Looks good on the tramway, just a few nodes in the city look dragged when maybe shouldn't have been?
---
#REVIEWED_GOOD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/96648418

That's fine if that's how it's defined in OSM. Technically there is a small section at the north without a cliff, and there is a gap on the east where a road goes up, so mostly surrounded by cliff but not completely.
---

Published using OSMCha: https://osmcha.org/changesets/96647702

Yep and I disagree with natural=valley only being a single way along the valley floor. This changeset and others are experimenting with alternative and improved ways of mapping valleys with the goal to eventually do a proposal to change the wiki. We need some mapped this way to see what works and what doesn't and to demonstrate the style. I'll try and start writing up a draft proposal on this style of mapping.

I understand that, but I still think there is no harm and a lot of benefit to OSM containing our best effort of mapping the extent of these valleys based on both local knowledge and government sources.

Do you have any specific concerns about the Jamison Valley after it was refined in relation/12107413 if you can raise specific points or issues with this one then we can try to work through them.

GNB designates this as a pass. Instead of tagging this as a generic unpopulated place name (locality), I think that mountain_pass=* is a better tag, indeed that's how I've tried to map passes and helps data consumers looking to find named mountain passes.
---
#REVIEWED_GOOD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/96583917

> We map what actually exists in the real world, and “original research” by surveying locations in person is our gold standard.

While I 100% agree with this and frequently argue for this approach. If community standard and expectations change, then this might not always hold true and it's fine for the community to agree to situations where this doesn't or can't apply.

Marine Parks have been mapped around Australia and these are not verifiable (nothing exists to mark their bounds or names on the high seas) but there are enough stakeholders and mappers who want to see these mapped and work to map them based off government sources.

RE point v area if one really don't have any clue on the bounds then agreed best to map as a point (eg. Hunter Valley I'd only be guessing for an area as I don't have enough local knowledge), it's just than where possible an area conveys much more information, about size and rough extent, this helpsfor cartography (knowing rough size and extent for label placement) and reverse geocoding (where am I type queries).

RE GNB. Great suggestion, that's a solid source as a reference to help refine the boundaries more precisely. I'm neutral to either a way or relation. Either way it can be defined by the ridge. With a way it can share nodes, with a relation it can be a member.

RE. Kangaroo Valley, absolutely you can argue if it should it cover just the "floor" or extent up to the ridge or cliff tops. That's a whole separate discussion worth having. But for now as a first pass just so we have something in OSM instead of nothing, either options or anywhere in between is good enough and better than nothing in my opinion.

Sure I get the argument that placing a node puts in in the right valley of the DEM, and then a program could choose to follow that point out from the valley floor, but I'm not sure if that's foolproof.

Still I appreciate the comments here and discussion raised.

I know this is a complex topic, and there are a variety of different opinions and approaches used globally. Generally I subscribe to approach 1 at osm.wiki/Forest#Approach_1 but I admit that this might be too simplistic and not enough to cover the finer details.

But at a high level, if there are trees there then it should be covered by natural=wood and if it's managed forest for timber production then covered in landuse=forest (and if both then can be covered in both tags).

Usually I'd just snap the admin boundaries to the river because mostly the boundaries do follow the river.

are you sure https://osmlab.github.io/osm-deep-history/#/way/172996908 permits motor vehicles? Usually firetrails are accessible by authorised vehicles only, usually determined either by signage or by a locked gate. If there is a locked gate here, then best to leave motor_vehicle=private to stop it routing the general public in 4wd vehicles through here, while still saying it's accessible by fire trucks.

good. but ps. you dragged the address node here (https://osmcha.org/changesets/96564672), but I fixed that in changeset/96566925

all good, just I don't think layer is needed unless there are overlapping power lines and you want to say which power line is over other power lines.
---
#REVIEWED_GOOD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/96564832

Tagging looks good, but don't think the layer tags are needed, layer=* is for distinguishing the physical vertical relationship between features, so doesn't make sense to apply it to the social facility object, unless it was a building under or over another building, or something like that.
---
#REVIEWED_GOOD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/96566166

Great thanks for checking it and fixing.

I think it's a good idea to map them in OSM, but agree we don't have a good tag for public land reserve. But I don't agree with blanket tagging these leisure=park, as a kid rocking up with his soccer ball or a family with their picnic rug expecting a local park might be dissapointed.

hi you've significantly changed the Badgerys Creek - Luddenham suburb boundaries, previously running through the airport which you changed to follow The Northern Road. What's your source for this change because the DCS Suburb dataset still has the boundaries as they were. Unless you have any other reason or evidence to suggest it should follow The Northern Road, I'm inclined to leave it as it was before to match the DCS data.

We haven't received permission to use NSW Planning data. Although it's CC BY licensed per https://blog.openstreetmap.org/2017/03/17/use-of-cc-by-data/ we still need a waiver. There is a list of organisations we have a waiver for at osm.wiki/Australian_data_catalogue but as far as I'm aware no one has asked for one from Planning yet. So we shouldn't be using any copyrighted data from Planning until we have that waiver in place.

Most of these I couldn't consider leisure=park based on leisure=park. If they are council reserves, I'm not sure the best tag to use, in the past I've used some combination of natural=wood, boundary=protected_area, leisure=nature_reserve, or even just untagged with a note.

I checked the boundaries and they seem okay, @ZeLonewolf can you point out specifically where it's broken?

Island name restored in changeset/96228931