2011/11/21 Václav Řehák <rehak...@gmail.com>: >> I think, the problem is, that Locus MAY BE used in a bad way by bulk >> downloading. > > That's not true anymore. Downloading is disabled by Locus author in > the recent versions and online browsing is limited to a certain number > of tiles per day. > >> As long as it's used like a browser it would not be a problem and the lock >> could be removed (I'm pretty sure then that would not be a problem even for >> the admins). > > Sure, but then it would be nice to hear from the admins what rules > should the app follow to be unblocked. So far the Tile usage policy > says no apps are allowed without prior permission.
This is a very contentious issue in OSM; I and others believe that OSMF needs to make its resource use more clear to outside providers. I encourage you and others to get involved in the Foundation (ie become members, get active on the lists, tell the board your feelings, vote) if you feel strongly. Right now the rules for usage are very muddy and are in large part just "As the administrators see fit"- which is in one sense reasonable, but sometimes hits innocent folks like yourself. > Maybe the way for future is writting an app that sends OSM credentials > to the server as Richard says (I don't know about such app). Or I can > add my custom providers.xml entry to Locus with some extra http header > (e.g. including my osm id) if the admins say so. But I have not heard > of such suggestions nor recommendation which app to use. The only info > I have is "do not use Locus and a few other popular apps". In my view, for users who are looking for simple navigation aids, a map that is old is usually not a problem. Most users won't notice the difference between one week and the other and for those people, the suggestions of using a commercial map provider, or having the author provide the map themselves is reasonable. I'd say even in situations where "up to the hour" maps are needed, such as in humanitarian situations, we have reasonable solutions. For real-time mapping though, the real answer is that the app should not be using tiles, but rather rendering the data locally. And then, we would benefit from apps which make updating vector areas easier. Matt Atmos and Kompaza have put work into what are essentially "data tiles", which would fit your usage needs fairly well. I spent some time on this myself, but haven't gone back. The limiting factor here is mostly developer time/effort. Right now, most mobile map app authors are entirely focused on displaying tiles. If they'd put some effort into getting data to/from their apps in more novel ways, I'm sure they'd be embraced by the OSM community. One last "OSMF ping/tweak". If this need continues to arise and mobile authors do not come up with a solution, then maybe it's time OSMF does. Kothic needs work, but I can imagine that a month or two of work by a high quality developer could result in a highly robust solution to this particular need, and therefore a robust solution to the issue of expensive map/ API calls and the issue of caching, which, if successful, could not only result in reduced need for tiles, but allow for better apps in general. - Serge _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
