[Talk-ca] Définition de zones fluviale / cotières vs chenaux de navigation ou baies

[Martin Chalifoux]

David, I would think when there is no consensus and approval the sensible thing to do is …. not to do it. 

Why you didn’t simply add all the bays and straights as POI's as a starter to get that info on the map, and leave the relations stuff for later and if you were eventually able to get consensus ?

I recently tried to make some beaches mapped better on the west coast (they were all mapped entirely under-water [to the right of the coastline]) and editing the mess of ways and relations forming the coast (bay, strait, coastline, beach, wood…) was not pleasant. I still see no use to this added complexity.

> On Mar 6, 2022, at 13:48, David E. Nelson via Talk-ca <talk-ca at openstreetmap.org> wrote:
> 
> Again, my stated goal for this was to give every single named body of seawater in Canada an area definition.  And for the vast majority of these, before I started on this, they originally had no presence at all in the OSM database, not even as points.  Defining bodies of seawater as just points may be simpler and easier on the database, but there is a tradeoff.  A lot of detail is lost in using point definitions, like size, shape and relations with adjacent bodies of seawater.  And I also found, as a consequence of the CanVec imports, bodies of seawater that had multiple point definitions, going against the guideline of "one feature, one OSM element".  As well in the course of doing this task, I ended up being able to reduce the number of members in quite a few island relations, such as Victoria Island, which went down from roughly 950 member ways to only just over 200.  There may have been no consensus on talk-ca for me to proceed with this, but there was also no consensus against me doing this, meaning I had to proceed entirely in the absence of a broad opinion either way.
> 
> Pierre, I will proceed with your request to restore the Middle St. Lawrence Estuary to the way you want it, with no huge swaths cut out of it.  But please understand, it is not my intent to go out of my way to inconvenience other editors with this overarching task.
> 
> - David E. Nelson