[Tagging] Layers (was Eruvs etc.)

[Kevin Kenny]

Graeme Fitzpatrick:

> So when the road doesn't track between those boundaries on the map, what
> do we do? Move the road, move the boundaries, or say, it's not worth
> worrying too much about it?
>

Do any of the inconsistent data sources have metadata that say at what map
scale they should be interpreted?  You may simply be dealing with data that
were digitized from an inappropriately small-scale map.  It happens.

Also, are both sources referenced to the same datum plane?  (Or worse, have
geometries referenced to the wrong datum crept into the sources?)

If the metadata don't give any indication of the quality of the source (and
the source isn't TIGER:  in my experience, when TIGER disagrees with
anything else about geometry, TIGER's wrong!)  then at some point I say,
"beyond my pay grade, let the respective agencies sort it out if they
actually care."

Minh Nguyễn:
> However, the lesson I draw is not that things shouldn't be connected to
each other explicitly, but rather that mappers should be mindful about how
existing data came to be. Even ignoring the issue of boundary alignment,
there are many boundaries that don't make a ton of logical sense, so we
have to be vigilant about mappers naïvely oversimplifying them.

Indeed!

In New York, I unglued the highways from the municipal boundaries because
the TIGER import (or a mechanical edit to "fix" its duplicate nodes) had
promiscuously glued the boundaries to everything nearby: not just roads and
watercourses, but driveways, building corners, utility poles, pipelines,
you name it.  They'd not been glued by any deliberate decision of a mapper.
If a mapper makes the deliberate decision to reglue them because the metes
and bounds are defined in terms of them, that's an improvement.

Given the mess that I found, it appeared to me that having the boundaries
unglued was better than having everything fused together willy-nilly.  It
was definitely an improvement for many municipal boundaries that were
obviously following the edge of rights-of-way rather than the road center
line;  when I see two village lines that appear to be precisely 22 yards (4
rods) apart with a roadway in between, that's a dead giveaway that both
villages would rather have the township go to the trouble and expense of
maintaining the road.  TIGER had a lot of these boundaries fused with the
centerline.

I'm pretty sure that Preemption Road, which appears to follow multiple town
lines, is supposed to be aligned with the historic Preemption Line, and so
perhaps the boundaries ought to be fused, but I think that in doubtful
cases, it's better to leave them separate.  I've seen a lot of "County Line
Road" and "Town Line Road" that _mostly_ follow a boundary but occasionally
stray from it.

Then again, New York has a handful of county and township lines that
someone drew on a map in the 18th century and nobody has successfully
surveyed.  Nor does anyone much care where the line is through places like
the St. Regis Swamp or the Adirondack High Peaks - so nobody is likely to
survey and monument it. When it becomes an issue, the courts sort it out.
Gore Mountain is named because it was located in a surveyor's gore - in
neither the township to the east nor the one to the west. Nobody cared
until Barton found garnets there.  The municipalities were in court for
years over who had the right to tax the mines.

 Vermont and New Hampshire disagreed about the state line from 1764 to
1933.

Beyond my pay grade, indeed.

-- 
73 de ke9tv/2, Kevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20220907/091f1a06/attachment.htm>