1 & 4: I don't doubt that ALKIS is official from the government (that doesn't make it correct) or that you managed to use Strg+C correctly (instead, I highly doubt that someone would call the place Glücksburgweg when there is a Weg called Glücksburg at walking distance to the north).

2: okay so others made the same mistake, is that a reason to continue?

3: not all points are historic, I'm sure many of them are also current, but you said yourself (and I agree) that, without local knowledge, it cannot be said which belong in which category.

Hmm, dann werde ich mir das morgen noch mal anschauen. Ich errinere mich nicht Bepflanzung gesehen zu haben, und bin mir sicher dass ich das untere Pfad genommen habe. Fortsetzung folgt :)

Thanks physi for the response and thanks Tux for the links!

--- Tagging scheme ---

I see the community is quite divided about whether these are useful, and even less sure about whether place=locality is a good tag. Adding these locality=ALKIS_Lage tags until we found a good tagging scheme, and then mass editing them later, sounds okay to me as a tagging scheme.

Of course, this assumes that you then also make sure a suitable scheme is found (because you are the one that wants to add this), and clean up the temporary tags afterwards. Otherwise the temporary tags will just stay there forever. So far, I am not seeing a lot of participation to make this happen, but I do see you are continuing to add nodes on the map for more than a year since the discussions started.

--- Currentness ---

I see there is a post from physi in one of the threads, saying:

"Inhaltlich bin ich der Ansicht, dass die Eintragung einer historischen Flurstück- oder Gewannbezeichnung dann sinnvollerweise erfolgt, wenn der Name noch gebräuchlich oder bekannt ist. Dazu sind aber sehr lokale Kenntnisse notwendig."

This makes sense to me, so then we should stop copying these names from a database and instead use local knowledge, right?

The ALKIS database might contain these labels in some "current" layer because it is the latest info it has available, but that doesn't mean people on the ground call it by this name anymore. It just means that ALKIS users might use the name to refer to the place. Copying labels from a database like this, without any knowledge of local use, seems similar to tagging what3words labels as localities. Maybe someone uses it, maybe not.

In another changeset discussion (changeset/122333151) you wrote "Frage der Gebräuchlichkeit halte ich dagegen für relevant bei der Darstellung dieser Informationen". This is not true: the renderer can't choose that old names are irrelevant, because it's not labeled as an old name. If a place is only formerly known by a name, then the name=* tag is simply not accurate. For example Westerwald (node/9815156407) placed outside of a Wald... whether that Feld is still called Wald, is something only locals can say.

--- Accuracy ---

You haven't commented on the issue I raised above, where Camp Astrid allegedly is a locality called Glücksburgweg by the locals, while in reality there is a real Glücksburgsweg (and bus stop called Glücksburg) only a few minutes walking to the north. Not being from Germany originally, I cannot say whether the names near me are accurate, but this is one example where it seems clear that this ALKIS data placed a label in the wrong place.

I don't know what the threshold is before OSM considers a data source reliable and suitable to copy from (even if there are, of course, occasional mistakes). It might be good to investigate ALKIS' currentness and accuracy before using it.

--- Specificity ---

Labels such as "An dem Rothenbach" (e.g. node/9816379762) obviously refer to an elongated area, namely the length of the brook called Rothenbach. Adding a locality here rather than a name or old_name to the brook would be more correct and help users a lot more to find something that is described as being "by some brook". Picking a random point and calling that place "by the brook" is not very specific. Since the brook currently has no name, if this is the brook that ALKIS refers to, then adding the name there would be a useful addition to the map!

Similarly for the Eschweiler forest, where the Stadtwald is allegedly also called Gemeindewald, it would make more sense to add the loc_name to the existing Wald object (relation/106123) rather than adding a duplicate object to a few locations that are seemingly chosen at random (or for/by the ALKIS renderer). I would see this as a useful addition!

--- Duplication ---

"The repetition in some cases is due to the large area of that object. It is comparable to a long street which also uses the street name several times, not only once."

That's because they indicate a defined region, from point X until point Y this road is called Hauptstrasse, and for technical reasons the way object needs to be split into multiple segments (to have different tags for different parts). This is not the same as adding duplicate localities for the same locality.

If there are adjacent, duplicate locality names shown in ALKIS data, and it is known to be current and accurate, then one should add a single node at a suitable central point, rather than adding multiple nodes for the same place.

(Update: while I was writing this, I see you added more changesets, and this one for example with even more duplication than I saw in past changesets... changeset/122553475 Please stop that, I really don't think this is correct.)

--- Community ---

Working together always requires communication and discussing with others.

Aixbrick requested 10 days ago: "Bitte höre vorerst auf, Gewannnamen zu ergänzen." (changeset/122076044)

Vademecum said: "Sich der Beteiligung an der Diskussion zu verweigern, wenn es um einen von dir erfundenen strittigen Value geht, finde ich im Rahmen eines Gemeinschaftsprojekts bedauerlich." (same changeset)

flohoff wrote 3 days ago: "Lagebezeichnungen zu übernehmen die typischerweise ausserhalb des Katasters niemand benutzt macht IMHO keinen Sinn." but asked for discussion (https://forum.openstreetmap.org/viewtopic.php?id=75785)

I think I also raised a few points that should give pause regarding adding more of these names from ALKIS.

The OSM community can be conservative (particularly regarding imports) and we have to be weary of ossification or excessive bureaucracy. In this case, multiple people question the source (whether the data is good at all) and everyone seems to agree that the tagging scheme isn't good either. The latter is okay temporarily, if it can be machine-edited later, but more than a year is not very temporary.

Because of the questionable source quality, combined with repeated requests from different people:

you should stop adding more of these names, at least until the data source's quality is known (then we can still see about the tagging scheme and mechanically edit it).

Looks good, thank you!

Please see osm.wiki/Keep_the_history when updating objects on the map. This edit destroyed a reference on the german osm wiki, see "[1]" on osm.wiki/DE:Tag:place%3Dlocality

I just noticed another questionable locality name: Glücksburgweg (node/9805658383). That sounds like a street name to me, and indeed there is a street "Glücksburg" just to the north. Do people really call this field (that you placed the node into), or the greater forest or Camp Astrid area, "Glücksburgweg"?

Hmm, I'm seeing three places in the same forest all called Gemeindewald. That seems duplicate to me, considering this best practice: osm.wiki/One_feature,_one_OSM_element
Has this issue been discussed in the (german) osm community somewhere?

I've also got doubts about adding places allegedly called "at the little street". This is not at all unique for a locality (osm.org/search?query=Am%20Str%C3%A4%C3%9Fchen) and there are also many real streets called as such (e.g. way/186657234). This will just muddle regular address searches and I wonder if anyone really calls a locality such a name. How do you see this?

Hi! I see you named this forest "forest". However, the 'name' field is meant for names. Setting the name of this forest to "forest" is a bit like calling a person like me 'human' instead of 'Luc' :D

pfad existiert schon? Ich hab da gestern noch gelaufen und es hat auch seit dem Fluten (von letztes Jahr) immer existiert

way/1062201155 This way should probably be 'service' rather than 'unclassified', as per the wiki "highway=unclassified is used for minor public roads" (this isn't a public road, it just provides access to boat parking which fits the definition for highway=service). The name 'unclassified' is a bit unfortunate and the meaning is based more on practical use rather than that a good name was chosen initially.

Edits look pretty great in general! :)

...ariable.

Thanks, iD, for warning me that I'm over the changeset comment length limit

Hoi, thanks voor je bijdrage! Het voetpad was wel toegevoegd, maar het zuidoost uiteinde (node/7917740559) was niet verbonden met het fietspad terwijl het er wel heel dichtbij lag. Ik ken de "Go Map" editor niet, normaliter zou dit niet moeten kunnen eigenlijk. Ik heb het nu verbonden; misschien iets om op te letten in de toekomst :)

"barrier is a wall not a hedge", that description should have said

"the Address here is documented at the Building as a way btw"

Oh, damn actually I missed that. Mapping in the netherlands a lot in the past, I'm used to it being all separate nodes from the government address import. Makes sense then, thanks!

Hey, thanks for adding the restaurant! Question, though: is there any reason you didn't add the tags to the node that already contains the address? Now the icon competes with the address for being displayed on maps, and finding the address programmatically is more complex. But that's just my perspective, I don't know if there was a specific reason you structured it this way.

feel free to edit if you think the tag should be different, I might very well have selected the wrong forest preset

Oh, that's quite confusing but okay, thanks for catching this flother! It might also be good to change it on Wikipedia, since that's still saying it's a municipality and that's where I got confused. Do you have a link to something like a news article or this combined population stat, something that could be used as a source on Wikipedia?

Reykjanesbær is a municipality consisting of Keflavík, Njarðovík and Hafnir; it's not a city. The region is already defined by boundary=administrative (relation/8467955). Not sure we need the extra node at all, but it seems to be actively maintained so I figured maybe update it rather than removing. Feel free to comment if you don't think this is correct, it's the first time I edit a place=* :)

Next to Grindavík, you mapped a peak called Húsafell: node/4363604803
Is there a reference for this being called that? When looking at Wikipedia (EN, IS) or various websites, Húsafell is a town further north, I never see any peak mentioned. (It keeps having me confused because the peak is the default result in Nominatim, but of course if the peak is really called Húsafell then the osm search engine is just doing its job.)

Looking at the previous discussion, specifically maps.is which is now map.is, it's labeled as Húsafjall instead. Was it a typo?

Okay so I assume that means coordinates and not other tags like house numbers. Thanks for responding, then I know I can safely remove these after checking the feature's position and outline :)