Hello norge09,

if this mapping is based on NPI CC-BY data please stop using this and revert your changes - CC-BY data is not compatible with the ODbL by default, you need explicit permission.

Even if permission is obtained please discuss use of this data with the community first. Much of the NPI mapping is quite outdated and therefore less accurate than newer satellite image based mapping in OSM (which has partly already been performed on Svalbard).

Hallo Micha,
löblich, dass Du die übergroßen Landnutzungs-Polygone hier ein wenig zerlegst. Bei schmalen Wegen, welche keine Lücke im Bewuchs bilden wie hier:

osm.org/way/35825921

sollte jedoch auch keine Lücke in der Erfassung des Waldes sein, der Weg befindet sich ja im Wald und die Baumkronen bilden ein geschlossenes Dach.

Bei den breiten Forststraßen mit grade2 sieht das natürlich anders aus.

Hello jptolosa87,

there is a newer version of the baseline declaration of Greenland on page 126 of

http://www.un.org/Depts/los/doalos_publications/LOSBulletins/bulletinpdf/bulletin56e.pdf

Hello Brian,

this is not really an appropriate use of natural=glacier. See the wiki:

osm.wiki/Tag:natural%3Dglacier

Hallo Alpinfuchs,

ich denke natural=beach ist hier falsch, beach (Strand) ist eine durch die Wellen geformte Struktur an der Küste, eine Sandbank ist kein Strand, das vorherige natural=shoal ist hier deutlich passender.

Und kommentiere bitte Deine changesets damit man erkennen kann, was Du jeweils gemacht hast.

Hello bebbi,

you here removed the previous glacier mapping (together with changeset 34702512) and replaced it with an approximate mapping of the snow extent in Bing images - as it was already done in changeset 28662999.

My previous mapping was based on comparing various literature sources and photos in addition to the aerial imagery. Could you please restore that - unless you have more accurate information on the actual glacier ice extent.

Hello,

this and many of your other recent changesets contain edits that are factually incorrect and do not comply with OSMs tagging conventions. Please try to be more careful verifying what you map and use meaningful changeset comments indicating the reasons for your changes.

Hello ulil,

once more your edits left the coastline unconnected, both here and in

osm.org/changeset/34862721

Please read osm.wiki/Tag:natural%3Dcoastline and make sure you understand the formal requirements on coastline mapping as well as its meaning - the coastline has to be continuous and it is not OK to map straits and tidal channels near the coast as natural=water and draw the coastline merely as an outer hull.

If you cannot edit the coastline without breaking it you should refrain from modifying it. If you have something that needs changing and you do not feel confident to be able to properly change it ask for help on mailing lists or forum.

ulil removed the coastline tags here and left the coastline unconnected. See:

osm.org/changeset/34822310

Geometry looks unchanged so you should be able to add it back without problems.

Hello ulil,

this and some of your other recent changes broke the coastline on a fairly large scope:

http://tools.geofabrik.de/osmi/?view=coastline&lon=-15.37967&lat=11.84141&zoom=10

Please look into fixing that and try to be more careful with future edits.

Hello fx99,

good to see you are trying to improve mapping in northern Canada. The Canvec data conversion you seem to be using as a basis here is deeply flawed though, in particular for the hydrography. For example what you tag as waterway=riverbank/natural=water + intermittent=yes here:

osm.org/relation/1656894
osm.org/way/369713173

is not an intermittent river but a river floodplain which in the dry climate of northern Canada is - if at all - only partially water covered during snowmelt. For reference here an up-to-date image of the area of the second sample:

http://www.imagico.de/files/LC80532472015223LGN00_exp.jpg

In addition tagging all waterways as waterway=stream independent of size is not a good idea.

And for the ice shelf areas you should follow the tagging conventions from the Antarctic (osm.wiki/Antarctica/Tagging) which were also used elsewhere in the Arctic - creating large multipolygons like this

osm.org/relation/5489293

including both ice shelves and glaciers is generally not a good idea.

On a general note the Canvec data in this region is also usually very old, in most parts even older than Bing images. Importing this without cross checking with up-to-date data is highly questionable. And in any case please use a separate import account.

Hello JFK73,

you specified RGI as source here - please note that the legal suitability for use in OSM is doubtful - there is no clear license for the RGI (in contrast to GLIMS which is PD) and some of the contributing works use doubtful sources like 'Google Earth'. It is certainly fine to use RGI to identify missing glaciers but i'd be careful with using it as a basis for mapping.

Für ausgesprochene Trockenrasen-Gebiete ist natural=grassland eigentlich sehr passend, ggf. könnte man das auch noch durch ein grassland=xeric ergänzen. Der ein oder andere Strauch tut dem meiner Meinung nach keinen Abbruch.

Im Allgemeinen werden die aber zumindest sporadisch schon geschnitten - sonst überwuchern sie nämlich nach spätestens 5-10 Jahren schon. Gelegentlich werden solche Flächen auch abgebrannt.

Ich würd wahrscheinlich natural=grassland vorschlagen, wenn dort regelmäßig (also mindestens jählich) gemäht wird, auch wenn dort teils auch höhere Stauden wachsen. Falls holzige Vegetation vorhanden ist, zum Beispiel auch Brombeeren und anderes rankendes Zeug, wäre natural=scrub hingegen durchaus passend, selbst wenn es relativ niedrig ist. natural=heath ist eigentlich nur für natürlicherweise sehr niedrig wachsende Gehölze, was am Kaiserstuhl kaum vorkommt.

Hallo q_un_go,
schön, dass Du den südlichen Kaiserstuhl verbesserst, natural=heath ist hier allerdings nicht ganz passend, denn bei regelmäßig gemähten Böschungen dominiert hier im Allgemeinen die krautige Vegetation und wo nicht gemäht wird gibt es schnell auch höhere Sträucher.

Yes, when you want to import additional data please include this on the wiki and open this for further discussion on the imports ML. Those things indicated for import on the wiki seem mostly solid but those imported beyond that have various issues. Getting input from the international community will help you avoid mistakes that have previously been made elsewhere.

Hello Stephen,
the most accepted taging for moraines is geological=moraine, see osm.wiki/Key:geological

natural=landform is primarily used by a Canadian import, is not well defined and should not be used.

You might also otherwise want to re-evaluate the linz2osm tagging for antarctic features - see your other changeset using man_made=windmill instead of power=generator

If it is correct to tag this as coastline is of course a question of how exactly the definition of natural=coastline is interpreted.

Arguments against this are:

* It would be the only artificial channel tagged this way - the Corinth Canal for example (which has no locks either) is mapped differently.
* With this mapping there is no easy way to determine from the data that this is an artificial channel and not a natural strait. With a polygon tagged natural=water + water=canal this would be easier.
* While there is in principle free water flow along the channel there is actually no significant water exchange across the entire length - it is way too long and too shallow for that. In other words the whole situation would not be much different if there were locks.

Please do not newly tag inland water areas as coastline as mentioned on:

osm.wiki/Tag:natural%3Dcoastline

You are greatly affecting anyone who uses the coastline data.

city is not appropriate here, an argument can be made for village but by common standards this is a hamlet (<200 permanent inhabitants).