openstreetmap
Warn when attempting to edit imported address nodes
Maybe just like one cannot tamper with locked store brand names (7-Eleven etc.), have a warning
This node contains an
addr:*:datasetfield, meaning it was from a mass address import. Any edits might get overwritten in the future.
this sounds like a problem with the import methodology. In other countries the regular import & conflation process respects the improvements that other mappers have made, instead of blindly overwriting them
Well fine, need very fancy conflation algorithms, vs. independent official government address nodes without anything glued to them (yes, by well meaning users), just like they exist in government GIS's.
https://wiki.openstreetmap.org/wiki/Import/New_Zealand_Street_Addresses_(2021) says
To add the ref:linz:address_id=* tag to 40,000 addresses where the tag was not added during the initial import, or where it was removed when address points were manually merged with buildings
E.g., the government address node is now the corner of some building which would become triangular if deleted. Yup, can't even adjust the node's position without distorting the building. So need fancy algorithms to work around mom and pop editors gluing it to things. Can't blame them. Nothing warned them.
The address was imported from LINZ, but since then LINZ has deleted the address. However, an OSM mapper has merged the address node into a building. We won't delete the building, but we will remove the address tags from it.
Fancy algorithms needed!
The address exists twice in OSM: once on an imported node, and once on a building. We will merge the node into the building. This is usually caused by StreetComplete users.
More fancy programming. Yes, somebody already wrote the fancy program.
Anyway, in my ideal world, mappers would already notice the address node, and think "good, one less detail I need to type in", and leave it alone, just like those survey marks on the sidewalk that say "do not disturb, else fine: $5000".
Now let's sit back and think about the difference between a drinking fountain and an address:
- The drinking fountain: just hitch it to the wall and map it.
- An address on the other hand comes from the government. If the public were in charge they would just use their favorite lucky number. Then they would hide it, keeps debt collectors away. And who knows what they would do to it on the map.
Now let's look at that https://wiki.openstreetmap.org/wiki/Import/New_Zealand_Street_Addresses_(2021)#Process_Diagram
That kind of software is very hard to maintain.
With the Dan Plan® (that's me, Dan) all that fuss is eliminated.
The massive diagram just becomes
- Delete old government address nodes.
- Import fresh government address nodes.
Yes you could say that is rather wasteful, of node serial numbers, that's all! What if the node is 100% the same? Just comparing it to be sure isn't worth the fuss... (Except perhaps making some changes file larger.)
Such simplicity is enabled by keeping the darn public's grubby hands off of the nodes.
Four out of five times they will thank you because they weren't really sure if they should glue something to such a node or not.
Now let's think of lamp and power poles. Would the utility company be happy that you have attached even your hammock to one of them? You get the idea.
Hey, I was just filling out the address fields for grandpa's store (way or node).
Well, there should be a warning there too, that please don't bother filling these in if there are government address nodes nearby!
But it isn't 100% clear that his store's address is #4. OSM has fields for the phone number and address. So what should I fill in?
Er, well complain to the county government that the node is ambiguously placed.
It is 100% to edit imported address nodes, if import breaks or overrides edits then such import is broken
This is not a good idea, I'm afraid. If imported data should not be edited by mappers, it should not have been imported in the first place. [^1] See also #9255.
[^1]: not that data consumers would be able to replace osm's addresses with the "official" dataset in post-processing.
