Playing devil's advocate here.
But how many objects are there?
By their very nature I image they are concentrated in certain areas, such as
parks?
Would it be as easy to simply use the data as survey points, that way you would
capture not only the barbecue object, but also associated picnic tables, bins
and paths to route to them?
Phil (trigpoint)
On 11 October 2018 11:48:15 BST, Mike N <[email protected]> wrote:
>On 10/10/2018 3:37 PM, Kevin Kenny wrote:
>> In the couple of imports that I curate, I leave external ID's in OSM,
>to
>> give me a quick search for the object that I previously imported. (I
>can
>> then check that the user that last modified the object was the
>import.)
>> If the object has been changed in the external database, and is
>> unmodified since the last import, I can skip a manual conflation
>step,
>> and simply apply any changes and present the modified object for
>review.
>>
>> The ID's would be useless in a 'one time only' import, but I attempt
>to
>> keep the import up to date as new versions of the external data
>appear.
>
> As the OSM community grows in the region containing a curated
>dataset, use cases must be managed:
> OSM user adds a new object they observe.
> OSM user adds a duplicate object without first checking for an
>existing object.
> OSM user removes an object they observe is no longer there.
>
> It all requires complete conflation, just as if there had been no
>external IDs in the database.
>
>_______________________________________________
>Imports mailing list
>[email protected]
>https://lists.openstreetmap.org/listinfo/imports
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
_______________________________________________
Imports mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/imports
