[Talk-us-massachusetts] more on Lynn Woods

[Tom Parent]

I am guilty of in the past re-using cadastre ways in a separate
landcover relationship.  My logic at the time was: I know the "woods"
extend past the "public lands" boundary BUT I'm only interested in
landcover mapping within the park, etc.  I now feel this is/was motivated
reasoning and basically just tagging for the renderer and therefore wrong.
I personally no longer landcover map "the woods" as I feel there are much
more important things to spend mapping time on.  But, it's a valid OSM
pursuit so not gonna begrudge anyone that does want to do that...as long as
it doesn't affect the validity of other things (public lands cadastre being
most important to me.)

I do feel OSM has some glaring technical deficiencies that make it way too
easy to screw things up...especially for casual users not using JOSM.  OSM
could greatly benefit from "layer" functionality.  IMHO landcover tags
should be a completely separate layer where one is unable to make
relationships with other ways (roads, cadastre, etc).  The following
should all be separate layers: land cover, structures, boundaries,
hydrography, transportation, other infrastructure (power lines, etc)





On Wed, Jul 27, 2022 at 8:54 AM Greg Troxel <gdt at lexort.com> wrote:

>
> I think you should not be using landuse objects as part of a landcover
> relation, and if you avoid that you will have vastly less trouble.  I
> can see your point that they happen to line up precisely in this place,
> right now, but in general they don't, and they don't reliably move
> together over time.   Anybody adjusting one thing will change the
> other.  Future validators that complain about landuse/landcover on same
> objects will object.
>
> Are you really using the entire way that is the reserveration boundary
> as outer for natural=wood, even though there are significant parts where
> that just isn't so?  That seems wrong.  I suspect you are running into
> odd cases and bugs, and my advice is don't do that.
>
> If you want to reuse specific parts of the boundary in the outer, when
> it is really the case that the woods line and the boundary match (I
> still think this should not happen), then you could
>
>   - create a relation for the outer boundary, with one component the
>     current closed way
>   - same thing on golfcourse
>   - split the ways where the sameness changes
>   - add the way pieces to the woods outline relation, so the closed way
>     thus represented has woods to the inside and !woods to the outside
>     all all along it
>
>   - figure out if
>
>       relation: tag woods
>         outer: relation, no tags
>            members: a bunch of ways, individually not closed, together
>            closed
>         inner:
>             relations which have way segments as members
>             regular ways e.g. pond, swamp
>
>     really works in the render toolchains.  After concluding it doesn't,
>     just draw ways, perhaps reusing nodes for the landcover and don't
>     try to reuse ways.  You'll be done before you would have figured out
>     the relation stuff.  While you can argue that it ought to work, OSM
>     has grown organically rather than with formal semantics and when you
>     are too far outside existing practice things break.   You have
>     arrived, I think :-)
> _______________________________________________
> Talk-us-massachusetts mailing list
> Talk-us-massachusetts at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us-massachusetts
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us-massachusetts/attachments/20220727/01464021/attachment.htm>