Modify

Opened 2 months ago

Closed 8 weeks ago

#19589 closed defect (fixed)

bogus validator warning duplicate housenumbers

Reported by: dieterdreist Owned by: team
Priority: normal Milestone: 20.08
Component: Core validator Version: latest
Keywords: duplicate housenumber Cc:

Description

the JOSM validator warns about a "duplicate housenumber", but actually there is nothing wrong with duplicate housenumbers.
In this case it is https://www.openstreetmap.org/way/58525242 and the node here: https://www.openstreetmap.org/node/1505516457

Attachments (0)

Change History (13)

comment:2 Changed 2 months ago by dieterdreist

The address of the house also has additional information (staircase B), so it is distinct. On the other hand, the gate has the required information (street and housenumber) for the gate. I still believe you should drop this test. It is too complicated to get it right globally, and you would have to know all keys that can be used to define a "POI", plus all the possible qualifiers that make housenumbers unique. And it would still fail for cases with actual duplicate housenumbers (the latter could be seen as a feature).

comment:3 Changed 2 months ago by skyper

It does not help if a new tag is used with low numbers (~50). Hey, why is the wiki page in main name space? Would you, please, move it under proposed and only redirect from main name space, thanks.

Last edited 2 months ago by skyper (previous) (diff)

comment:4 in reply to:  2 Changed 2 months ago by Klumbumbus

Milestone: 20.08

comment:5 in reply to:  2 ; Changed 2 months ago by dieterdreist

Replying to dieterdreist:

The address of the house also has additional information (staircase B), so it is distinct. On the other hand, the gate has the required information (street and housenumber) for the gate. I still believe you should drop this test. It is too complicated to get it right globally, and you would have to know all keys that can be used to define a "POI", plus all the possible qualifiers that make housenumbers unique. And it would still fail for cases with actual duplicate housenumbers (the latter could be seen as a feature).

FWIW, this is not a new tag and I did not add it to the object in question (I hardly ever used this tag) and while I have appearently added documentation for this to the wiki in 2017, it was also part of a proposal IIRR, and was in use since 2010. The reason why this is not present any more in significant numbers lies in an undiscussed automated edit towards the end of 2017 as you can see here: https://taghistory.raifer.tech/#***/addr:staircase/

Last edited 2 months ago by dieterdreist (previous) (diff)

comment:6 in reply to:  2 Changed 2 months ago by Klumbumbus

Replying to dieterdreist:

drop this test.

-1. The test is useful. I already found some tagging mistakes thanks to the test. I'll disable it if tagged on barrier=entrance or barrier=gate.

comment:7 in reply to:  5 Changed 2 months ago by skyper

Replying to dieterdreist:

Replying to dieterdreist:

The address of the house also has additional information (staircase B), so it is distinct. On the other hand, the gate has the required information (street and housenumber) for the gate. I still believe you should drop this test. It is too complicated to get it right globally, and you would have to know all keys that can be used to define a "POI", plus all the possible qualifiers that make housenumbers unique. And it would still fail for cases with actual duplicate housenumbers (the latter could be seen as a feature).

FWIW, this is not a new tag and I did not add it to the object in question (I hardly ever used this tag) and while I have appearently added documentation for this to the wiki in 2017, it was also part of a proposal IIRR, and was in use since 2010. The reason why this is not present any more in significant numbers lies in an undiscussed automated edit towards the end of 2017 as you can see here: https://taghistory.raifer.tech/#***/addr:staircase/

I apologize, I did not want to get personal.

Thanks for the information, I miss it on the wiki page.

Still I have two questions:

  • Numbers did raise quickly, too. Was this imported data?
  • Why was the undiscussed automated edit not reverted?

comment:8 Changed 2 months ago by dieterdreist

I just discovered it (I am also not completely sure it was not discussed, but I can't remember it at least, and generally automated edits almost never are discussed even less approved), and I was not able to find it yet. (the history planet extract I have locally is just Italy and in some other older data I happened to have on the disk the tag wasn't present). I am not sure how these tags came into use, it does not seem to be a single edit which added them, but I agree the line is quite steep. If you are able to find the edit which removed these, it would be interesting to contact the author.

comment:9 Changed 2 months ago by dieterdreist

To add more detail, there was a proposal 2011 where addr:staircase was initially contained, but it was then removed in the RFC phase before voting. More details here: https://wiki.openstreetmap.org/wiki/Proposed_features/addr_keys_%282011-04%29

Last edited 2 months ago by dieterdreist (previous) (diff)

comment:10 in reply to:  9 ; Changed 2 months ago by skyper

Replying to dieterdreist:

To add more detail, this was voted in 2011, and while addr:staircase was originally contained in the proposal, it was then removed in the RFC phase. More details here: https://wiki.openstreetmap.org/wiki/Proposed_features/addr_keys_%282011-04%29

So the status is still proposed. I have quite some OSM wiki edits on my ToDo and start right now. I see, if I'll find time for addr:key but I have not much insights about addr tagging.

I think my energy is better used by updating the JOSM wiki which will be quite some work as there are many new features and some major pages really ugly.

comment:11 Changed 2 months ago by dieterdreist

another area where there are potentially problems with this test (not tagged by me, but arguably not wrong, although it might be incomplete, not sure about the internal numeration). https://www.openstreetmap.org/way/58521218 (and around there)

comment:12 in reply to:  10 Changed 2 months ago by dieterdreist

Replying to skyper:

So the status is still proposed.

actually it isn't contained in any proposal as far as I can see, or do you mean "proposed on the tagging mailing list in 2011"?
The status is "in use" and crippeled by undiscussed automatic tag fiddling.

comment:13 Changed 8 weeks ago by Klumbumbus

Resolution: fixed
Status: newclosed

In 16829/josm:

  • fix #19579 - Ignore generator:output:electricity and plant:output:electricity
  • fix #19589 - Don't warn about duplicate housenumbers if one is tagged on a barrier=entrance or barrier=gate node
  • fix #19604 - Warn if the value of alt_name is the same as of name (patch by Famlam)

Modify Ticket

Change Properties
Set your email in Preferences
Action
as closed The owner will remain team.
as The resolution will be set.
The resolution will be deleted.

Add Comment


E-mail address and name can be saved in the Preferences.

 
Note: See TracTickets for help on using tickets.