I also want to apologize for being to harsh. I saw your post and interpreted it as you being elitist, rude, and trying to shut down all discussion because you’ve been here longer and talked about it on mailing lists. Most of it came from “discord kiddies who are unable to use google”
2 Likes
I don’t see why railways don’t use lifecycle prefixes like everything else and get treated the same way as other objects with razed or demolished lifecycle (deleted as necessary). OHM has a new rail layer if you are looking to map what used to be present.
7 Likes
Disparaging a portion of the community for using their preferred communication platform isn’t in the spirit of the project either. A person came her to discuss things. You don’t have to berate them for using Discord. Definitely doesn’t set a good tone.
If there’s good pointers to previous discussions just post them and move on. Everyone hasn’t seen everything. Sometimes things change and we should discuss them again. Etc etc.
21 Likes
An old railroad grade has a distinct look vs a general embankment. These can serve as important landmarks when navigating in the backcountry, e.g. “after about a half mile you will come to an abandoned railroad grade…”
However, if there is no evidence of a railroad having existed at a location, I wouldn’t map it, but I also wouldn’t delete it, as someone with a keener eye than I may have some on-the-ground evidence that I missed.
6 Likes
There is a difference between “planned” and “proposed.” If something has been approved and funded by some government entity (in other words, it is “planned”), and there is official documentation to that effect, then I think it is ok to map. On the other hand, to me “proposed” means means an idea has been submitted, which may or may not get approved and implemented. In fact, there may be multiple concurrent competing proposals that the authorities are considering. Given the uncertainty around a “proposed” feature, I would say they should not be mapped in OSM.
1 Like
That doesnt have to be tagged as railway=razed, perhaps you could add some embankment tag that signifies its a former railway embankment. Also is the differences from a former railway embankment significant enough to not just be tagged as an embankment?
2 Likes
So far there hasnt been much support for the status quo usage of railway=razed and dismantled. Is there anyone who thinks that nothing should be done and why? Also does anyone want a different replacement proposal for those tags?
I think that a few people here misunderstood why should we deprecate railway=razed and railway=dismantled. The reason they should be deprecated is a piece of historical/outside data is required order to be certain that razed/dismantled railway exists. You cannot verify a razed railroad by surveying, which is contrary to the project verifiability rule and cause a lot of debate – the OSM discord just had a heated discussion by those that were adding nonexistent razed railways by “scars” on farmland. These useless features also clutter up the editing interface and make it very difficult for people to tag these old railroad tracks as highway=track or highway=path.
Razed and dismantled railways currently on OSM would also make a lot more sense if old stations and infrastructures are present on the database, but we map these demolished features on OHM not OSM, and this is also the reason why razed/dismantled railways should belong at OHM. If it is not controversial to remove demolished train stations on OSM, even when considering lifecycle tags, why then it is so controversial to remove these railway features? Perhaps this debate still exists by a small minority of editors who will always refuse change to their mapping habit, but that’s beyond me.
In the same vain, highway=proposed should also be depreciated, but we will not get into that in this thread for now.
9 Likes
I understand that, but I think that mapping those with railway tags is wrong. I’m not saying that those shouldn’t be mapped, but think that they should be mapped with an historic tag.
No, that’s not wrong.
This is
highway=cycleway
railway=abandonedaccording to Railways - OpenStreetMap Wiki
5 Likes
The wiki isn’t gospel; a potential outcome of a discussion on this very
forum could be a change in the wiki pages.
Personally I am in favour of getting rid of these tags because they
invite abuse; I’ve recently done a quick database query in Germany to
find all razed/abandoned railways that run under buildings (!) and found
almost 20k spots where this was the case; while there were some false
positives (bridge or railway maintenance buildings), in most cases the
railways seemed to have been completely built-over with no trace
remaining except in the heads of those who used to live there when there
still was a railway. And that is clearly wrong, but it is something
that people think they can do because these tags exist.
12 Likes
If that is the case then the title of this topic is misleading. What you have written is “Proposing to deprecate railway=razed and railway=dismantled”.
As I pointed out, deprecate means replace one tag with another. I think you need to have another read of Deprecation process - OpenStreetMap Wiki to work out what you need to be doing.
2 Likes
So you are saying that a group of mappers work done over the last 15 years should be wiped out because you don’t like it?
2 Likes
and found
almost 20k spots where this was the case; while there were some false
positives (bridge or railway maintenance buildings), in most cases the
railways seemed to have been completely built-over with no trace
remaining except in the heads of those who used to live there when there
still was a railway.
I looked at a few of these hits you provided and often it was clear from just looking at the map where the railway had been, because of current features like road layout, tree rows and similar, also names, and even if there were some spots where the former line was intersecting with buildings. In the German discussion we spoke about railways completely built over by residential quarters or a supermarket but in the map you provided it was much more often the case that these intersections of longer lines were at a few spots while generally they were not overbuilt.
Also former tramways in cities are often cited as abuse (and I tend to agree), but these are very different from heavy railways outside built up areas, and it is less likely to find traces of former trams on the current surface. Usually there weren’t tunnels or embankments built just for trams, they had much fewer impact on the landscape.
4 Likes
(Different person)
In cases where there are no remains whatsoever?
Yes, these features definitely should be removed from OSM, no matter how much effort went into mapping them as these were never wanted nor welcomed.
Effort alone does not make something worthwhile.
(Yes, I removed several railway=razed/dismantled/abandoned where no trace remained, one of cases was where people mapped railways replaced by open pit mine that obliterated any traces of railway and entire village where it was located)
Note that such mapping is welcome on OHM.
And no, just because in some cases problem was not spotted or people were scared to fix it does not mean that it should not be fixed.
(if there are some in-place remains more substantial than “town has Railway Street” then it is trickier but many cases are straightforward)
PS: case where feature was recently removed and is visible on aerial imagery may fall under exception of keeping recently gone features
5 Likes
I think this is wrong too. We don’t tag castle as castle=abandoned we use historic=castle. Why can’t be repurposed or left behind railway structure be tagged with historic? I think this is a tag that could help us keep information while giving the right information too: one here there was a railway, what’s left behind is this.
Deprecation often means removal of something with no replacement whatsoever.
Ironically, that wiki page may benefit from making it more clear.
edit: to be clear, I mean “deprecation” in general use, not limited to osm context
2 Likes
Deprecation often means removal of something with no replacement whatsoever.
as this is “often” you surely can provide a long list with examples, or at least some?
Hi,
So you are saying that a group of mappers work done over the last 15
years should be wiped out because you don’t like it?
In so far as this work has gone against OSM’s “on the ground rule”, i.e.
recording things that cannot be verified on the ground, yes, but not
because “I don’t like it” but because “OSM doesn’t like it”.
If such data is in OSM, then it is there because (in the best case) of a
misunderstanding and (in the worst case) of a willful violation of the
“on the ground rule”.
Note that I am speaking about railways that have been over-built with
other developments, and nothing remains on the ground of the original
railway. Such data never had any place in OSM.
If someone has spent a lot of work adding such data, it appears that it
is relatively easy to move it to “OpenHistoricalMap”, a project that
complements OSM and has specifically been created to record information
about things that may have vanished. So the work isn’t lost; it’s just
that there’s no place in OSM for information about things that are not
there anymore.
I acknowledge that there are many cases where someone with “an eye for
it” can identify a former railway line from on-the-ground cues and this
is a different scenario. In these cases, there is room in OSM for the
things that can still be seen, and sometimes there is discussion about
just how much there needs to be seen to justify some form of railway tag
(i.e. is this just some kind of embankment I see there, or is it a
former railway line?).
Bye
Frederik
10 Likes
I will lock this page for a period of a day in order to both lower the tension I sense, and provide people the chance to collect relevant examples from OSM data that will help move the conversation forward.
I am hoping this thread will evolve into a very robust, detailed, and respectful conversation after the lock ends.
6 Likes