sent from a phone

> On 10. Nov 2018, at 07:01, دار الآثار للنشر والتوزيع-صنعاء Dar Alathar-Yemen 
> <hubais...@outlook.com> wrote:
> 
> I am talking about nodes not areas or ways. Most of seas in OSM are tagged on 
> node element. Also countries most of them are tagged on node element


these country nodes come from the old days when we hadn’t administrative 
boundaries. Today we have all or maybe almost all countries as detailed 
boundaries.


It is true the seas are mostly mapped as nodes although there are some parts 
mapped as polygons, e.g. https://www.openstreetmap.org/relation/7156290

From a data point of view, adding an angle to a node to specify a label 
orientation will likely be rejected as tagging meta information that is not 
objectively observable, but one could say the tag is about the principal 
orientation of the represented waterbody and it would be much harder to 
question it ;-)

As a node doesn’t have orientation nor extent, the information could be 
valuable.

From a practical point of view, for rendering, a line for label placing seems a 
better solution as it gives indications about scale, orientation but also shape 
(e.g. curved). There are tools around that can calculate the line from areas. 
Drawing them yourself would also be an option, there are not so many of these 
if you are only interested in seas and oceans. It might be interesting to see 
alternatives to osm for those, e.g. Natural Earth has global simplified water 
areas with names.

Cheers, Martin 
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
  • ... Kevin Kenny
    • ... دار الآثار للنشر والتوزيع-صنعاء Dar Alathar-Yemen
      • ... OSMDoudou
        • ... دار الآثار للنشر والتوزيع-صنعاء Dar Alathar-Yemen
          • ... Kevin Kenny
          • ... Mateusz Konieczny
            • ... Dave Swarthout
      • ... Mateusz Konieczny
        • ... دار الآثار للنشر والتوزيع-صنعاء Dar Alathar-Yemen
          • ... Marc Gemis
      • ... Martin Koppenhoefer

Reply via email to