SteveC <steve <at> asklater.com> writes: >Speaking personally about what large orgs and what they want, I think it's >pretty simple. Have a look at commercial data and OSM and do a diff, what are >the main things missing? Addressing for geocoding and turn restrictions for >routing.
For addressing, I guess it is usually sufficient to have a street name - the exact addr:housenumber stuff is not needed I assume? OSM already has plenty of tools for 'noname' hunting but it is harder to track down streets which are missing from the map altogether. In the United Kingdom we are fortunate to have the OS Locator database to check against, although even that is by no means complete. Without such a secondary source, noticing that a particular obscure side street is not on the map has to wait until a dedicated OSM contributor happens to walk past it. In a seemingly-complete area, that could take a while. But I think that name searches (from a website or a geocoding API) can help with the task of finding missing addresses. With access to a big dump of all name searches done from the Bing website (suitably anonymized) it would be possible to feed them all through Nominatim and see likely candidates for missing streets. Of course this procedure can produce only hints for resurveying, it can't add the streets to the map or even provide proof that they exist, but for getting the last 2% of missing addresses in an area we have to take any help we can get. Are you and the other Microsoft people able to generate any bulk data of this kind? Turn restrictions are also hard to survey manually. A mapper on foot or bicycle might not pay much attention to them, and again, it is hard to know when you have all of them. They might possibly be suggested from analysis of GPS traces, provided we have a large number of traces for an area and they are clearly tagged to show which ones are for travelling by car. This is one reason why a standard tagging scheme for GPS traces is needed. -- Ed Avis <e...@waniasset.com> _______________________________________________ talk mailing list talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk
