I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
I've updated the GWR extracts
Thanks, Simon!
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA.
I once learned that it is frowned upon to include foreign keys to OSM, but did not agree with the reasoning. Would be fine with me...
There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer
Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Could you explain a bit more? What is their purpose? How do they look like? When do they change? Etc. I guess, some background knowledge would be necessary in order to come to a well-founded conclusion.
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
I am generally interested. However, to me it's still unclear, even if we get the tagging and pre-processing all sorted out, how would we bring this data into OSM? I.e. what tools would we use and how much effort would this take? Would this effort better be spent on importing addresses or could we get both in one go?
Best regards
Lukas
Am 26.03.2023 um 21:34 schrieb Lukas Toggenburger:
I've updated the GWR extracts
Thanks, Simon!
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA.
I once learned that it is frowned upon to include foreign keys to OSM, but did not agree with the reasoning. Would be fine with me...
I was just pointing out that Jonas had suggested this. As a tendency I would argue that given that the addresses themselves are nearly always an unique key it isn't really necessary for update purposes.
There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer
Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Could you explain a bit more? What is their purpose? How do they look like? When do they change? Etc. I guess, some background knowledge would be necessary in order to come to a well-founded conclusion.
They are all just integers, the entrance number is zero for buildings with just one entrance. I don't believe any of them have internal structure, they are just allocated sequentially. Entrance data:
EGID EDID EGAID DEINR ESID STRNAME STRNAMK STRINDX STRSP STROFFIZIEL DPLZ4 DPLZZ DPLZNAME DKODE DKODN DOFFADR DEXPDAT 1 0 100000334 20 10000533 Grossholzerstrasse Grossholzerstr. Gro 9901 1 8910 0 Affoltern am Albis 2676465.141 1235849.103 1 2023-03-21
2 0 100000335 1 10000537 Im Feld Im Feld Fel 9901 1 8910 0 Affoltern am Albis 2676541.16 1235979.043 0 2023-03-21 .....
and the building data:
EGID GDEKT GGDENR GGDENAME EGRID LGBKR LPARZ LPARZSX LTYP GEBNR GBEZ GKODE GKODN GKSCE GSTAT GKAT GKLAS GBAUJ GBAUM GBAUP GABBJ GAREA GVOL GVOLNORM GVOLSCE G ASTW GANZWHG GAZZI GSCHUTZR GEBF GWAERZH1 GENH1 GWAERSCEH1 GWAERDATH1 GWAERZH2 GENH2 GWAERSCEH2 GWAERDATH2 GWAERZW1 GENW1 GWAERSCEW1 GWAERDATW1 GWAERZW2 GENW2 GWAERSCEW2 GWAERDATW2 GEXPDAT 11513432 ZH 1 Aeugst am Albis CH540120777857 0 1573 1199 2679647.268 1237500.347 901 1004 1020 1121 2000 8019 166 32 7430 7530 860 2002-09-05 7630 7530 860 2002-09-05 2023-03-21 11513433 ZH 1 Aeugst am Albis CH587820017717 0 1543 1198 2680635.9 1236936.229 901 1004 1020 1110 2000 8019 90 21 7430 7530 860 2002-09-05 7650 7560 860 2002-09-05 2023-03-21 .....
Attribute documentation see https://dam-api.bfs.admin.ch/hub/api/dam/assets/7008785/master
Currently I'm just using the building data to get the municipality id GGDENR, which is, on the topic of foreign keys is what we maintain on the municipality boundaries (slightly misnamed for historical reasons).
Simon
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
I am generally interested. However, to me it's still unclear, even if we get the tagging and pre-processing all sorted out, how would we bring this data into OSM? I.e. what tools would we use and how much effort would this take? Would this effort better be spent on importing addresses or could we get both in one go?
Best regards
Lukas
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer
Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
PS: it seems as if coordinates for building centroids exist for these municipalities, worst case I'll build a workaround from them.
Am 13.05.2023 um 09:06 schrieb Simon Poole:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator +
Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Am 13.05.2023 um 09:12 schrieb Simon Poole:
PS: it seems as if coordinates for building centroids exist for these municipalities, worst case I'll build a workaround from them.
I've regenerated the per municipality files using the building centroid coordinates. Some spot checking would seem to indicate that this works reasonably well, if anybody sees any other issue pls get in contact with me.
Simon
Am 13.05.2023 um 09:06 schrieb Simon Poole:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator +
Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Just a further note on this. I cross checked a couple of the "missing" addresses with swisstopos "official" address data, and surprise (not really) they seem to have corresponding entrance coordinates and are not relying on the building centroids.
Simon
Am 13.05.2023 um 11:08 schrieb Simon Poole:
Am 13.05.2023 um 09:12 schrieb Simon Poole:
PS: it seems as if coordinates for building centroids exist for these municipalities, worst case I'll build a workaround from them.
I've regenerated the per municipality files using the building centroid coordinates. Some spot checking would seem to indicate that this works reasonably well, if anybody sees any other issue pls get in contact with me.
Simon
Am 13.05.2023 um 09:06 schrieb Simon Poole:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator +
Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Hi Simon
Thanks for your very quick reaction! Your workaround is very helpful. However, if a building has more than one entrance and thus more than one address, the files you have newly created include only one address, e.g.:
Viale Castagnola 21a/21b/21c/21d, 6900 Lugano - https://map.geo.admin.ch/?lang=en&topic=ech&bgLayer=ch.swisstopo.pix...
Note that Swisstopo (layer "RDB: building status") also has a problem with buildings that have multiple addresses in Ticino: it stacks all the addresses in the centre of the building. On the other hand, the Geoportale Ticino (layer "Indirizzi degli edifici (Misurazione ufficiale)") doesn't have this problem and displays the addresses on the entrances:
https://map.geo.ti.ch/?lang=en&baselayer_ref=Carta%20Nazionale%20(bianco...
Another peculiarity (but not a problem) i found in Ticino is that the RDB data (on Swisstopo and in your files) include non-unique addresses without housenumbers (only street name, postcode and place name), e.g.:
https://map.geo.admin.ch/?lang=en&topic=ech&bgLayer=ch.swisstopo.pix...
The Geoportale Ticino doesn't display an address for these buildings:
https://map.geo.ti.ch/?lang=en&baselayer_ref=Carta%20Nazionale%20(bianco...
Best regards Raphael
On Sat, 13 May 2023 at 11:09, Simon Poole simon@poole.ch wrote:
Am 13.05.2023 um 09:12 schrieb Simon Poole:
PS: it seems as if coordinates for building centroids exist for these municipalities, worst case I'll build a workaround from them.
I've regenerated the per municipality files using the building centroid coordinates. Some spot checking would seem to indicate that this works reasonably well, if anybody sees any other issue pls get in contact with me.
Simon
Am 13.05.2023 um 09:06 schrieb Simon Poole:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator +
Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Am 14.05.2023 um 20:20 schrieb Raphael:
Hi Simon
Thanks for your very quick reaction! Your workaround is very helpful. However, if a building has more than one entrance and thus more than one address, the files you have newly created include only one address, e.g.:
Viale Castagnola 21a/21b/21c/21d, 6900 Lugano - https://map.geo.admin.ch/?lang=en&topic=ech&bgLayer=ch.swisstopo.pix...
This seems to be an issue with how the OSM xml files are generated, they are present in the geojson files (from which the XML is produced).
Another peculiarity (but not a problem) i found in Ticino is that the RDB data (on Swisstopo and in your files) include non-unique addresses without housenumbers (only street name, postcode and place name), e.g.:
https://map.geo.admin.ch/?lang=en&topic=ech&bgLayer=ch.swisstopo.pix...
This is "normal" and is the case in other cantons too, they are useful for checking street level localisation, outside of that just ignore them.
The BfS is looking in to the issue BTW.
Simon
The Geoportale Ticino doesn't display an address for these buildings:
https://map.geo.ti.ch/?lang=en&baselayer_ref=Carta%20Nazionale%20(bianco...
Best regards Raphael
On Sat, 13 May 2023 at 11:09, Simon Poole simon@poole.ch wrote:
Am 13.05.2023 um 09:12 schrieb Simon Poole:
PS: it seems as if coordinates for building centroids exist for these municipalities, worst case I'll build a workaround from them.
I've regenerated the per municipality files using the building centroid coordinates. Some spot checking would seem to indicate that this works reasonably well, if anybody sees any other issue pls get in contact with me.
Simon
Am 13.05.2023 um 09:06 schrieb Simon Poole:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator +
Eidgenössischer Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
Am 15.05.2023 um 12:14 schrieb Simon Poole:
... This seems to be an issue with how the OSM xml files are generated, they are present in the geojson files (from which the XML is produced).
It seems that the Node.js lib I'm using eats nodes that are at the same location during the conversion. I'm going to away for most of the week, so I'm not quite sure if I'll have time to fix the issue by writing something myself.
Simon
OK I hacked together a quick replacement that seems to work better, if there are still issues with the conversion, please open an issue here
https://github.com/simonpoole/geojson2osm
Simon
Am 15.05.2023 um 12:41 schrieb Simon Poole:
Am 15.05.2023 um 12:14 schrieb Simon Poole:
... This seems to be an issue with how the OSM xml files are generated, they are present in the geojson files (from which the XML is produced).
It seems that the Node.js lib I'm using eats nodes that are at the same location during the conversion. I'm going to away for most of the week, so I'm not quite sure if I'll have time to fix the issue by writing something myself.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
I'm just happy to hear anything about addresses in the Ticino ... from my recent mapping experience there, most villages have no adresses.
Am Samstag, 13. Mai 2023 um 09:07:55 MESZ hat Simon Poole simon@poole.ch Folgendes geschrieben:
I just knew that changing to the new data format went too smoothly: for a number of municipalities in Ticino, for example Locarno, the data from the BfS is missing coordinates and as a consequence the per-municipality files are empty.
I just checked the data from today and the coordinates continue to be absent. I've contacted the BfS about the matter, lets see what they have to say.
Thanks to dafadllyn for pointing this out.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer
Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
_______________________________________________ talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
I've update the GWR extracts to data from yesterday. If you see any issue please report them here.
Note that I haven't resolved the issue wrt the street data yet, it seems as if it will boil down to conflating data from the GWR and swisstopo.
Simon
Am 21.03.2023 um 15:46 schrieb Simon Poole:
I've updated the GWR extracts that are available from http://qa.poole.ch/addresses/ch/ to data from the 20th of March of this year. As the BfS changed how they provide the data last year (that's the reason that this update was delayed), I used to opportunity to improve and simplify how I update the data. I hope this will lead to more frequent updates, at least till the next time the BfS changes everything.
The caveats with respect to the data quality still apply, in particular because of the rather questionable street data for some mainly mountainous municipalities some local knowledge is a good thing to have before using the data.
On a related note, Jonas (loremo) suggested at one point in time that we should include one of the GWR ids with the data for referencing updates/QA. There seem to be two possibilities how we could do that:
- EGID + EDID (Eidgenössischer Gebäudeidentifikator + Eidgenössischer
Eingangsidentifikator)
or
- EGAID (Eidgenössischer Gebäudeadressidentifikator)
The former would seem to be more useful as it allows to determine if the entrances belong to the same building. Any comments? Tagging suggestion?
Further because of the changes it would be possible to generate building type and level count information, it is a bit unclear how well this would map to OSM tags, but is somebody is interested we could look in to this.
Simon
talk-ch mailing list talk-ch@openstreetmap.ch http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
-
Lukas Toggenburger -
Raphael -
Sentalize -
Simon Poole