Closed acka47 closed 4 years ago
As this is about entries we derive from Wikidata, I suppose we resolve this together with #85.
To dos:
Another big task for restructuring was requested by H.F. via email on 2019-05-01:
Was ... die Position der Verwaltungsgliederung anbelangt, so erschien es uns schon immer als weniger gute Lösung, sie in der Hauptgruppe 9 ganz hinten in der Raumsystematik anzusiedeln, also hinter den kirchlichen und historischen Gebieten. Aufgrund der Gliederung des AGS ergibt sich aber nun eine gute Gelegenheit, sie ganz nach vorn zu bringen. Beginnen doch die Schlüssel des AGS für unser Bundesland NRW mit den Ziffern 05, um in den weiteren sechs Stellen Regierungsbezirke, Kreise und Orte numerisch strukturiert abzubilden. Die Gruppe 05 soll daher in Zukunft von unserer Verwaltungsgliederung eingenommen werden.
Damit dieses Konzept greifen kann, müssen wir allerdings die bislang unter 05 firmierende Gruppe "Westfalen" auf die neue Raumsystematik-Gruppe 04 umnotieren. Das alles kann natürlich nur automatisch durch Sie im hbz erfolgen. Wir denken, dass dafür auch jetzt der geeignete Zeitpunkt ist, ehe die von Ihnen geschilderten weiteren Arbeiten in Abstimmung mit der Verbundgruppe / Verbunddatenbank erfolgen.
H.F. also provided a "Auszugsskizze", see Skizze_NWB-RaumSysGr.0-mit-Vw.Gliederung-gem-AKS-u.-AGS.pdf:
Die Notationen ab der Ebene der Regierungsbezirke entsprechen den amtlichen Schlüsseln, wobei es neben dem bereits genannten Amtlichen Gemeindeschlüssel (AGS) auch einen Amtlichen Kreisschlüssel (AKS) (https://deacademic.com/dic.nsf/dewiki/71259#Aufbau) gibt. Der Kreisschlüssel ist fünfstellig, der Gemeindeschlüssel achtstellig. Beide Schlüssel finden sich auch in den Datenübersichten der Wikipedia-Artikel zu einzelnen Kreisen und Städten (vgl. z. B. Kreise Kleve und zugehörige Gemeinden).
Für einige Altkreise suchen wir noch nach ggf. älteren Schlüsseln. Über die Position der Euregio stimmen sich die NWBib-Redaktionen noch ab. Einige Kreisschlüssel müssen wohl noch in Wikidata von uns ergänzt werden.
Restructuring will occur in this file: https://github.com/hbz/nwbib/blob/lobid-vocabs-85/conf/non-90s-qids.json
With https://github.com/hbz/lobid-resources/pull/1006 we made sure that "Lippischer Wald" will be matched with the corresponding Wikidata entry. The data is now updated in lobid-resources but NWBib has to be adjusted as well, see e.g. https://nwbib.de/BT000059116 and https://nwbib.de/spatial#Q31312912.
Here is another reason for updating the NWBib: As reported by I.N. via email today, there are two new places that have been added by the NWBib editors:
Am 15.07.19 um 11:23 schrieb I.N.:
uns ist aufgefallen, dass zwei neue Ortsteile (Sendschotten, Twismecke ) noch nicht in der NWBib-Raumsystematik angezeigt werden.
Die Raumnotationen wurden in der HBZ-Titeldatei wie üblich im Feld 700n „99 | Drolshagen-Sendschotten“ und 700n „99 | Schmallenberg-Twismecke“ erfasst.
While both are missing in https://nwbib.de/spatial and Sendschotten is not shown in detail views (see e.g. https://nwbib.de/HT020122055), the detail view for "Twismecke" strangely looks good, see https://nwbib.de/HT020127780
For the last step "Move Regierungsbezirke, Kreise, Gemeinden etc. under 05", we just have to add skos:broader nwbib-spatial:N05
to the seven Regierungsbezirke and Euregio and create a new entry nwbib-spatial:N05
with skos:prefLabel "Regierungsbezirke, Kreise, Orte"@de
while moving the existing N05
to nwbib-spatial:N04"
.
[...] while moving the existing N05 to
nwbib-spatial:N04
Hm, but there is no nwbib-spatial:N04
(yet). Or do you mean nwbib-spatial#n4-7
?
but there is no
nwbib-spatial:N04
(yet). Or do you meannwbib-spatial#n4-7
?
Yes, the existing N05
becomes N04
:
nwbib-spatial:N04 a skos:Concept ;
skos:broader nwbib-spatial:N0 ;
skos:inScheme <https://nwbib.de/spatial> ;
skos:notation "04" ;
skos:prefLabel "Westfalen"@de ;
foaf:focus wd:Q8614 .
Should now contain everything from the original comment, except the last point, which moved to https://github.com/hbz/lobid-vocabs/issues/91:
For testing, I've set up usage of the new SKOS file and disabled adding other entries here:
I think it would only be consistent to move the Dekanate (N37) under the respective Diözese (N36) as we do with Kirchenkreise and Bistümer. However, we will ask NWBib editors before adjusting this.
Yes, I think that was agreed upon in the last meeting, so I think we can go ahead with that (https://wiki1.hbz-nrw.de/display/SEM/NWBib-Raumsystematik-Treffen+2019):
"Einstimmigkeit, dass Dekanate und Kirchenkreise hierarchisch unter die Diözesen bzw. Evengelische Kirche Rheinland/Westfalen sortiert werden"
Also, some are already sorted under Diözesen: https://test.nwbib.de/spatial#N36
We should probably wait with deploying this until that is done too.
I added the hierarchy for Dekanate with https://github.com/hbz/nwbib/pull/483/commits/0aee324dcef447befa55d3fa34b5f24f962a7e16.
I did this manually now but we should do this autmatically. We basically just have to use the property P708 "Diözese" for obtaining hierarchical information. i used this SPARQL query to get all information I need: https://w.wiki/7FU
Deployed with latest changes:
https://raw.githubusercontent.com/hbz/lobid-vocabs/85-spatial/nwbib/nwbib-spatial.ttl https://test.nwbib.de/spatial
As discussed, I created a new issue for the remaining TODO: https://github.com/hbz/lobid-vocabs/issues/93
+1 (We will have to rename the label for 36, currently "Katholische Kirche. Diözesen". Will ask NWBib editors after deployment to production what label we will use in the future.)
OK, so when you approve https://github.com/hbz/lobid-vocabs/pull/92, I'll merge the Turtle file and deploy the new setup based on that file to production (https://github.com/hbz/nwbib/pull/483)
Done
Fixes part of https://github.com/hbz/nwbib/issues/452 ("Dekanate und Kirchenkreise hierarchisch unter die Diözesen bzw. Evengelische Kirche Rheinland/Westfalen sortieren").