[Tagging] coastline v. water

David Groom reviews at pacific-rim.net
Mon Nov 23 14:10:54 UTC 2020

See comments below:

------ Original Message ------
From: "Eric H. Christensen via Tagging" <tagging at openstreetmap.org>
To: "tagging at openstreetmap.org" <tagging at openstreetmap.org>
Cc: "Eric H. Christensen" <eric at aehe.us>
Sent: 18/11/2020 20:19:51
Subject: [Tagging] coastline v. water

>After a few days of much work, a recent collaborative project to turn the Chesapeake Bay from a nothing space outlined by natural=coastline to what we considered to be a more accurate relation of natural=water, we've received some negative feedback.
>The difference of opinion seems to lie in the definition of what we're mapping.  The use of coastline is for "seas"[0] while the use of water is for "inland areas of water"[1].  Even though the Chesapeake Bay is tidal, there is no question that it is an inland waterway (it is completely surrounded by land except for the mouth at its southeast side).
Using this logic the Mediterranean Sea, the Red Sea, and the Persian 
Gulf should all have the coastline tags removed from their defining ways 
and converted to water areas!   Italy, Greece, Libya, Egypt and a large 
group of other counties would find they had no coastline, which might 
come as a surprise to anyone lining there.

>The idea of using coastlines for basically creating an edge between the land and the nothingness of the ocean makes sense when, as far as the eye can see it's only water.
>Now, some of the feedback that has been presented[2] is that because it is tidal it is part of the sea.  I have pointed out that many rivers and streams (and ditches!) are tidal; does that make them part of the sea?  I would not think so.  In fact, there are named seas on this planet that are not even connected to other water formations (the tiniest, according to the National Geographic, is the Sea of Marmara which has an area just less than 12,950 sq km, larger than the Chesapeake Bay).
>But, tagging the Chesapeake Bay, and its tributaries, as "water" brings several benefits to the map and the users.  First, it helps identify the sections of water that exist in these areas (this can't really be done with node points as there is no way to define start and end points of an area).  There are many defined bays, rivers, and streams that make up the greater Chesapeake Bay area.  What one may see as one large mass of water is actually many smaller defined segments each with their own history.
This is irrelevant to the question of whether the ways should be tagged 
as natural = coastline.  You have had to create a multipolygon 
containing the ways which form the "sections of water", its perfectly 
possible to add the "name" tag to this multipolygon without removing the 
coastline tag from the ways

>  Second, we can speed up any updates (fixes) to outlines of the polygons that happen in these water areas without having to wait for the entire Earth's coastlines to be re-rendered.
Changes to tagging should not be done to facilitate easier rendering on 
one particular map.

>  I suspect having less coastline to render would also speed up the rendering of coastlines as well?
Very unlikely.

>I would like for the tagging community to clarify the different between "water" and "coastline" and when to use each.  The definition on water seems to say to use it on inland water but there seems to be, at least, and open interpretation of the word "sea" for coastline that is dragging many inland waters into that category.
>Eric "Sparks" Christensen
>[0] https://wiki.openstreetmap.org/wiki/Tag:natural%3Dcoastline
>[1] https://wiki.openstreetmap.org/wiki/Tag:natural%3Dwater
>[2] https://www.openstreetmap.org/changeset/94093155#map=10/37.1620/-76.1581
>Tagging mailing list
>Tagging at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20201123/58ce1fa2/attachment.htm>

More information about the Tagging mailing list