Thanks, definitely an improvement. I'll do any further refining on a smaller scale in future.

Whoops, yeah I keep forgetting to add that tag to passes. I'll keep it in mind. Cheers

Here's the relation for Jamison Valley I've come up with. relation/12107413

Whilst the GNB descriptions aren't the best, it is unlikely that there will be on-ground indicators to where a valley starts and stops (I say this as a local who went down into the Jamison and Kedumba valleys 2 weeks ago). Agree that we can do better than an arbitrary straight line between The Col and Kedumba Pass; but there is a well-defined ridge line that fairly clearly separates the catchment areas that roughly follows a straight line. Not saying all the valleys here can be mapped in this way, clearly it has to be looked at case-by-case. Alternatively, yes they could simply be mapped as points, but that's no fun and being ambitious in mapping is part of the spirit of OSM that you mention. I suggest to start off, we at least try mapping Jamison Valley bounds as a multipolygon - to me it looks pretty straightforward. I could have a crack at it and link the changeset here.

On further inspection, GNB has provided fairly rigid descriptions of valley bounds (example for way/889441951 is https://proposals.gnb.nsw.gov.au/public/geonames/d8218040-81e9-4828-aff8-e0db1d2797ad) -generally bounded by easily identifiable features such as escarpments and ridges, a lot of which already have existing ways. In this case perhaps they should be mapped as a multipolygon?

Yeah I think leaving as single nodes rather than ways would be much easier unless someone is prepared to trace catchment areas (which in turn are difficult to define). One example is that the Jamison Valley (way/889441951) and Kedumba Valley (not included in your edit) are part of the same larger valley, I suppose they are separated by Sublime Point Ridge but who really knows.

Thanks, yeah seeing as the river defines the bound it seems reasonable. My comment was mostly for osmcha watchers to explain any flags; I didn't unglue anything.

Thanks, I see now. The problem was that there are several boundary ways on top of each other in that spot and so I must have included the wrong one in the forest relation. I'll get onto the straight bits progressively - there's a lot to do. Cheers

Hey thanks for your comments. Did you mean to remove ALL the landcover/landuse information from that part of Lidsdale State Forest? I'm not sure I understand what your reasoning is. State forests have a harvest cycle over decades so they aren't dynamic with respect to the OSM database. Regardless of the administrative bound, the bounds of the *managed* portion of Lidsdale State Forest haven't changed in years from what I can see on the aerial imagery. In that case I don't think you should have deleted way/888019432 and the majority of way/888019420. By that logic all the tracks in this state forest should also be deleted? Your changes also don't seem to be consistent with the other state forests that are tagged as landuse=forest.

"I also render them differently" can you elaborate on what you mean?

Hi, "overtraced" I meant as a synonym of "overfitted" https://en.wikipedia.org/wiki/Overfitting
I am all for detailed forest bounds -see my work in Kanimbla and Hartley area. Arbitrarily "joining the dots" between random isolated trees I don't believe is useful for OSM nor adds meaningful detail. In this case, the forest bound was crossing over houses and all kinds of things - it was not good quality geometry. I removed those nodes (rather savagely) because I was planning to redo the forest bound and residential area bound in a structured and orderly manner as I have done for some of the other towns up the mountain. The detail would have been added back by necessity.

Can you elaborate on what part of the forest relation was broken? I checked the relation for defects such as unclosed parts after I had finished my work and I did not see anything obvious.

Next time can you discuss on the changeset before reverting? As you are probably aware I am very active on OSM and would have given a quick reply. Cheers

Whoops, yeah that was a mistake. I'll fix it. The challenge id is 15439.

Thanks, I'll check it out in person and see if I can figure it out. Cheers

Thanks for the fix and heads up. I recently realised a lot of the wetlands I created in this area when I was a beginner mapper need to be made inners of surrounding land covers. I certainly have a bit of fixing up ahead of me.

hi there, welcome to OSM and thanks for your edits. Just letting you know you've put a bunch of buildings on top of existing ones here. Don't worry about fixing them yourself - I will fix this soon in order to preserve the OSM history of these objects. Cheers

Hi, thanks for your contributions. Just to let you know, I removed the website you added to the Picketts Valley node as it does not appear to be an official website. I've also fixed up a few things, like changing a path to a service road, and removing the redundant name "Private Road" from a private road. (see changeset/94946592). Cheers
---

Published using OSMCha: https://osmcha.org/changesets/94919097

Holy moly osmcha throws a lot of warnings for this one. As far as I can tell, nothing was "dragged".
---

Published using OSMCha: https://osmcha.org/changesets/94932901

Amazing work you've done in Armidale. One of the best mapped towns in NSW for sure. Big props

Thanks, I will fix

Hi there, welcome and thanks for your contributions. I've just made a few fixes to your edits, which you can have a look at over on: https://osmcha.org/changesets/94532717/
Keep mapping :)

Good idea, cheers