building=yes should only apply to the building, if you're drawing the area to cover the grounds/site there shouldn't be a building tag. I think the iD editor adds this by default as part of the preset, so may need to be manually removed.

I've restored the landuse=commercial tag, there is no other landuse tagged here and I think commercial is the best fit for a childcare based on landuse=*

the original tree nodes seem to be well mapped and had further details confirmed via a survey so I've reverted this change.

While in this case I think it's worth retaining tree nodes, they were originally added from imagery and likely weren't accurate anyway (2 nodes added where there was one tree, or vice versa) therefore I've left this change

I think in this setting the trees can be mapped as nodes and to restore the deleted nodes I've reverted this changeset.

I've reverted this to retain the original tree nodes.

I've reverted this change.

I think in this case the trees which were mapped as nodes are sufficiently distinct to be mapped as nodes, especially here where they are sporadic and not in a strictly regular interval along the nature strip.

"Saturday: Closed,Sunday: Closed" can be added with "Sa off; Su off" this makes it explicit that it's closed on the weekend.

I dispute the point that the tree nodes didn't have any other details on them, node/12135349872 had leaf_cycle=deciduous which was lost in this change.

Regardless I agree with others points here that in this case the individual trees should be retained, even if they were just natural=tree.

I agree that it's useful to map tree rows as features especially the kind in the photo at natural=tree_row

I have at times combined both methods using mapping a way with natural=tree_row with the nodes along the way set at the position of each tree within the row and then applied natural=tree to those nodes. That captures both the individual trees and the "tree row". This approach is described on the wiki.

I'm not sure if these trees should be considered a tree row though.

What shouldn't be individual natural=tree nodes? a dense hedge.

hi I see you've removed the Aboriginal site tags, I raised this with the original mapper at changeset/127996117 if the site is private property and public access is restricted then it's good to add the access=private tag, however the Aboriginal site should still have the relevant tags preserved, especially since this is a publicly documented site according to https://www.saveglenlee.com/

What was your reasoning for removing the tags?

Yeah according to https://www.saveglenlee.com/ it's private property so I think you were right to delete the nature reserve way/1106545342/history

However, the site also mentions a Aboriginal sites, so it does sound like there are some publicly documented sites here. In this case I think it's important to retain the tags for the rock carvings but agreed that access=private should be added if not open to the public.

motor_vehicle=designated implies that the public can drive their cars here? Is that the case here? If it's only for management, maintenance and emergency vehicles then motor_vehicle=private should be used.

ah sorry I see you've done that now.

some of these still had proposed:highway=*, this should be removed if the road is already constructed now, or if construction is underway then changed to construction:highway=*.

What about motor_vehicles? My preference is to not add a top level access=no but instead specify each mode, you've added bicycle=no, foot=yes, but it's now unclear about access by emergency services, given the gates here it's probably motor_vehicle=private ?

The node https://osmlab.github.io/osm-deep-history/#/node/10126074223 was edited, do you have any comment on that change?

It's best to add access=private + ownership=private.

This is more than just a single parking spot for a residence, it's a few marked spots for a community centre, so while private is still useful to me mapped.

Having it mapped also helps with landuse, mobility and economic research, analysis and studies.

As such I've restored the parking and added the private tags.

Driveways are likely considered acceptable to be mapped, so they should not be deleted if already mapped by others unless they have been demolished. Others may find these useful data.

They are visible to the public from the street and visible on aerial imagery.

Personally I find them useful for pedestrian routing and analysis, since driveways create a hazard for pedestrians so having them mapped allows for analysis looking at the safety of walking routes.

They also help with routing as they provide hints about where a house may be accessed from the street.

As such I've reverted this change to restore the driveways.

Thanks, I can see in that document it's referring both to "Chatswood North" as the name of a "Local Area Traffic Management (LATM)" which I probably wouldn't think as warranting a place=neighbourhood but also as the "Chatswood North precinct" which probably does warrant place=neighbourhood.

place=neighbourhood has varied use in Sydney ranging from GNB assigned urban/rural place names that don't form formal suburb/localities, usually as a area/region of a suburb, but also for named residential estates.

I think then it's okay to map "Chatswood North" as a neighbourhood here, though improved tagging practices and tags to better categorise "neighbourhoods" would be good for OSM.