[Talk-us-massachusetts] Mapping stone walls

[Greg Troxel]

Marc Sevigny <marc.sevigny at gmail.com> writes:

> Has anyone thought about mapping stone walls in Massachusetts?  I've helped
> to add to New Hampshire's efforts to map stone walls there.  They have over
> 3,800 miles of stone walls mapped now.  See
> https://www.arcgis.com/apps/webappviewer/index.html?id=ec12d9d5583c4709be446a77115bf0a2

That demands a signin.  Is that open data?  I am curious about who that
group is and what they are doing (am guessing not OSM).

> I think it would be useful to have a similar effort in Massachusetts.
> These historical walls have value as heritage objects that deserve
> protection.  The first step to protection is to know where they are.  We
> could also then track their loses due to development or theft.

I have been adding some, and I think Joe in Acton has also.    Nothing
fancy, just linear features with barrier=wall, wall=dry_stone, IIRC.

I have added some from imagery, where especially if you have been on the
ground and made a few notes, if the leaf-off imagery is particularly
good, you can see them.  However, I added a wall that doesn't exist in
one conservation area; after one hike I made a note to check it, and the
next time it was clear there was no wall.  There were some big rocks
that appeared natural sort of in a line.

I have also added them from GPS measurements.   Joe and I went to one
place to understand the boundary (the deeds and reality say one thing,
and MassGIS L3 parcels is wrong).   As part of this we stopped at the
ends of many walls and I took averaged waypoints.
  https://osm.org/go/ZfIcib~yB--
Here, the E-W property line  is from L3 parcels, and the walls are from
our GPS observations.   While its almost certainly true that the wall
is the property line in this case, I don't have any reason to believe
that the GPS measurements are more accurate the the L3 parcels data.
Mapping stone walls this way is very labor intensive.

I am very curious what you are doing, and how.  It would probably be
good to have a writeup of how and why someplace, probably on the OSM
wiki, perhaps in a mass-specific page, like the MA conservation lands
mapping description.  Or perhaps a US-specific page as I suspect that
how to do this is not state specific.

> Also, what do people think about mapping cellar holes?  I searched for tags
> that would be suitable for mapping cellar holes but didn't find any.  I
> love in an historic district and mapping the cellar holes adds to the
> historical context of the district.  Any ideas as to how to add cellar
> holes?

I'm not sure and you could wade into the tagging list.  It seems one way
is building=cellar_hole which is a bit abusive but probably renders :-)
Then there are various ruins-type tags.  I don't remember who did this,
and I don't claim it's right.   For many of them not so celebrated and
marked, it's surely a bit much.
  https://osm.org/go/ZfIflkQwb

There is also the separation of "how to tag" and "how people should
render what I tagged".  If you start mapping things that aren't already
mapped a lot, you are quickly into "the default render doesn't show my
tags", and you set up your own render that does what you want.  And then
after having some experience with it perhaps submit a change request for
the default render.   It probably is sensible for a cellarhole to get
mapped as a thin closed line, sort of like a building without the fill.
Or perhaps a different shade of fill, or as the trend seems to be "CH"
in the middle.  You also might look at prior art in USGS topos or other
maps in terms of what cartographic norms are.