Re: [talk-au] Suburbs: Nodes, Areas, or both?
On 6/11/21 20:30, Simon Poole wrote: Yes, Gruyère is a cheese, it's named after the town of Gruyères see https://en.wikipedia.org/wiki/Gruy%C3%A8res, very nice place BTW. That's why I was wondering :-), but I suppose the dairy industry explains it. I think Wikipedia is wrong on this one. It's probably named after the town because that's where one of the early big landholders was from: https://en.wikipedia.org/wiki/Paul_de_Castella Turns out the de Castella family is still in the wine industry: http://www.jeanpaulsvineyard.com.au/ ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
Am 06.11.2021 um 10:22 schrieb fors...@ozonline.com.au: Quoting Simon Poole : PS: wondering why Gruyere has that name. Good question. The town is named for a variety of cheese, as the area's history is in the dairy industry. Cahillton Post Office first opened on 20 August 1892. It was renamed Gruyere in 1950 and closed in 1960 Wikipedia Yes, Gruyère is a cheese, it's named after the town of Gruyères see https://en.wikipedia.org/wiki/Gruy%C3%A8res, very nice place BTW. That's why I was wondering :-), but I suppose the dairy indsutry explains it. Simon Tony OpenPGP_signature Description: OpenPGP digital signature ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
Quoting Simon Poole : PS: wondering why Gruyere has that name. Good question. The town is named for a variety of cheese, as the area's history is in the dairy industry. Cahillton Post Office first opened on 20 August 1892. It was renamed Gruyere in 1950 and closed in 1960 Wikipedia Tony ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
This is a somewhat unsolved issue in OSM modelling, as both area (extent) and node (assuming it is not simply the centroid of the area) convey geometric information that the other cannot. IMHO best would be to have a similar concept as we do for administrative areas that works for "places" in a more general sense, that is that allows simple deduplication of the elements. A rather epic discussion on the topic can be found here: https://github.com/gravitystorm/openstreetmap-carto/pull/2816 Simon PS: wondering why Gruyere has that name. Am 05.11.2021 um 07:53 schrieb cleary: Ideally suburbs would have a relation for the boundary PLUS a node for the "label node" as part of the relation. I'm not so familiar with Victorian locations, but this example for South Albury in NSW is an example: https://www.openstreetmap.org/relation/5901488 Where there is a boundary and a separate place node, I would add the place node to the relation and its role would be "label node". On Fri, 5 Nov 2021, at 2:15 PM, Dian Ågesson wrote: Hey all, I would appreciate the thoughts of the community with regards to suburb representations. In a recent change set (https://www.openstreetmap.org/changeset/113355648) a node was introduced for Gruyere. Gruyere is on the urban boundary, but is technically in Metropolitan Melbourne. As such, it straddles the border between what could be considered a bona fide suburb, and an independent town. Mick has correctly pointed out that many of the other localities in the area are represented by both an area and a node. Is this the way all suburbs should be represented? Or is it an urban/rural distinction? Dian ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au OpenPGP_signature Description: OpenPGP digital signature ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
On 5/11/21 18:06, cleary wrote: Sorry. I should have written ...add the place node to the relation and its role would be "label". Done: https://www.openstreetmap.org/relation/2428804 ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
Sorry. I should have written ...add the place node to the relation and its role would be "label". On Fri, 5 Nov 2021, at 5:53 PM, cleary wrote: > Ideally suburbs would have a relation for the boundary PLUS a node for > the "label node" as part of the relation. I'm not so familiar with > Victorian locations, but this example for South Albury in NSW is an > example: > https://www.openstreetmap.org/relation/5901488 > > Where there is a boundary and a separate place node, I would add the > place node to the relation and its role would be "label node". > > > > On Fri, 5 Nov 2021, at 2:15 PM, Dian Ågesson wrote: >> Hey all, >> >> I would appreciate the thoughts of the community with regards to suburb >> representations. >> >> In a recent change set >> (https://www.openstreetmap.org/changeset/113355648) a node was >> introduced for Gruyere. Gruyere is on the urban boundary, but is >> technically in Metropolitan Melbourne. As such, it straddles the border >> between what could be considered a bona fide suburb, and an independent >> town. >> >> Mick has correctly pointed out that many of the other localities in the >> area are represented by both an area and a node. >> >> Is this the way all suburbs should be represented? Or is it an >> urban/rural distinction? >> >> >> >> Dian >> >> ___ >> Talk-au mailing list >> Talk-au@openstreetmap.org >> https://lists.openstreetmap.org/listinfo/talk-au > > ___ > Talk-au mailing list > Talk-au@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk-au ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
Ideally suburbs would have a relation for the boundary PLUS a node for the "label node" as part of the relation. I'm not so familiar with Victorian locations, but this example for South Albury in NSW is an example: https://www.openstreetmap.org/relation/5901488 Where there is a boundary and a separate place node, I would add the place node to the relation and its role would be "label node". On Fri, 5 Nov 2021, at 2:15 PM, Dian Ågesson wrote: > Hey all, > > I would appreciate the thoughts of the community with regards to suburb > representations. > > In a recent change set > (https://www.openstreetmap.org/changeset/113355648) a node was > introduced for Gruyere. Gruyere is on the urban boundary, but is > technically in Metropolitan Melbourne. As such, it straddles the border > between what could be considered a bona fide suburb, and an independent > town. > > Mick has correctly pointed out that many of the other localities in the > area are represented by both an area and a node. > > Is this the way all suburbs should be represented? Or is it an > urban/rural distinction? > > > > Dian > > ___ > Talk-au mailing list > Talk-au@openstreetmap.org > https://lists.openstreetmap.org/listinfo/talk-au ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au
Re: [talk-au] Suburbs: Nodes, Areas, or both?
Hi Dian, I have an interested in mapping what I call, for want of better terminology, fuzzy names or sense-of-place and comment in that specific regard. In summary: if the suburb has a defined boundary, use an area, if it doesn't use a node. I would certainly NOT however use both to represent the same suburb. From experience with Russian cities, that makes if very difficult to make maps without pre-processing OSM data to remove duplicates. For things like airports and islands where this often happens accidentality due to the evolution of the map or simple misunderstanding, I can and do remove the node and merge any extra info onto the area. I would comment that from my understanding, Australian suburbs are somewhat unusual in often having defined admin/postal boundaries. A more common situation is a "sense of place" that can really only be mapped with a node. As an example, my UK home town has an area mapped as a suburb called the Weston Estate. In the 1930s(!) it was a defined new housing development. Everyone know where it is, north of the river and to the west of the road out of the valley. But does it include later development? Does it include the older houses and a couple of farms? Hard to say and who you talk to gets a slightly different answer. So dangerous to map an area because then the map is making the landscape. Perhaps this is the case with Gruyere? (I genuinely don't know). If anyone has an interest in sense-of-place mapping, I've experimented with is_in:* tags to map physiological regions, often historic but still relevant or loosely geographic. The idea being to end up with a point cloud that can then be processed according to need. I find that if you ask someone who lives there, "Are in X?", they can give a straight and usually consistent yes/no answer. But if you ask "And where does it end?", you'll get either a very vague answer or a look of panic. But I am wandering off topic, so will leave it there. Mike On 2021-11-05 04:15, Dian Ågesson wrote: Hey all, I would appreciate the thoughts of the community with regards to suburb representations. In a recent change set (https://www.openstreetmap.org/changeset/113355648) a node was introduced for Gruyere. Gruyere is on the urban boundary, but is technically in Metropolitan Melbourne. As such, it straddles the border between what could be considered a bona fide suburb, and an independent town. Mick has correctly pointed out that many of the other localities in the area are represented by both an area and a node. Is this the way all suburbs should be represented? Or is it an urban/rural distinction? Dian ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au ___ Talk-au mailing list Talk-au@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-au