It sounds like there are two applications here: The mobile editor itself, for which there are various applications which could be used/extended. Given that it's mobile (no/costly network available) it would presumably want to store an OSM diff file of changes made, ready for upload when back at home.
A tool for applying these diff files to the OSM database. I think this would be a GUI application rather than a simple merge tool, because the underlying data could have changed since the mobile device was last loaded with maps, the item might already exist, it might have been recently added, it might be in the OSM data but not displayed on the mobile device, the way you're trying to name might have been split and renumbered since you recorded information about it (especially if there are several people mapping an area simultaneously) And of course, some mobile devices might not know about OSM data or way-IDs at all. The program might want to store "road at this lat/long should be oneway" but if it's only displaying map images then the resulting diff file won't even know the way ID until you look it up later. So it might be useful to have a desktop application that can take an OSM diff file and let you confirm each change, showing a map of the area, highlighting/selecting which road you mean, adjust the position of a POI from away from the road-centre where were standing when you entered it, that sort of thing.. On Tue, Aug 26, 2008 at 3:32 PM, Nick Whitelegg <[EMAIL PROTECTED]> wrote: > Hello everyone, > > Was thinking of a few ideas for an OSM mobile editor which could work as > follows: > > * Ability to allow user in the field to add new attributes to a way (e.g. > road name if it's missing, or one way) or correct existing attributes. > * Allow a user to add new POIs. > * Could work by either downloading OSM data live from the server (though > this would have problem of being relatively expensive for the user) or by > the user preloading OSM data to the phone before they go out. > * Allows user to add annotations to Openstreetbugs (e.g. "missing footpath > here") > * Because the inbuilt GPS in phones is not as good as dedicated devices, > it would not be a priority to develop features to allow surveying of new > ways. However this could be built in, in preparation for inbuilt mobile > GPS improving in the future. > * Java ME based for maximum cross platform support. > > Does this seem like a good approach? As can be seen the idea is not so > much to allow addition of new ways (due to the inbuilt GPS on phones not > being great) but more to add POIs and tags to existing ways. If there's > interest and - more crucially - if I have the time (always difficult!) I > could start work on it. > > Nick > > > > _______________________________________________ > talk mailing list > [email protected] > http://lists.openstreetmap.org/listinfo/talk > _______________________________________________ talk mailing list [email protected] http://lists.openstreetmap.org/listinfo/talk
