail.com
Sat Jan 29 15:22:37 GMT 2011

2011/1/29 John Smith <deltafoxtrot256 at gmail.com>:
> On 30 January 2011 00:32, M∡rtin Koppenhoefer <dieterdreist at gmail.com> wrote:
>> can you point me to this decision? In my mapping I almost never see
> http://trac.openstreetmap.org/ticket/2873

you are pointing me to an open ticket for which there might be good
reasons _not_ to realize it in order to prove your statement "That
definition hasn't been true since use of surface=* was expanded beyond
highways for things like golf bunkers" ?

Come on, it "was" never expanded, you would like it to be expanded.

>> it is IMHO not the case that surface for landuse is a "well
>> established" feature that now would require intense changes of tags.
> I tag most beaches (that are sand surfaced) as natural=beach,
> surface=sand etc, I doubt I'm the only one.

Don't know. I don't actually care for beaches if they are tagged
surface or landcover, but I think that it would be easier for
everybody to just use one key instead of 2, and I think that landcover
is generally better suited for all kinds of values and surface is not
yet established so it wouldn't be a big "change".

>> there is golf=bunker which seems to perfectly fit the needs.
>> http://wiki.openstreetmap.org/wiki/Proposed_features/Golf_course
> How much is it actually used? Is all bunkers made of sand? Of course
> rendering surface=sand as a yellowish area would be a lot easier than
> trying to render every possible use of sand.

this is not only about rendering, it is about the meaning. If you
wanted to make a map of a golf course, you would maybe want to
distinguish between casual sand and a bunker.

> This is where SteveB likes to suggest we are actually tagging for
> renderers, at least to some extent, otherwise why bother having the
> Map Features page and tagging presets other wise?

to unify the mapping, to make the data interpretable. This has in
second place to do with rendering and is not "tagging for the
renderer". Any kind of data evaluation should be possible.


More information about the Tagging mailing list