Why did you delete this golf course? Did it close? The holes didn't go anywhere; does someone else operate it now?

Looks like they were traced from the USGS Topo map, but they no longer exist since the quarry tore them up. I have removed them.
There are still some other streams that don't seem to terminate anywhere; I wonder where the water ends up or if it just floods the quarry.
See changeset #83049200

Hi,
Thanks for adding so many solar farms in NC!
Are you sure that Bing is what you're tracing over? Most of these panels are misaligned from all imagery and are not even the correct size. In this area, Esri World Imagery is the nearest to 'truth' alignment as far as we can tell. Mapbox is also pretty good, but Bing is off by a bit and DG was always off. Check that things are lined up before added lots of things like solar panels.

Thank you,
Will

It is, actually. Please review tagging guidelines before altering random things that you are not familiar with.

Got scammed by formatting up there but hopefully it's still readable. :/

I'm also not sure that this is even TNRIS data, as the landcover I can find on their data portal is NLCD data which is created by USGS (https://data.tnris.org/collection/89b4016e-d091-46f6-bd45-8d3bc154f1fc)

> It isn't an import, I made it.
> it is derived from semi automated classification of aerial imagery ... by the organization in question which is distributed as a faster file under an open license.
You performed the processing that took it from one form to another, and then uploaded it. You did not generate/create the data yourself, TNRIS did and made it available.
Imports do not have to be automated, they can be and often are manual. OSM has guidelines (osm.wiki/Import/Guidelines) for importing data from external sources which clearly state **Community Buy-in**, **Documentation**, and **Import Review**. Just because it *is* licensed properly and *can* be imported doesn't mean it should be.

> As the features were created solely by me through geoprocessing and as **I copied them into the map myself** I do not think it qualifies as an import ...
Copying is importing if it is not your data.

As for quality. There is very little that separates these islands of landcover from the area next to them. Just from a single spot check, ways 699813571, 699812272, 699807593, and 699811350 in no way appear to be differentiated from the surrounding terrain. This data would not make it through any community quality review, had it happened.

Bottom line, large imports of data must be reviewed by the local and/or larger community, and I have serious doubts that this would have been approved.

Hi,
Did you document this import anywhere, and discuss it with the community? I can't find a page on the wiki for this TNRIS data.
If not I would suggest that this data (all of it) be removed because it is VERY low quality, inconsistent, and just generally not good.

Hi,
are you aware of the quality of building you are adding? Many of these buildings are just rectangles that approximate a more complex building. You are effectively performing an unchecked import of the Microsoft buildings database. Please adjust the building outlines to more closely match reality before uploading.

This seems like it broke a LOT of existing landuse multipolygons...

No I traced these by hand. Microsoft building imports are done in the RapiD editor and must be labeled as imports. The dataset is also not available for outside the U.S. osm.wiki/Microsoft_Building_Footprint_Data

Added a lot more than a hairdresser; dunno why that's the comment.

Hi,
please know that the Esri Clarity layer uses imagery (at least in this area) from around 2011. At least a few of the buildings you added here do not exist and are not visible on any other layers. Clarity is also offset by a couple meters from the other layers and probably reality.
Even though it's sharper than the other layers (thus named Clarity) it is almost never the newest or third-newest imagery.

... the directional prefixes are also not present in TIGER 2017, 2018, or 2019.

Hi,
please make sure that you are using the newest imagery, or are aware of it. Bing imagery in these rural Southern African areas is usually quite dated.
At Sirana, the A12 takes a different path, which is not overly important, but in other cases the road might be drastically different in reality than in Bing satellite.
Maxar is almost always the newest available imagery in rural Africa. At Sirana the imagery is from 2016-17; Bing may be as late as 2012 but they do not provide tile dates.
So please be aware of other imagery sources that are newer and that Bing is generally not new, even though it looks the sharpest.

Hi,
please be careful when splitting large ways like this. You kept the 'surface=ground' tag on all three segments, when both the bridge itself and the way forward of it are both entirely asphalt. I have fixed this.

Hi NAZIR-SH, and welcome to OpenStreetMap!

Please be careful when editing things. Don't trace over roads that are already on the map, this just creates duplicate features. Also check the type of road that you are adding; most of these are short driveways, not tertiary roads.

I have changed these back to the correct road classification. Let me know if you have any questions. Thanks for editing, and keep mapping.

---

(Google Translate)
Привет NAZIR-SH и добро пожаловать в OpenStreetMap!

Пожалуйста, будьте осторожны при редактировании вещей. Не отслеживайте дороги, которые уже есть на карте, это просто создает дубликаты объектов. Также проверьте тип дороги, которую вы добавляете; большинство из них - короткие дороги, а не третичные дороги.

Я изменил их обратно на правильную классификацию дорог. Дайте знать, если у вас появятся вопросы. Спасибо за редактирование, и продолжайте отображать.

Hi, is this import discussed somewhere? I can't find anything on the wiki and your source is wrong; these are not buildings.

Is way/53040848 broadleaved or needleleaved?
It was fine before, with `leaf_type=needleleaved` but you changed to `leaf_cycle=needleleaved` + `leaf_type=broadleaved`, which is an incorrect use of leaf_cycle.

Please stop deleting random features without explanation. "FXFX" does not provide any useful information about what you are doing. Why did you remove these in particular? Have you been there and seen them gone?

I had not been doing any editing in this area. I found the changeset using a tool claled OSMCha (https://osmcha.mapbox.com), which lets people inspect changesets and and helps with finding potentially problematic changes.

I'm not sure what offset problems you were having. The surrounding buildings were traced from Bing, which in this spot has an offset from Maxar of only about 1-2 feet. Any buildings that were in 'green fields' are there because they do not exist anymore. Those should be deleted.

You are correct, that sometimes it is faster to delete and recreate then to fix a badly mapped building. However this causes problems for other mappers who try to follow the history of an object, as well for anyone tracking edits in an area. It's just bad practice.