3 ms·

That's not the case in my local area. The suburbs boundaries align quite accurately with the boundaries provided on the local council's GIS. The OSM boundaries

by timv 8y ago

That's not the case in my local area. The suburbs boundaries align quite accurately with the boundaries provided on the local council's GIS. The OSM boundaries are line-strings, so they don't have all the contours from the council, but it's clearly a representation of the real boundary, not the ABS boundary. What areas are you looking at.

rmc 8y ago
OSM boundaries are stored in the OSM DB as polygons (which are made up of linestring)
squiggleblaz 8y ago
Hi. If you read my comment, I acknowledge that I can't simply replace the existing data with an upload because they've probably been fixed - the problem is that the suburb boundaries that are in OSM are a mish-mash of suburb/locality boundaries and state suburbs boundaries. This [probably] occurred as a matter of history because they started as "state suburbs" and then people have corrected them in metropolitan areas. Consequently, today they represent nothing - neither statistical areas nor address areas - because you never know whether this particular area (a) represents the SS (b) represents the locality (c) represents both because they're identical (d) represents the SS on one front and the locality on the other. One place to look at is Mirboo. I've just picked it to avoid yielding too much PII, but it barely respects the existence of surrounding rural districts. (It might even swamp them like others I know - I'm afraid I don't have my full system here.) I know of other examples (locations to which I have an attachment), but you'll forgive me if I avoid giving too much PII on a public forum. I'm afraid I can't give an exhaustive list because I haven't produced one. Also because I anticipate that if I produce a report it will come to nothing - I usually find that public databases are curated by people who will refuse to accept that any major corrections are necessary because they've already worked on the problem for years and therefore a major correction is taken to be an insult. As an example, if I can only produce examples in rural areas, I fear that someone will say "we have thousands of correct addresses, but there's only twenty addresses affected by this example, and thirty by that example - it's not important enough". At that point it seems there's no point even producing a report. OSM will just fall into the category of useless databases, where, like Wikipedia, accuracy and convenience are not valued.
- timv 8y ago
  I read your comment, but in the absense of any specific examples it was hard to know the details of the problem. The OSM Wiki contains a Data Catalogue for Australia. https://wiki.openstreetmap.org/wiki/Australian_Data_Catalogue https://wiki.openstreetmap.org/wiki/Australian_Data_Catalogu... The issue right now is that both the PSMA data (that you want OSM to use) and the ASGS data (that you infer it is using - and in some cases that might be the source of the boundaries that were entered, but it is not officially in use by OSM) are licensed under CC-BY, which is not 100% compatible with the ODbL that OSM uses. The necessary solution is to request an explicit waiver from the data provider, which has been requested for PSMA, but (as far as I know) never received a response.

by timv 8y ago

rmc 8y ago

OSM boundaries are stored in the OSM DB as polygons (which are made up of linestring)

squiggleblaz 8y ago

Hi. If you read my comment, I acknowledge that I can't simply replace the existing data with an upload because they've probably been fixed - the problem is that the suburb boundaries that are in OSM are a mish-mash of suburb/locality boundaries and state suburbs boundaries. This [probably] occurred as a matter of history because they started as "state suburbs" and then people have corrected them in metropolitan areas. Consequently, today they represent nothing - neither statistical areas nor address areas - because you never know whether this particular area (a) represents the SS (b) represents the locality (c) represents both because they're identical (d) represents the SS on one front and the locality on the other. One place to look at is Mirboo. I've just picked it to avoid yielding too much PII, but it barely respects the existence of surrounding rural districts. (It might even swamp them like others I know - I'm afraid I don't have my full system here.) I know of other examples (locations to which I have an attachment), but you'll forgive me if I avoid giving too much PII on a public forum. I'm afraid I can't give an exhaustive list because I haven't produced one. Also because I anticipate that if I produce a report it will come to nothing - I usually find that public databases are curated by people who will refuse to accept that any major corrections are necessary because they've already worked on the problem for years and therefore a major correction is taken to be an insult. As an example, if I can only produce examples in rural areas, I fear that someone will say "we have thousands of correct addresses, but there's only twenty addresses affected by this example, and thirty by that example - it's not important enough". At that point it seems there's no point even producing a report. OSM will just fall into the category of useless databases, where, like Wikipedia, accuracy and convenience are not valued.

timv 8y ago
I read your comment, but in the absense of any specific examples it was hard to know the details of the problem. The OSM Wiki contains a Data Catalogue for Australia. https://wiki.openstreetmap.org/wiki/Australian_Data_Catalogue https://wiki.openstreetmap.org/wiki/Australian_Data_Catalogu... The issue right now is that both the PSMA data (that you want OSM to use) and the ASGS data (that you infer it is using - and in some cases that might be the source of the boundaries that were entered, but it is not officially in use by OSM) are licensed under CC-BY, which is not 100% compatible with the ODbL that OSM uses. The necessary solution is to request an explicit waiver from the data provider, which has been requested for PSMA, but (as far as I know) never received a response.