I understand that the title is very vague. I’m also not sure I’m explaining the content correctly because I don’t know the difference. If my explanation is vague or not clear enough for you to understand, please let me know and I’ll try to supplement it.
Why do some countries, mainly in the West, not attach attributes to the names of map features?
For example, why do train station signs say “Gare de OOO” (In English, ‘OOO Station’), but the information on the map says “OOO”, or why do they leave out the “-river” for things like river names?
In this case, a search for a river ‘OOO’ in the region ‘OOO’ and a train station ‘OOO’ would be the same.
On the other hand, some map features are still appended with ‘-station’. (I understand that this is used idiomatically or to distinguish them)
I would like to know why this is the case.
For comparison, at least in Korea, where I live, a river with the same name in the ‘AAA’ region is called ‘AAA river’ and a train station with the same name is called ‘AAA station’.
This is more than just a curiosity, it’s a pretty important point, and the Korean community is in a pretty big mess because of the consensus(?) in OSM not to attach attributes to place names, etc.
If we say that we don’t attach attributes to names because the tag already contains the attribute, then that means that we shouldn’t attach ‘-School’ to something like ‘school’ because the tag already contains the attribute.
For example, “Seoul Station” and “Seoul Elementary School”,
‘public_transport=station’ and ‘name=Seoul’(Don’t use ‘-Station’ because the property is already station.)
‘amenity=school’ and ‘name=Seoul Elementary’(The property is already school, so ‘-School’ should not be used.)
And because in East Asian countries I know, “OO Station” is “OO Station” and not “OO” in the name itself.
In a service with advanced search like Google, you can search differently and it will find everything, but in OSM, where you have to search straightforwardly, if you search for ‘OO station’, nothing comes up, but if you search for ‘OO’, it will search for ‘OO’ area, ‘OO station’, ‘OO river’, etc. all at once. I think this is also quite confusing and inconvenient.
I think you should fill in the official name of the place. So if the offical name is “OOO”, write down OOO. But if the official name is “OOO station” write down OOO station.
Having this standardized is important. Fixing the limitations of the osm search is a whole other issue.
No descriptive words in names is not an absolute rule. Name is what people call it, and that can include descriptive words.
The central train station of Rotterdam is actually called Rotterdam Centraal Station, appearing on signs as Rotterdam CS. Removing the Station part is not ground truth.
On the other hand, the train station of Nieuwerkerk aan de IJssel is usually just signed as “Nieuwerkerk aan de IJssel”. Adding “station” is not signed ground truth. But people will use “station Nieuwerkerk aan de IJssel” outside the context of traveling. That is where mappers will disagree about including the descriptive part.
If multiple features carry the same name, usually named after an important POI like a prominent landmark, people will usually add a qualifier. In those cases, inclusion of the qualifier in the name would be practical and verifiable as ground truth.
I think communities will settle on a basic rule with some exceptions. Exceptions could address categories, such as train stations, or specific situations, such a multiple features named after the same landmark or just “everybody calls this castle ‘Rainwood Castle’”.
Can you explain more? I could agree from “Rotterdam CS” , but I do see it using “Rotterdam Centraal”. Same for other “Centraal” stations in Netherlands? File:Rotterdam CS lage zon.jpg - Wikimedia Commons
I would agree in general for Hbf in Germany. An example from the same opposite part of world in Asia is “Taipei Main Station” using “台北車站” in MRT.
Of course, note Nangang, and Banqiao, the other 2 stations on TRA, and THSR, doesn’t include “Station”. However, the Zuoying station for Kaohsiung can have significantly conflicting signage.
Let’s not discuss the examples I gave, but the principles involved. I wanted to illustrate that there is no single absolute rule, so mappers and communities make practical choices. Usually including the perceived effects in currently dominating end user applications such as OSM Carto, Nominatim and the routing engines on osm.org. Since these tools are positioned as a testing environment for mappers, that’s understandable.
With rivers that’s different - and probably different from language to language. There’s Relation: La Moselle (390416) | OpenStreetMap. Neither in French nor in German “river” is part of the name and it is not used anyway. It’s “La Moselle” in French and “die Mosel” in German. While “La” is used in the French name, “die” is not used in the German name. That is valid for nearly all rivers in France and Germany that I’m aware of.
But exactly, I don’t understand how were the princples applied in that case. Yes, no absolute rules. But guidance and suggestions need to be better than that. How is “Removing the Station part is not ground truth.” there? That’s a very serious and critical claim you are making, from an OSM principle, which I don’t see how it is true yet.
I noticed that in Italy, the type of river is often included in the name. It appears that to Italians, it is very important if the river ends in the sea or in a bigger river! In Nederland, all rivers are interconnected, joining and splitting at random, so we don’t bother to make a difference.
But official names may suffer from the same inconsistencies as informal language, and it may be hard to tell an “official name” apart from an “official description” . The nearest large railway station to me has a huge sign over the main entrance saying “Estación de Málaga María Zambrano”. But the platform signs say only “Málaga María Zambrano”. Neither the “on the ground” rule or the “official names” rule really help here.
Because many people and signs use Station… as the name. Other people and signs don’t, not in all cases, but still, the name including Station… is found and heard on the ground.
I think in this case, leaving it out hurts more than leaving it in. If you ask a person "What’s the full name of this building/complex?"I think the answer would be Rotterdam Centraal Station. I have lived over 65 years in and around Rotterdam, so I think of myself as Ground Truth expert here!
Other mappers may think differently, of course. Just illustrating that there is not one absolute rule.
name=Wisła instead of name=Rzeka Wisła is used because actual common name is “Wisła”, not “Rzeka Wisła”. And definitely not “Wisła Rzeka” or “Wisła River”.
If it has Korean, Chinese or Japanese name, maybe it will have part mentioning its object type.
And in turn “Grzybowiecki Potok” has Polish word for stream in its name because it is apparently its common local name.
In either case actual common name should be used. Note that there are universities using different word than university and so on.
not always
sometimes official name is some bizarre monster not in actual use
official_name=Park Krakowski im. Marka Grechuty name=Park Krakowski
And for why it is inconsistent? If someone expects consistency from languages they will be typically really disappointed. Even for formal, constructed ones. To say nothing about actual spoken languages.
And this is not changing that into name goes name, not its description.
If you ask anyone what a station building is around the world, I would guess they will almost always mention “station”. I don’t see definite proof from that alone.
You need to distinguish pedestrian wayfinding signs on streets, and signs at the station itself. Outside and far away, “Station” is usually included to highlight what they are.
There are several questions possible. Can be broken down, as seen from what you mentioned.
Should a name be included
name= or official_name= , or some other *_name=
The original question in this post, whether generic form “type” viz “Station” are included
A “full name” is often official_name= only. Including “Station” might be better worded as “full title”.
As I raised, looking at signs inside the station could be more suitable. Signs on the exterior of a station still convey what they are to the outside world, as street signs do. It will be more obvious on platform name signs, line diagram, network maps, and I haven’t mentioned tickets.
On the opposite end of what @alan_gr mentioned, Rotterdam CS appears to have “Centraal Station” on largest sign. Obviously, this doesn’t mean it should be name=Centraal Station only. File:Rtd CS-III.JPG - Wikipedia
I’d like to take a moment to summarize the discussion and continue.
Of course, ultimately my question is about whether or not we should name them with attribute words, but that’s a later question, and the point of this question is to ask why some cultures don’t label their cartographic objects with attribute words.
(Judging from the discussion so far, it doesn’t seem to be a big deal, just a cultural custom…)
In any case, I’ve heard this view because it’s not only in OSM but also in Google Maps that you can see that some map features don’t have the part of the word that indicates their attributes, and the most obvious example of this is “station names”.
From your comments here, it seems that even in other countries, when referring to stations in the OO region, they don’t just say “OO”, but “OO station”, so why is it that the map only shows “OO”? Why did this convention come about, how does it differ from not saying ‘OO stations’, and what are the advantages? (The Bing translation used here is definitely a bit odd. )
But we usually have the qualifier already expressed by tags: railway=station name=Rotterdam Centraal
The railway=station part specifies that it’s a railway station, which can be rendered on a map with an appropriate symbol; in search results, it can be displayed as:
Railway station Rotterdam Centraal
Or, if you prefer another language:
Rotterdam Centraal bahnhof
Overall, the qualifier can be inferred by both software and user, and explicitly including it in data can be harmful - it clutters the map and may be redundant in search results.
Personal opinion would be to include in the name=* tag the most used/spoken name of the feature and in official_name=* however it is signed/officially referred as. This way, you would be ok with the “no descriptive names” rule, which as some already told, it’s not an absolute rule anyway.
I’m not sure the comments here lead to that conclusion. I’d say the overall message is that it depends on the context and it’s hard to distinguish between names and descriptions.
There are relatively rare exceptions, such as Gare du Nord, where there is really no way to refer to them without the attribute. But in general “it depends”, e.g.
If asking for directions on the street, I might use not just “station” but “bus station” or “train station” to ensure I am directed to the right form of transport. That doesn’t mean I think those attributes are part of the name.
When displayed on platforms, tickets, and timetables, I would normally not expect to see “Estación” or “Station” (with the noted exceptions).
Similarly I would never say “Estación de…” when buying a train ticket. It’s obvious that I am referring to a station.
Whether the attribute is included on signage at station entrances seems to vary wildly between countries and even within countries, and can’t be taken as a reliable indicator of the station name.
Direction signage is even less reliable and may simply say “station”. (There are still some old signs to be seen in Spain pointing to a mysterious place called RENFE, from the period when RENFE ran both the trains and the stations).
In normal speech people may include an attribute if there is ambiguity, e.g. there is a train station near me called simply “Aeropuerto”. It is not, in fact, an airport.
In summary, real language is messy and the OSM conventions are an attempt to achieve some kind of order. I wouldn’t read too much into it about cultural conventions.
For what it’s worth, in Spain there have been long debates about whether the names of schools should include abbreviations for the type of school (“IES Don Quijote”) or be spelled out (“Institución de Educación Secundaria Don Quijote”). But there seems to be general agreement that it is OK to include one or the other in the name, even if they are arguably attributes. Perhaps that is partly because common OSM tags don’t map neatly to the types of school indicated by these abbreviations, but that’s merely a guess.
Others have made similarly good points, but there are a couple things in your comment that particularly caught my attention and I wanted to respond to them.
(Please understand that I am not commenting because I find something particularly wrong with your opinion.)
If I’m translating and understanding your intentions correctly, I agree and applaud your perspective.
I think you are trying to be principled, focusing on objective factors rather than subjective factors, and trying to be principled.
But I’m also sorry to say that your logic is where my questions begin.
it seems that people feel differently about whether the attribute of ‘station’ is part of the name. If I say ‘Seoul’, would anyone among you think of ‘Seoul Station’? Maybe some people would think of ‘Seoul City’ when I say ‘Seoul’. In that case, ‘Seoul’ can be the name of a city, but it is difficult to see it as the name of a station.
I think it’s a different matter whether the attributes are explicitly and customarily used together. If I say ‘Seoul’ or ‘Busan’ in ‘Seoul Station’, you might hear it as a station name, but I don’t think anyone would think of ‘Seoul Station’ if I said ‘Seoul’ in New York.
I believe that signs are minimally formal, and therefore the names on them should be viewed as formal names. However, different signs are often labeled differently depending on the nature of the sign, which is where the confusion comes in.
while some cultures call map features by their names without attribution(If you don’t say ‘Gare de Metz’ and just use ‘Metz’), others almost always do so with attribution, i.e., in situations where it’s clear that it’s a station (train tickets, signs inside the station, etc.), they may not attribute it, but in most situations the name “Seoul” would never represent “Seoul Station” (at least in my country).
)
