[OSM-talk] QA assessment (was Re: You are still free to continue to use or interpret this tag as you see fit since OpenStreetMap does not have “banned features”)

Roland Olbricht roland.olbricht at gmx.de
Wed Mar 31 15:41:02 UTC 2021


Hi Marc,

> please share how you deal with it

Thank you for the question. I would like to share my approach, others
may have different approaches. The primary aim is always to encourage
the mapper to map as understandable as possible.

First, check whether the mapper is still active:
https://www.openstreetmap.org/user/Kira999
The last changeset has been closed 11 months ago. Thus, there is only a
reduced chance that the mapper will respond to messages.

Then, check how much the mapper has mapped of the constellation in
question. In this case, the tags "diaper" and "changing_table" are
involved. In addition, as the last edit is from May 2020, I decided to
look at the OSM data as of 1st June 2020:
https://overpass-turbo.eu/s/15DP

Result:

count	diaper	changing_table
1		limited
32		no
1		yes
6	no
1	no	limited
20	no	no
1	yes
1	yes	no

Thus, the mapper has only tagged one object in a contradicting way.
Given that and that the mapper appears to be inactive, I would just
leave it as it is.

If sending a message is prospective then
- ensure that you understand and meet the standard of the local community
- choose the probably right language for the mapper
- ensure that the message is welcoming and encouraging

 > every day there are contributors who add a postal code to an POI because
 > the information is stored in another tag, otherwise (postal_code on the
 > commune relationship or on a postal_code relationship)

Again a good example. There exist no postcode areas in Germany, the
postcodes are intentionally mapped by the postal service immediately to
addresses. AFAIK this also applies to the UK and Ireland and probably
many other countries.

Thus, an app that works propely in Germany should consider postcodes on
the POI first, and postcodes of surrounding objects second.

Best regards,

Roland



More information about the talk mailing list