[Talk-us] parcel data in OSM

[Ian Dees]

On Sun, Dec 30, 2012 at 8:50 PM, Jeff Meyer <jeff at gwhat.org> wrote:

> What's frustrating about distinguishing between useful & not useful data
> imports is that there isn't much information available on the wiki & other
> documentation about how to distinguish between the two.
>
> At least, I haven't been able to find much of the good information that's
> in the minds of the old-timers and the comments of this thread on the wiki
> or other documentation.
>

I think we've relied on the OSM principle of "map what others can verify on
the ground" to define what is acceptable to import. In general, most data
is acceptable to import when properly discussed and reviewed, but in this
particular case we all got crazy because "parcel data" apparently means
many different things to different people and (at least in my view) it's
almost universally not something that belonged in OSM.


> So... people can take the time to come up with what they think is a good
> idea, to go looking for guidance, to follow what guidelines are available,
> to prep data, to find what seem to be relevant & comparable past imports,
> and then be told in email that what they've been working on is
> unacceptable.
>

The seventh step of the imports checklist is to discuss your plans with the
mailing list. This is really quite early in the process and if it puts
people off from contributing to OSM then I don't think we really want their
import to happen anyway.


> I think this problem is fixable by transferring this knowledge onto the
> web pages, which is what some of us are trying to do with the wiki.
>

I hope that the working group Serge is spearheading will generate a
document (preferably not on the wiki) describing how to move forward with
imports in our community.


> In the mean time, I'm saying it can be frustrating to follow every rule
> you can find and then to be told you're still doing something incorrectly.
>

As I said before, if the community's concerns about your import put you off
from contributing it to OSM, then you should seek help to improve it or
continue doing local surveying instead of being completely put-off from OSM
as a whole.


> FWIW, I'm also sympathetic to the frustrations of people who end up
> cleaning up after the imports that do go awry.
>

The "old-timers" are loud when it comes to imports because OSM has had
*very few* imports that work and we've had to deal with them all. Whether
its my crappy county lines import (abutting polygons need to be broken
apart and turned into relations), extraneous data (useless source data
leaking in to OSM tags), over-noding (over-digitized but not necessarily
accurate source data), improper licensing, improper interaction with the
OSM API, or any number of other problems that arise, it's very hard to pull
off.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20121230/3f034bde/attachment.html>