[Imports] [Imports-us] Vermont, U.S. address import

[Greg Troxel]

The wiki page says

  Addresses that already exist in OSM will be removed from import data.

but this is much easier said than done.  I would expect that there is
some program, that you had to write or modify, that takes as input:

  a file from the state

  an OSM datafile

and produces a set of nodes with address tags to be added.

I can't find a description of this software or a link to download it.  I
think that should be available for others to run to evaluate this.

And, while separate from the import guidelines, it would be useful to
share the software within the community for others (not in VT) to use
for similar situations.


If you aren't using software like that, then I don't understand what you
are actually doing, and how one can have confidence that there aren't
going to be duplicated points.




Also, while I'm not really on one side or the other of this debate,
there are frequent objections to including things like
ref:vcgi:esiteid.  The usual rationale is that conflation has to happen
later regardless of where things came from, and that it's not necessary,
and therefore that it is noise.

It is now established practice that "source" does not belong on nodes,
but in changeset comments.   The planned changeset comments should be
documented.


Later it says

  Conflation will be avoided if possible. Any preexisting addresses will
  be left as-is. New addresses will be imported as standalone nodes (not
  conflated with existing building outlines)


I don't know how to read 'if possible'.  I think you are proposing to
add nodes for addresses that don't exist.  That means you will not be
modifying address tags that do exist but are partial.  That's ok, but
"we won't do this in this phase" is different from a vague "if possible"
which means "we might do some other random thing later and say it was
part of this approval/review".


I wonder about building conflation vs. not.   In MA we generally have
address points on buildings when a lot has a single large building, and
this is mostly right.   But I don't think there is any strong consensus.

When there are multiple address points for a building, it doesn't really
make sense to put them on the building if the E911 locations are
different.  But if they are the same locaition, then tagging the
building with addr:unit=1;2;3;4 might make sense.

Have you evaluated whether there are points in the database with the
same location, what you are going about that, and why?

Greg
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/imports/attachments/20220917/1d593352/attachment.sig>