Proposal talk:Deprecating demolished railway tags

Information loss

Never, for now. This proposal is about erase specific historic information inside OSM. OHM is not OSM. Why we should lose this information? Why we won't be able to understand the route about that special cycleway (in Spain, for example we have 'Vías Verdes' we have a lot) with stations because some people wants to remove all this important information that help us to understand it? Who assure us the transfer of the information to an other external project? Have you seen the "present" of the map you mentioned? Here it is NOT Madrid,the capital of Spain , an European city , capital of an European country. Here it is Berlin. Nowadays, for now, (I hope in a future would be a compatible project, like wikidata, but not for erase the data) is a failed project (there are a lot of reasons). The only way you can assure all this data is keeping it on OSM, not outside of it. --Yopaseopor (talk) 06:06, 26 February 2024 (UTC)

Totally agree with that; contrary to other historic data, which have their place on OHM, old railways data often explain current structures, such as long cycleways, so maintaining them in OSM still have sense, despite their historic aspect. Edit: besides, given the ongoing debates on this issue, I'm afraid this proposal will fail to reach enough approvals to be put in effect. Basically, I'm afraid that you're losing time and effort. ;-) --Penegal (talk) 07:21, 26 February 2024 (UTC)

If something, like old railway stations still exist - they they can be mapped in OSM. Where something no longer exists then it should not be in OSM. People continue to map non existent things .. I have had them cross existing roads with railway embankments, that not longer exist, making the road look broken. Many old historic things that could explain present objects shape/presence exist .. but OSM is not the place for it. Warin61 (talk) 09:29, 26 February 2024 (UTC)

"This proposal is about erase specific historic information inside OSM. OHM is not OSM. Why we should lose this information?" Simple answer to that question. It is historical information and not current information. OSM it not OHM. If the thing does not exist there anymore then it has no place on OSM. OSM is a map of current things. Utterly removed railways are not current things. I've had to deal with this locally to me where someone insisted on putting bridges that were removed 50 years ago and an embankment which went at around a similar time into the current OSM map. Oh in between those bridges there is most definitely a prominent viaduct still in place, so that's tagged appropriately. It physically exists and it is obviously railway infrastructure. The bridges over the roads on the other hand? Completely demolished decades ago and therefore have no place in OSM. Davidpnewton (talk) 19:01, 10 March 2024 (UTC)

Historical information is current information. Why exist a tunnel in the middle of nowhere in a "natural" (better called outdoor) cycleway? Because it was a railway. Can you determine the nowadays inexistence of Berlin Wall? Can you explain the existence of pieces of railway in the middle of Spain without the abandoned railways?

Completely demolished decades ago but Oh surprise someone tries to add it again and again based in old information. If you have the historic information inside OSM you will not have this problem.

Forget OHM, it is not about OSM so any data of OSM by now can be deduced to finish there (if OHM does not change at all). I don't want to lose information. Do you? --Yopaseopor (talk) 14:04, 18 March 2024 (UTC)

History is the past, not the present. A new bridge replaces the old bridge, if the old bridge is gone then it should be gone in OSM. It is not the function of OSM to keep 'historical information'. OSM records things that exist today, including things that record old events such as Node: Battle of Rorke's Drift Memorial & Cemetary 1879 (95643816) but the feature has to have a physical presence today. If it is gone today then it should not be in OSM. Warin61 (talk) 09:32, 19 March 2024 (UTC)

Sorry, but I'm not agree with you. History is present, as you can see in every museum and every monument (why does the humanity make monuments or memorials if not to remember someone, somewhen, something?). You can remember whatever you want in the way you want. About the functions of OSM...read the main page of the wiki: "Welcome to OpenStreetMap, the project that creates and distributes free geographic data for the world" (it not says anything against historic information) . "We started it because most maps you think of as free actually have legal or technical restrictions on their use, holding back people from using them in creative, productive, or unexpected ways. " Are you trying to restrict this like other maps? . Or in About Openstreetmap "The OpenStreetMap License allows free (or almost free) access to our map images and all of our underlying map data. The project aims to promote new and interesting uses of this data" (why historic and existing information would not be an interesting use of this data?). --Yopaseopor (talk) 12:01, 19 March 2024 (UTC)

I repeat "history is the past" .. meaning what occurred before today. An old railway line that is removed and now has houses, shops, car parks and roads over it should not be in OSM. Existing things should be in OSM such as present day museums, monuments and existing railways. From the OSM wiki 'Map Features': OpenStreetMap represents physical features on the ground. A removed railway line is no longer 'on the ground'. Warin61 (talk) 08:00, 20 March 2024 (UTC)

"History is present" and currently present features can be mapped! Problem starts when some people want to map no longer existing features which left no identifiable traces Mateusz Konieczny (talk) 17:25, 20 March 2024 (UTC)

I don't understand why it would be necessary to remove this data. They don't appear, but they're still geographical elements that can be used in research (for example, to follow the evolution of these railways). On the point of verifiability, it's very often local and historical knowledge. It's not verifiable by definition, because the more time passes, the more difficult it will be to verify again, but it is possible to check with other sources whether these lines really existed. --Sylv1_Durif (talk) 10:54, 26 February 2024 (UTC)

After some research, (talk/2022-October seems to describe the problem well) I think people should be encouraged to stop mapping this tag while keeping their page in the wiki (there will always be some who will do it and it doesn't matter, the majority will stop) and in the same way, I don't see any harm in deleting it when you come across it during your mapping (just let people know that it's not useful and poses many problems that are close to grief). --Sylv1_Durif (talk) 10:54, 26 February 2024 (UTC)

Features gone without trace are not mappable in OSM (except short period after feature is demolished but for example visible on aerials). Long gone features without trace in terrain are not mappable in OSM, should be removed and will be removed from map data Mateusz Konieczny (talk) 12:39, 26 February 2024 (UTC)

Maybe Wikidata and Wikimedia Commons is better place for it. See:

https://m.mediawiki.org/wiki/Help:Map_Data

https://m.wikidata.org/wiki/Wikidata:Map_data

Something B (talk) 22:40, 26 February 2024 (UTC)

Counter-proposal

In my opinion, this proposal is basically flawed: as said in past discussions, ex-railways are frequently important large structures in the landscape, but we lack an agreed way of representing them in OSM. Removing them is not a good approach as we would be losing valuable data in OSM.

Some time back I described a way of mapping an ex-railway as a site relation, in detail, in a commento on a wiki page (Ex-railway_and_cycle_route_relations_-_an_example

Instead of repeating again the same argument here, I kindly invite you to read that old quasi-proposal.

In particular I invite you to also have a look at the satellite imagery of the example ex-railway, which speak for themselves.

--voschix (talk) 11:17, 26 February 2024 (UTC)

Discussions

See https://community.openstreetmap.org/t/rfc-feature-proposal-deprecating-demolished-railway-tags/109757 and https://community.openstreetmap.org/t/rfc-feature-proposal-deprecating-demolished-railway-tags/109757 for discussion about this on forums. Mateusz Konieczny (talk) 12:42, 26 February 2024 (UTC)

Map what's on the ground

I recently noticed an abandoned railway in the map near my house. I organized a hike to go discover it, maybe take some cool pictures. When I got there, there was pavement, buildings, sidewalks, shops, and exactly zero traces of an abandoned railway. A map that tells me "a railway is here, in a state of abandon" is plain wrong. (Of course, when I got home I corrected the map.)

"Penegal" argues that "old railways data often explain current structures." But, how does this argument account for current structures are explained by former structures that were not railways? If the fence around a plot of land matches what used to be the river edge a hundred years ago, would you argue that we should tag the area as "river=rerouted" to explain why the fence is there? Why should we make an exception and keep non-current data only for railways, and not for the rest of history?

Worse even: If I was lost in the woods and the map tells me that I can follow along a railroad, but there's no railroad there anymore, the map will make my situation worse by having me look for signs of a railway that is not there. This is a concrete, real-life scenario in a forest near my place, where there are plenty of tracks, all similar to each other, and only one of them corresponds to a historical railroad dismantled 80 years ago (which was still in OSM until I deleted it some weeks ago).

A train station building, still there, should remain in the map because it's still on the ground. On the other hand, the fact that it was a route stop for a railway that ceased operations in 1934 is not a reality today, it's not part of local knowledge, and is just a historical fact. So I am planning on deleting the corresponding OSM relation.

OHM data-archeologists are welcome to browse the history of OSM to scavenge for deleted railways. Perhaps a technical improvement could be for OSM to automatically upload deletions of some features to OHM, eg. when the changeset has a "#not_on_the_ground_anymore" tag or similar.

In my view: Keeping things in OSM that do not exist as of today goes against the goal of OSM.Dani CS (talk) 13:30, 26 February 2024 (UTC)