[OSM-newbies] Church, or not?

[Randy]

Isaac Wingfield wrote:

>
>On Nov 8, 2009, at 4:00 AM, 
>newbies-request at openstreetmap.org wrote:
>
>>On Sat, Nov 7, 2009 at 5:24 PM, Isaac Wingfield 
>><isw at witzend.com>
>>wrote:
>>>It's becoming clear to me that some of the entries on my local part
>>>of
>>>OSM are based on old, out of date information, and I've been fixing
>>>these as I become aware of them. Some of the entries are easy -- a
>>>"Golf Club" marker on what is now clearly an industrial site -- but
>>>one I'm unsure about is when a location has a "place_of_worship"
>>>designation, but on-site inspection reveals that there is no obvious
>>>church there, or even a sign designating one.
>>>
>>>I've tried web searching on the names (wondering if the icons somehow
>>>got displaced), but that's been very frustrating. It would appear
>>>that
>>>a lot of "information sites", such as lists of churches offered by
>>>religious organizations, take their data from the same obsolete
>>>information sources that OSM used; they all point precisely to the
>>>place where the OSM "church" icon is, where there is nothing
>>>resembling a church (I even found the same thing regarding the "golf
>>>club" I mentioned above, on a site offering information to traveling
>>>golfers).
>>>
>>>So here's the question: It's easy enough to tell a golf club from an
>>>industrial site just by walking past it, but a "place_of_worship" is
>>>perhaps not so clear; could one by located in a private home, with no
>>>external markers? If I find nothing but a house at a location where
>>>OSM has a place_of_worship icon, can I just delete it?
>>
>>I've recently seen some non-traditional places of worship; regular
>>meetings in hotel conference rooms and what would otherwise be retail
>>shops.
>>
>>I'd fall back to the OSM "observable / verifiable" guidelines.  If you
>>can't point to the sign that proves it on the ground, perhaps it
>>doesn't belong in OSM.
>>
>>Have you contacted the user that added the items to get their
>>perspective?
>
>I don't know how to find out who that was, but if I highlight the
>icons the data seems to have gotten there as part of an original
>"bulk" import from TIGER or something similar; the gnis:create date is
>1999.
>
>Isaac

GNIS indicates that it is from the US Bureau of Geographic Names.
http://geonames.usgs.gov/  Their data, apparently, is notorious for being 
incorrectly located, something I wouldn't expect from the folks at USGS. 
And, if it was 1999 data, then it could easily be obsolete.

-- 
Randy