Re: [OSM-dev] [josm-dev] Problem with large changeset

2010-07-06 Thread Andy Allan
On Tue, Jul 6, 2010 at 5:56 AM, Alan Mintz alan_mintz+...@earthlink.net wrote: If you retrieve a node with http://www.openstreetmap.org/api/0.6/node/54530540 , you get a longitude with 7 decimal places. If, as JOSM does, you retrieve objects by bounding box with

Re: [OSM-dev] [josm-dev] Problem with large changeset

2010-07-06 Thread John Smith
On 6 July 2010 18:54, Andy Allan gravityst...@gmail.com wrote: suggests* that double results are printed to (a maximum of) 9 significant figures, which is fine for latitude and fine if all your testing longitudes cover, say, UK and Germany. You do realise, regardless of the latitude, anything

Re: [OSM-dev] [josm-dev] Problem with large changeset

2010-07-06 Thread Andy Allan
On Tue, Jul 6, 2010 at 10:10 AM, John Smith deltafoxtrot...@gmail.com wrote: On 6 July 2010 18:54, Andy Allan gravityst...@gmail.com wrote: suggests* that double results are printed to (a maximum of) 9 significant figures, which is fine for latitude and fine if all your testing longitudes

Re: [OSM-dev] [josm-dev] Problem with large changeset

2010-07-06 Thread John Smith
On 6 July 2010 19:16, Andy Allan gravityst...@gmail.com wrote: whether 7 decimal places is too many (most times I've heard it discussed some people have asked for more) is something entirely different. At this point it's almost pointless to have more than 6dp of accuracy since we're not likely

Re: [OSM-dev] [josm-dev] Problem with large changeset

2010-07-05 Thread Alan Mintz
At 2010-07-05 19:35, Alan Mintz wrote: At 2010-07-05 17:34, Alan Mintz wrote: ... One of the nodes involved is id 54530540. When I bring up the conflict resolution dialog for this node, it says 33.8099363, -117.4006172 My version coordinates 33.8099363, -117.4006170 Their version coordinates