From OpenStreetMap Wiki
Jump to navigation Jump to search
Proposal status: Draft (under way)
Proposed by: Hungerburg
Tagging: highway=scramble
Applies to: way (way)
Definition: Sections of a footpath that need use of hands to get along.

Draft started: 2022-11-27
RFC start: 2022-09-14
Vote start: 2022-11-09
Vote end:


It is proposed to create the tag highway=scramble as a base tag for sections of pedestrian paths, that cannot be walked throughout their course, where use of hands is required, be that for keeping balance or be it in order to advance.


The proposed tag is intended as a means to refine existing mappings. The conflict in scope with highway=path|footway is deliberate: Highway=scramble covers a very minor subset of what currently gets mapped as "path|footway", namely, their scrambling sections. In a way, it can be conceived of as a companion to highway=steps.

Many data consumers that have no interest in non-walkable paths evidently have been failing to filter those out for years now, either for lack of a means to tell them apart or due to lack of controlling for ambiguous attributes. Instead, consumers should actively select for scrambles. This tag will make it easy for them and for mappers, that are aware of this unfortunate failure on both sides.

Unlike with rendering, initial impact on routing may still be low: Both OSRM and graphhopper already include ways tagged highway=scramble in their directions, when the scramble is part of a hiking route relation, OSRM does so even in its bicycle profile.

Personal note of the author of this proposal, who is not afraid of scrambles: I consider scrambling a mode of transportation sufficiently distinct from walking, cycling and horseback riding, so much that it deserves a separate value in the highway key. A bit more so than stairs even.


Scramble here is understood as the section of a route, between where the scramble begins and where it ends.

Mappers likely will opt for tight mapping, much like we do with highway=steps now, where a "path" gets interrupted by scrambling sections: this is recommended, when scrambles are nicely delimited and spaced distantly. Mappers may opt for a more general approach, where the scramble extends to the last scramble in a succession or even the next notable POI (a hut, a summit, a branching node): this is agreeable, when scrambles are loosely delimited and spaced closely, respectively the POI not too distant. Location affects choice as much as zeal, so the proposal does not try to impose a single scheme.

A scramble starts where walking ends. The tag is meant to be used on footpaths that turn into a scramble, as are typically found in rocky terrain. Further advance needs use of hands, even if just for balance. (See pictures below to assess.) Mere occurrence of handrails or ladders on an otherwise walkable path, though engaging use of hands, does not make the highway a scramble.

To quote from the famous definition of waterway=stream, the distinction between walking and scrambling is to be based on where an active, able-bodied person, with head for heights, suitable shoes and dress is very likely to use hands for balance, comfort or progress.

A scramble ends where climbing starts. Not necessarily on location, but in theory at least. Endless discussion would ensue, what constitutes "climbing", if scramble was not by definition a subset of what currently gets indiscriminately mapped as a "path": so UIAA grade II sections will not disqualify (these are covered by the highest value in the approved sac_scale key), while UIAA grade III will. In a nutshell: Non technical terrain, no ropes and other gear necessary.

The source of the information if to walk or to scramble preferably is local knowledge, own hands on, trustworthy notes, accounts. As of today, 20,000 paths are tagged sac_scale=*alpine_hiking and thus contain scrambles by definition. It is not proposed to perform any kind of mechanical edit; tagging may be incorrect, or if correct, may not map a scramble, but a glacier traverse e.g.

Mappers are reminded to only map scrambles where on location signs of actual use by humans other than the mapper can be observed.

There exist a number of attributes already, that are useful in combination:

JOSM users can install a preliminary custom preset "Scramble" from the editor's templates list, here to the source code


Scrambling on Pike of Stickle - - 2053532.jpg


Prior art


Routers and Renderers may choose as they deem fit for the public they address.

Even the best pedestrian routers of today, that account for incline at least, fail with their duration estimates on routes that contain scrambles. They cannot be blamed though, how can they tell?

Renderers may want to use POI like icons to display scrambling sections in overview zooms, so unsuspecting users are able to spot them rather sooner than later.

Features/Pages affected

An article to document the tag, worded from text of the proposal will be created; articles on compatible attributes will link there.

Related articles

External discussions


Please comment on the discussion page.

Suspended Voting

  • I approve this proposal I approve this proposal. High time scrambles emancipate. --Hungerburg (talk) 21:55, 9 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. /While scrable=yes would have been backwards compatible, avoiding a breaking change, I don't want to hold up this proposal./ I changed my vote to no since I feel the breaking change is too much and it's already covered by sac_scale=demainding_mountain_hiking, or a potentially new scramble=yes tag. --Aharvey (talk) 22:27, 9 November 2022 (UTC)
@Aharvey: I know from previous talk that you use sac_scale incorrectly because alpine_hiking sounds too much over the top in your ears. The proposed tag though is not meant to play games with or appease routers and renderers, it is not the least conceived to provide a simple distinction, that can be much more reliably tagged, one that plays a prominent role in a number of grading schemes from different places around the world (SAC, CAI, YDS) and is less bound to the location it stems from. --Hungerburg (talk) 23:11, 12 November 2022 (UTC)
  • I approve this proposal I approve this proposal. -- Something B (talk) 22:38, 9 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Eerib (talk) 23:00, 9 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Tekim (talk) 23:01, 9 November 2022 (UTC)
  • I approve this proposal I approve this proposal. If adopted, existing path features would be converted to scrable features. This means they'll likely fall off render. I'm okay with this, because once you're climbing, you can't really call it a "path" anymore. There will be an implication for routers, but I also think that's okay because of the implication of routing someone over a scramble. This is a rare case where I think the breakage that might result from data consumers is a good thing that should prompt data consumers to respond appropriately or cause acceptable breakage if they do not. --ZeLonewolf (talk) 23:09, 9 November 2022 (UTC)
note that once you are climbing it won’t be a scramble either—Dieterdreist (talk) 00:06, 10 November 2022 (UTC)
This might be just an English language thing, but my working definition of "climbing" is broad enough to include scrambles. I would definitely say things like "climb up that pile of rocks" even if it's a scramble. --ZeLonewolf (talk) 01:32, 10 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Sound proposal. Diacritic (talk) 23:52, 9 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I don't like the idea to add a new type of highway=* for every nuanced use case. IMO this falls into the same category as highway=golf_cart_path or service=driveway2. The func'm ntion of those path are transportation by foot therefore highway=footway with appropriate sac_scale or eventually with a new scramble=yes should be sufficient. --Mashin (talk) 07:42, 10 November 2022 (UTC)
The whole point is, that it's not sufficient to use your foot and this argument also leaves the question why we have at least 15 different highway categories for transportation by car --Luzandro (talk) 09:05, 19 November 2022 (UTC)
@Mashin: I take this as a blanket permit to replace the image on Highway=footway with the image of people scrambling shown above :) --Hungerburg (talk) 21:00, 10 November 2022 (UTC)
@Mashin: Comments like this are water on the mills of people, that map true climbings routes as paths/footways with appropriate climbig:grade:uiaa=7. After all, even climbers are mostly using their feet, to get from A to B. --Hungerburg (talk) 22:54, 16 November 2022 (UTC))
  • I oppose this proposal I oppose this proposal. Per Mashin. Also, I have serious questions about if a "scramble" is actually a thing that can be verifiably mapped or not. Like one of the examples of something that I assume is a scramble is of someone Using their hands for balance, which for lack of a better way to put it is just ridiculous. If a scramble is a path requiring people to use their hands to pull themselves or similar, maybe, but people will use their hands for balance on any number of random paths. Including ones where other people wouldn't need to. That wouldn't be as much of issue if this didn't directly compete with highway=footway. Really though, it should be something like hands=required, hands=optional, whatever tag. Or maybe just a refinement of one of the 15 other tagging schemes that already exit for this type of thing. Since it really comes to down to the individual hikers ability to climb a path without needing to use their hands then it does something intrinsic to the path itself. Adamant1 (talk) 08:55, 10 November 2022 (UTC)
@Adamant1: The language you ridicule is taken literally from the worlds most reknowned difficulty rating schemes for mountain hiking. All the people in the field will instantly recognise this. It seems you are not familiar with the subject. This shall not deprive you of your right to disapprove. Note, that the proposal explicitly excludes such cases as you mention, where lay persons may wrongly assume use of hands. --Hungerburg (talk) 20:25, 10 November 2022 (UTC)
@Hungerburg: I'm not really sure what your talking about, but nothing I said was to ridicule anything. The proposal literally gives examples of a scramble with someone using their hands for balance, which obviously isn't a verifiable criteria since everyone has different abilities to climb hills. I don't see anywhere in the proposal where such cases are excluded either. Maybe that's just on the vague way the proposal was written. Either way though, whatever miss-understandings occur about what the proposal does or doesn't cover is on the person who wrote it though. It's rather ridiculous to claim I'm somehow mocking the proposal when your the one who failed to write it in a concise way from the start. Maybe don't take things personally next time. It looks like other people have the same issues with it that I do. Adamant1 (talk) 05:49, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Carnildo (talk) 09:12, 10 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Excellent and balanced proposal. With this wording and the examples I think 98,5% of mappers will agree on most cases, and nobody will map an easy path as a scramble! --Peter Elderson (talk) 09:22, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Sorry :( I really tried to convince myself. But as a hiker: there are many cases where it is impossible to define it well and scrambling status depends heavily on person, weather and so on. And it seems far more subjective, divergent and problematic than stream vs river. It would be survivable as property, but not when it is top level tag. It is common to see some able-bodied people scrambling and some not, on the same section of way. And I disagree with "I consider scrambling a mode of transportation sufficiently distinct from walking" and consider it subtype of walking, rather than as distinct as cycling. Also, not as reason for being opposed but as prediction: disturbance caused here in broken display/routing will be far worse than with highway=busway that at least is not used in normal operation by typical users. --Mateusz Konieczny (talk) 10:55, 10 November 2022 (UTC)
@Mateusz Konieczny: Having you on opposing side hurts indeed, so you are right to feel sorry.--Hungerburg (talk) 21:25, 10 November 2022 (UTC)
@Mateusz Konieczny: During today's commute, I was reminded, that a scramble mapped as path in close vicinity was tagged "foot=no" for some time. Perhaps the mapper considered it impossible to walk there ;) --Hungerburg (talk) 20:32, 11 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Sorry but I feel there are sufficient tags now to indicate track conditions and the skill levels required for walks/trails etc. The rendering issue can be taken up with developers. Tastrax (talk) 11:35, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I agree with all the four "no" reason listed above. Especially the part about the "individual hikers ability" and the fact that we have already a lot of different tags to describe paths --Ivanbranco (talk) 12:30, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. As others have mentioned, this tag is too subjective as it depends on individual hikers' abilities. The proposed definition leaves too much room for speculation and debate about individual cases. --501ghost (talk) 14:34, 10 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Currently highway=path is much to broad, covering everything from wheelchair-accessible urban walkways to dangerous mountain traverses. Segmenting out one extreme end of this spectrum into its own top-level highway=* tag will help prevent naive renderers and routers focused on the urban environment from directing their users onto very difficult terrain. (My previous comments with more details) --Adamfranco (talk) 16:17, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I agree that highway=scramble is too nuanced, and it also depends on local hiking culture/tradition if it's really seen different to mountain hiking. But as I agree with the general problem with demanding paths, I think it would fit better in a broader approach which was also discussed like highway=demanding_path, demanding_path=scramble --Eartrumpet (talk) 20:22, 10 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Whilst this isn't perfect, it meets a very important need. The strawman objections raised by folk voting no convinces me that this is absolutely necessary. SK53 (talk) 20:40, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. although the rendering issue doesn't bother me (renderers should follow the map, not the other way around), the vague line between when something becomes a scramble path makes me vote no on a tag like this. I'd prefer path=scramble or scramble=yes or so --Famlam (talk) 20:51, 10 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I'm on the edge, because it has Verifiability issues as noted, but also has utility. But the worry it might break routing and rendering is misplaced -- its main purpose (as I see it) IS to break routing and rendering! (e.g. you don't want your grandma to be sent to such a technical path, which many routers will happily do currently as they don't take into account additional tags). So highway=scramble shouldn't be routable (or renderable!) at all by default, but only in special apps that implement special support for it. If it weren't for that "break routing" feature, one could happily just use sac_scale=*, smoothness=* and friends. So I'd work on making it more general and verifiable. Also Hungerburg, do not take it personal. "No" is often more akin to "needs more work, improve & try again" then "never gonna happen!". Update: after extra consideration on Discourse discussion, I'm thinking something akin to highway=demanding_path + demanding_path=scramble would be much more versatile tag combination, while still equally clearly marking scrambles. Therefore, I'll vote no here, but would vote yes if next attempt was modified in such way. --mnalis (talk) 21:19, 10 November 2022 (UTC)
@Mnalis: Of course I do take this personally ;) I am not in for careering, I am in to make the best map of the world even better. I deliberately did not frame this as saving the local mountain rescue some needless operations, but if it worked out that way and fewer people would get led on inappropriate itineraries, I'd be perfectly fine with it. As for the "needs more work": If there were constructive comments from the opposition, that might help. Still waiting for them though. As to verifiability: Don't you think so too, that this here is much easier than sac_scale: If you can't decide a scramble, how can you tag alpine_hiking? --Hungerburg (talk) 12:07, 11 November 2022 (UTC)
@Hungerburg: "How can you tag alpine hiking." There's a fairly clear sac scale chart here for one. It doesn't seem to hard from that how to tag paths with a sac scale. At least it's clearer then this proposal. Maybe you can use it as inspiration for next time ;) Adamant1 (talk) 08:49, 12 November 2022 (UTC)
Using hands is a core element in SAC scale grading. It is the single most important element that separates T4 from T3. If, as you say, this cannot be verifiably mapped, then sac_scale cannot be verifiable mapped just the same. Note that the pictures above were selected by a member of the SAC trail classification working group. They focus especially on using hands and are very telling. At least for me. --Hungerburg (talk) 09:49, 12 November 2022 (UTC)
Sure, using hands is a key element for T4 from T3, but your leaving out the important fact that it's in conjunction with the existence of other, verifiable elements like ropes or chains. Whereas, for all intents and purpose there is literally nothing on the ground with this proposal that has to exist for someone to use the tag highway=scramble. So obviously, there's differences. Personally, I've never said I'd be against a tag for paths that require the usage of hands, but it just can't be the sole indicator. I'm not super into sac scale past T3 for similar reasons though. Like to me T6, "Mostly without a path and unmarked", is completely non-sensical at least for the purposes of mapping the "paths" in OSM since there's nothing ground verifiable about an unmarked path. In this case you need a balance between indicating that people might need to use their hands, but without making it into a completely separate type of highway just to trick the rendering. Which is why I suggested something like hands=required or whatever. It's a lot easier to map something that can't be 100% ground verified if it's done with key as a characteristic of a path instead of a main tag where your saying a scramble is legitimate type of path. Like with the various access tags for instance. It just works better to use access=car then tag something as highway=car, obviously. Part of that is because access is an ephemeral characteristic of a highway, not a type of one. I'd be 100% against sac_scale if it was a replacement for highway=path or whatever instead of an add-on key. Hopefully that clarifies things :) Adamant1 (talk) 10:39, 12 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. for many reasons already mentioned above--CJtmmr (talk) 21:29, 10 November 2022 (UTC)
  • I approve this proposal I approve this proposal. The analogy with highway=steps is compelling. I consider the alternative, sac_scale=difficult_alpine_hiking on a highway=path, to be a trolltag as it radically changes the expected meaning of highway=path. Introducing a new highway=* value avoids this problem. --JeroenvanderGun (talk) 22:52, 10 November 2022 (UTC)
  • I abstain from voting but have comments I have comments but abstain from voting on this proposal. highway=path should be replaced. However I personally don't like using highway=* as reflected. --- Kovposch (talk) 03:45, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. highway=path is not really appropriate for the kinds of ways that would be tagged highway=scramble. I am somewhat concerned about the transition period where the new tag will not be rendered on the many OSM based hiking maps. Unlike general purpose maps where not showing scrambles would be appropriate, for a hiking focused map it would be appropriate to show them. Significant outreach to hiking map makers will be needed for this transition to be successesful. -- Ezekielf (talk) 04:31, 11 November 2022 (UTC)
  • I abstain from voting but have comments I have comments but abstain from voting on this proposal.The current usage of path is much to broad. If alpine hiking trails should be part of OSM (and I think they should and in fact are), we need a better distinction. While hiking trails up to SAC scale T3 are generally considered as "path" by most people, this is not true for anything beyond and leads to misunderstanding and discussion. Please not that highway=path is displayed by any renderer, while SAC_scale is usually not. However, I agree with the comments by eartrumpet: Scrambling should not be top level, but rather a level below. I would prefer to have a top level such as demanding_path and scramble, mountaineering etc for further specification of this tag --*abq* (talk) 07:20, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. As others have mentioned, highway=path is much too broadly defined, so I am in favour of every attempt to split it into multiple well-defined tags. -- Discostu36 (talk) 07:25, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. I understand that some people are concerned about verifiability, but I feel that most hikers/mappers are likely to agree on the dividing line between path and scramble in 90% of situations. For the remaining "borderline" cases the choices is probably not that important. I also note that some other objections on the grounds of the function of a path/footway would apply equally to highway=steps. --Alan gr (talk) 09:03, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Lonvia (talk) 09:31, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Jrachi (talk) 11:41, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. while the reasons for rejection persist, after a re-evaluation of the weights I now believe the benefits outweigh the potential problems such as the reasons mentioned by Mateusz and others (not verifiable (at the edges)), I also hope it will not lead to people adding actual climbing sections as scramble. —Dieterdreist (talk) 11:57, 11 November 2022 (UTC)
Experience shows, mappers who do that are quite happy with highway=path, they wont switch. I do not want to get into an argument with those people, so the proposal explicitly rules out climbing. --Hungerburg (talk) 12:46, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Amadvance (talk) 14:13, 11 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Makes the map worse because it gratuitously introduces yet another tag, makes editing harder and the interpretation of the map fuzzier. We have a better tag for this already, and that is: sac_scale. The SAC scale is an internationally recognized standard defined by the Swiss Alpine Club, an institution with more authority in the matter than we can ever muster. People using the map are much more likely to understand the SAC scale than the proposed tag. Renderers willing to use sac_scale can already give a finer grained rendering than with the proposed tag. This proposal aims to deliberatly break renderers because you don't like their ways. You should talk to vendors instead and convince then to implement sac_scale. --Highlander (talk) 20:06, 11 November 2022 (UTC)
@Highlander:, please note that while SAC scale may be well understood in most European countries, other areas of the world use completely different scales and are not familar with SAC. -- Ezekielf (talk) 21:37, 11 November 2022 (UTC)
@Ezekielf: While I agree that the average person outside of Europe doesn't know what sac scale is the tag is used pretty widely all over the world. Which at the end of the day is what should matter when it to comes to the tags we use, not how educated "normal" people in particular countries are with the concept the tag is meant to convey. A good example would be how roads are currently tagged, which I'd argue isn't institutive to most "normal" people in the world outside of Europe, but still works reasonably well because of the support the tags have in various styles/software/Etc. Etc. In this particular case, like Highlander says it's more on the software to implant sac_scale in a more fine grained, understandable way then they currently do. Instead of us trying to shoe horn in another tag for something that's already ubiquitous mapped just for the rendering or whatever. At the end of the day if people can't tag for the rendering, we shouldn't be able to create proposals for the rendering either, and that's mostly what this feels like to me. Adamant1 (talk) 01:38, 12 November 2022 (UTC)
Indeed, one could consider highway=scramble an attempt to fix the problems, that OSM carto has with sac_scale. After all, its focus is on a single aspect, while sac_scale combines it with several other aspects. And this single aspect also known in other classification systems, e.g. YDS, that is popular in the US. --Hungerburg (talk) 10:14, 12 November 2022 (UTC)
The "problem" that carto has with sac_scale is that they ignore it. How is your proposal going to change that? There are 14 uses of highway=scramble against 696.118 uses of sac_scale. What are the odds that OSM carto is going to implement your proposal in preference of sac_scale? You will just break carto and lots of other renderers and gain nothing. tl;dr: Don't write proposals for the renderer. / This is not at all like steps: steps is verifiable on the ground and often even from airborne imagery. Scramble is subjective and unverifiable. / It is also less verifiable than SAC scale. (If you don't like SAC insert any other scale used by mountaineering professionals at your location.) You can ask any professional mountain guide or consult any publication and you'll get straight info regarding the local difficulty scale. Yours is a private solution that nobody else will ever use. --Highlander (talk) 12:36, 12 November 2022 (UTC)
  • I approve this proposal I approve this proposal. At least as different from the average highway=path as highway=steps. Deserves its own value. --Tordanik 21:23, 11 November 2022 (UTC)
  • I approve this proposal I approve this proposal. It's not perfect, but it's the least worst option. SomeoneElse (talk) 20:32, 12 November 2022 (UTC)
@SomeoneElse: Why do you consider it better than the option highway=demanding_path? :-) --Schoschi (talk) 03:05, 21 November 2022 (UTC)
It's more descriptive and less open to misinterpretation. "demanding_path" is more likely to be misinterpreted with different people having different views. "scramble" is more clear-cut. Also "demanding_path" may be included alomg with "path" by unthinking map app creators, which is the problem that we are trying to solve. SomeoneElse (talk) 13:50, 24 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Janko (talk) 21:00, 12 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. A solution in search of a problem. As a keen hiker my impression is that sac_scale=5/6 is more suitable and less misleading for non-native speakers than "scramble" to describe mountain paths where "the use of hands is required". --geow (talk) 22:45, 12 November 2022 (UTC)
@Geow: An avid hiker here too: Happy to not have to carry the equipment required for T5/6 if I am out for a mere scramble. As native speaker I'd call it Kraxeln. --Hungerburg (talk) 23:47, 12 November 2022 (UTC)
@Hungerburg: Just back from a nice hike with T5 scramble sections :-) As mentioned by others "scramble" is less precise and less verifiable than the widespread used and well defined sac_scale. Apart from that, (1) the average experienced person usually does not need extra equipment for UIAA I (sac_scale=T5) and UIAA II /sac_scale=T6 and (2) it is seldom advisable to choose your gear based on third party OSM-tags alone ;-) --geow (talk) 18:43, 13 November 2022 (UTC)
@Geow: You are not the first one to lecture me on this. I stopped counting the number of people who obviously do not know how sac_scale is decided. --Hungerburg (talk) 22:38, 13 November 2022 (UTC)
@Geow: On the apart part: You brought up an important point: Too many inexperienced people are doing just that, following router directions based on OSM-tags alone, all too often just plain path. I honestly do not expect this will ever change, we are a niche group occupying a majority tag. --Hungerburg (talk) 17:40, 14 November 2022 (UTC)
@Hungerburg: Please use more arguments and less name calling. If a lot of people are lecturing you maybe you failed to understand an important point. You failed to explain to us why we should use your private standard instead of an internationally acclaimed one. --Highlander (talk) 08:38, 14 November 2022 (UTC)
  • I abstain from voting but have comments I have comments but abstain from voting on this proposal. comments -- I found it bitter that we have to create a new tag to hide data from consumers that failed for years to properly address tags like sac_scale, trail_visibility and so on. Because they don't do their job thoroughly, us mapper will have to deal with two competing tagging scheme and a messy database for years. I understand some want to move forward, will respect the outcome of this proposal, but won't be part of this.Yvecai (talk) 05:28, 13 November 2022 (UTC)
@Yvecai: Thanks to your effort on skitours I got reminded of mapping invisible stuff. It makes me wonder, why the scrambling community is so set up on path. --Hungerburg (talk) 19:53, 13 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. sac-scale should do the trick. Kig8472 (talk) 11:10, 13 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I think there are already several tags available to describe paths characteristics. Moreover a scramble tagging criteria seems more related to human abilities than to something verifiable. --IlBano (talk) 15:03, 13 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Although I think that this would be an improvement, I don't think that it is ok to introduce a breaking change to hiking maps without trying to involve at least some of the maintainers of hiking maps in the discussion. --JIDB (talk) 21:58, 13 November 2022 (UTC)
  • I approve this proposal I approve this proposal. I don't think this will break a lot since it is barely used and it will be barely used. Nevertheless, it is a rather specific definition of a type of highway, which I would find useful in field. Definitely, while hiking, most of the time a highway to avoid, or one you should be easily aware you want to follow. --CENTSOARER (talk) 23:14, 13 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Strongly support separating these from walkable paths. --InsertUser (talk) 00:29, 14 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. There are already other tags to address this use cases. --Musuruan (talk) 08:49, 14 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Strongly support this. This sort of intuitive duck tagging is a much better approach than layering on yet more attributes to existing highway tags. --Richard (talk) 10:06, 14 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. sac_scale, uiaa_scale is enough. Hiding high sac_scales on maps is a renderer topic and not a tagging topic --Andreas Binder (talk) 10:27, 14 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. If you introduce a new tag for more difficult paths don't narrow its scope to UIAA II and below. I would support a new tag like highway=mountaineering that would only be valid if its accompanied by another tag describing the actual difficulties (uiaa_grade, yds_grade, hochtouren_grade or whatever you want to call it). With a combination like that you could accurately tag routes like the already mapped normal ascents to many of the peaks in the western alps (Piz Bernina, Monte Rosa, Mont Blanc, etc...) --Gebux (talk) 11:49, 14 November 2022 (UTC)
@Gebux: The scope here is both narrower and wider. Narrower: it is not meant to replace path, but instead to highlight scrambling sections therein. Wider: it is not limited to high mountain areas. --Hungerburg (talk) 18:11, 14 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I still don't see why we need to replace the combination of highway=path and SAC scale 4, 5, or 6 with a different highway key. The real problem is that at present many mountain hiking paths on OSM do not carry the appropriate SAC scale on difficult bits. --voschix (talk) 16:23, 14 November 2022 (UTC)
This does not prevent the use of sac scale. It tells routers and renderers that its not appropriate to carry your shopping up this bit of a route without them having to know ten different types of specialist difficulty ratings --InsertUser (talk) 09:03, 15 November 2022 (UTC)
@Voschix: You can look at it like that: For the OSM sac_scale key, trail_visibility got singled out into its own key. This proposal singles out walkability. It does so even a bit bolder, by occupying a value in the highway key instead of creating yet another attribute to path. Focusing on only one aspect should help mappers produce consistent tagging much more easily. In order to address, what you consider the real problem, to mark the challenging bits, much like we map stairs now, it should be a boon, shouldn't it? (PS: Nowhere the proposal says that whole routes should get based on this new tag, to the contrary!) --Hungerburg (talk) 10:30, 15 November 2022 (UTC)
"Much like we map stairs now." If your going to claim this will work because of the existence of some other highway tag, the tag should at least be somewhat similar to this one. The problem with using stairs as an analogy to scrambles is that there is no ambiguity about what stairs are and they can be ground verified. Neither of which can be said for this tag. I can almost guarantee that if you put two mappers in a room together and asked them if a particular path was a scramble that they either wouldn't be able to or would give different answers. At least outside of just re-tagging paths that are already mapped with the alternatives. Really the only way to know if something is a scramble or not is by seeing if the path is already tagged with sac scale rating and guessing. Which means you've essentially just duplicated sac_scale but with yet another "trail rating" tag. Adamant1 (talk) 10:59, 15 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. --Whb (talk) 01:01, 15 November 2022 (UTC)
  • I approve this proposal I approve this proposal.If this proposal gets rejected, I support a compromise. As people above me have already said, it IS important and has it's use cases :) --KoiAndBlueBird (talk) 19:08, 15 November 2022 (UTC)
The proposal will fail to reach the threshold needed for acceptance. This will not prevent you from using the tag "highway=scramble" where you deem fit. When doing so, even on paths tagged sac_scale=difficult_alpine_hiking, you will in fact add new information to the data. After all, the sac_scale tag does not tell consumers, where the scrambles are, nor it will tell them, if there are any at all. And tagging a scramble is certainly less subjective and a lot more reproducible than tagging sac_scale :) And BTW, tagging a scramble does not prevent you from tagging sac_scale, except T1 or T2, of course, these two values are not compatible. --Hungerburg (talk) 21:29, 15 November 2022 (UTC)
My recommendation would be to find a compromise tag that integrates the feedback you have received from the people who opposed the proposal instead of suggesting that people just use the tag anyway even if this ultimately isn't accepted. Otherwise, your essentially encouraging edit warring. It's totally possible to come up with a tag that would be a good middle ground between this and sac scale that will be accepted by the community if you put the effort into it. Adamant1 (talk) 09:19, 16 November 2022 (UTC)
We may use scramble=yes with appropriate sac_scale=*, surface, width etc on highway=path --geow (talk) 22:13, 16 November 2022 (UTC)
I think that completely misses the point of making unwalkable trails distinguishable without looking at many different tags in combination. --InsertUser (talk) 23:17, 16 November 2022 (UTC)
"We may use scramble=yes with appropriate sac_scale=*" Or you could just admit that "scramble" isn't the best way to describe/tag hard to traverse paths and find something else to use. It sucks the "vote" didn't go your way, but it's not like this isn't a solvable problem. It's just not solvable with this solution. It happens sometimes. Turning "scramble" into a key without the highway part still has the same problems highway=scramble does though. Get over it find something better. There's plenty of suggestions here to build something workable off of.
Personally, I like the idea of hands=* since it would allow for a gradient of options, like hands=recommended or whatever, but I'm sure there's other things that would be just as good. At least that would accommodate the issue of verifiability though. Whereas, something like scramble=recommended is just nonsensical. At the end of the day it doesn't really matter what specific word people use for the tag anyway as long as it can be verified and "scrambles" clearly can't be. Get over it and find something that can. Adamant1 (talk) 05:49, 17 November 2022 (UTC)
The suggestions offered here by opposing voters dismiss the problem this proposal is intended to solve: that scrambles are not paths and it is untruthful to tag them as `highway=path`. Opposing voters seem to be saying this is not actually a problem. Given this, I am not seeing the potential compromise you speak of. -- Ezekielf (talk) 15:51, 19 November 2022 (UTC)
@Ezekielf: I can't speak for anyone else, but the reason I'm "dismissing the problem" is because if we go down that route we could just as easily say scrambles aren't highways either. Which wouldn't help this out anymore then addressing the path "problem" would. Look at this way, according to highway=* "The key highway=* is the main key used for identifying any kind of road, street or path." What option out of those is a scramble if it's not a path? It sure isn't a road or a street, and you say it isn't a path. So why are they being tagged with a highway tag in the first place then? Adamant1 (talk) 12:28, 20 November 2022 (UTC)
@Ezekielf: I was refering to a different solution like highway=demanding_path, which is one of many suggestions that were made while discussing highway=scramble. And of course scramble is a kind of path; since 14 years the wiki says that path is also used for hiking trails; hiking trails include trails for mountain hiking, which according to very well developed and respected definitions like SAC scale can require the use of hands. So scrambles can be part of hiking trails, for which highway=path is used according to the Wiki. Of course the wide definition of highway=path is problematic, but saying it's not part of the current definition misses the point. Not only scramble is problematic; only looking at scramble and not looking at everything else which can be problematic for non-mountain hikers is a step too short IMO. In German we have the word "Steig" for everything which is more demanding than a level footway, this would be the logical differentiation in the German speaking alpine context; that starts with SAC scale T2. And of course a scramble is a part of a "Steig". --Eartrumpet (talk) 15:15, 20 November 2022 (UTC)
@Ezekielf: Several times, abstaining & opposing voters mentioned the alternative highway=demanding_path (or other wording) which does also reach main targets of proposal highways=scramble while avoiding some anticipated problems. Hence, I assume it will be acceptable by many supporters of highway=scramble as well as abstaining & opposing users, in other words, it could be a viable "compromise". --Schoschi (talk) 03:05, 21 November 2022 (UTC)
I have responded on this community forum topic -- Ezekielf (talk)
I nearly forgot: You will also map in a simple and concise way, the perhaps most important information: There is no path there to walk on, to ride a bicycle or a horse there. --Hungerburg (talk) 22:38, 16 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. --Map per wiki (talk) 21:20, 16 November 2022 (UTC)
  • I abstain from voting but have comments I have comments but abstain from voting on this proposal. I haven't followed the entire discussion on this topic but whenever I read further opinions, in the end I never manage to get off the fence between the pros and cons. I do agree with @Yvecai, though, it's a shame that we need to discuss how to "hide" stuff which should really be better observed and handled by the various downstream consumers and renderers in the first place. --Arminus (talk) 11:37, 17 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --EdLoach (talk) 09:41, 18 November 2022 (UTC)
  • I abstain from voting but have comments I have comments but abstain from voting on this proposal. I'm very on the edge, as I agree with the problems of verifyability and I'm also definitely against introducing a new value just to hide routes from renderers and routers. I can possibly agree with scrambles being functionally different from regular walkable paths, but I don't think most people will like to use the new tags if they're not displayed or routed. I personally would want to see them on the main map and I would possibly want to route over them. So either routers and renders will start supporting highway=scramble, which would make this new value largely useless, or they don't, which is also not great. --Jonathan Haas (talk) 12:00, 18 November 2022 (UTC)
  • I approve this proposal I approve this proposal. I also think that the combination of highway=path and sac_scale=difficult_alpine_hiking is akin to a trolltag. Introducing highway=scramble will fix this issue. --Timmy_Tesseract (talk) 13:58, 18 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Luzandro (talk) 08:27, 19 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I prefer a "scramble=yes" attribute --Jcr83 (talk) 08:43, 19 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --快乐的老鼠宝宝 (talk) 09:42, 19 November 2022 (UTC)
  • I approve this proposal I approve this proposal. As others have written: it's not perfect, but better than the status quo and i don't have a better solution. --Dafadllyn (talk) 17:47, 19 November 2022 (UTC)
@Dafadllyn: Have you been aware a solution highway=demanding_path is discussed? :-) --Schoschi (talk) 03:05, 21 November 2022 (UTC)
@Schoschi: scramble seems clearer to me than demanding_path. --Dafadllyn (talk) 17:26, 22 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Enough with sac_scale and/or scramble=yes --Javiersanp (talk) 23:00, 19 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I agree to tag and share trails that the inexperienced average climber cannot take. However, there are also short vertical cliffs with some safety devices in the middle of the general hiking trails. No one can divide the path that an unskilled climber can take just by grade or difficulty. I think it should be divided according to skill level and equipment need. --깨몽/dreamy (talk) 12:10, 20 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --EneaSuper (talk) 12:34, 20 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. use sac_scale and/or scramble=yes with a standard highway tag--JB (talk) 12:54, 20 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. highway=scramble, like highway=via_ferrata, will not be used often and eventually will not be rendered anywhere. Instead of introducing a new highway type, I would rather use a tag that, like smoothness, shows the actual subjective assessment of the path. This could happen for example with an additional tag like path_difficulty=scramble. But you always have to see that a "city kid" who is in the mountains for the first time is much more likely to be "crawling on his hands" than an alpinist who has done nothing but hiking in the mountains since childhood. The latter will probably complete a trail run on the same trail without using hands at all. In addition, a simple hiking trail can become a "scramble" in winter. Anyway, I think there are already a lot of tags that describe walkability in terms of subjective properties well like smoothness, sac_scale, incline, etc. --Mcliquid (talk) 12:58, 20 November 2022 (UTC)
My hiking map renders via_ferratas, I guess most of them do. --Hungerburg (talk) 23:37, 20 November 2022 (UTC)
I'm not aware of a single that doesn't. To drop a few names: Locus Map, bergfex, komoot,, Outdooractive (or branded versions), OSMAnd or OpenTopoMap all show highway=via_ferrata --Luzandro (talk) 13:12, 21 November 2022 (UTC)
@Mcliquid and Luzandro: A map that does not render via_ferratas is not supposed to render scrambles. I read this called a feature, not a bug. --Hungerburg (talk) 11:07, 22 November 2022 (UTC)
@Hungerburg: I must have been unclear here. I was referring to highway=scramble. highway=via_ferrata is represented by pretty much all maps except Carto itself. Currently there are 16 uses, it will take forever until these paths are displayed and used especially by route engines (just one example is Garmin). But completely independent of this, I stand by my opinion that the use of highway=scramble is incredibly subjective. In my area there are some "Instagram hotspots" in the mountains (e.g. a beautiful mountain lake with an old wooden hut embedded in the mountain). The number of photos now exceeds six figures. There you can find local trail runners next to foreign people who are there for vacation, have never been to the mountains before and climb the mountain in their heels or simple sneakers. As soon as it gets a bit steeper or there are still a few centimeters of snow in early summer, the latter crawl up to the lake on all fours. Let's imagine a narrow ridge, exposed, on the left and right it goes down a few hundred meters. Will the experienced trail runner go here on all fours? How will the tourists get on here? How big is the technical discrepancy between the two groups of people? @Hungerburg You are surely often at the Nordkette (Innsbruck, Austria). The very first time I was there in the mountains some decades ago, I had my via ferrata set with me and went up between the chimney tops (Kaminspitzen) on all fours sweating from fear. From behind, an older man overtook me, in his simple trekking shoes, without a harness and jumping on top of the ridge from rock to rock. How do you get both technical levels combined in one tag? I mean that would be like if I change a pass road in the Alps to highway=slippery in winter. The "German stereotype on summer tires" will be happy, the local with his 4x4 AWD does not understand the issue.
I hope that was a little clearer.
  • I oppose this proposal I oppose this proposal. SAC Scale --Reino Baptista (talk) 16:30, 20 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --scai (talk) 17:53, 20 November 2022 (UTC)
  • I approve this proposal I approve this proposal. Timmy_Tesseract sums it up. To the proposal author(s), I'd suggest to explain better why SAC Scale is not a solution. Many "no" voters based their "no" vote on referring to this, so they very likely did not understand the issue with this. --Westnordost (talk) 17:55, 20 November 2022 (UTC)
@Westnordost: The many SAC fans did not show up during two months of RfC. I do not think any explanations would have led to a different outcome. --23:37, 20 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Robybully (talk) 19:56, 20 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --OSMRogerWilco (talk) 20:56, 20 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. While I consider highway=scramble to be an improvement over current tagging, mainly by reducing the huge variance of meaning that highway=path has to cover (others word it as "path is overloaded") with many positive consequences, I consider it to be considerably inferior to the voiced alternative (see g.g. tagging mailing list post and post) highway=demanding_path with secondary tags telling what kind/characteristic of difficult path it is/has – IMHO one the first of these secondary tags would clearly be demanding_path=scramble (or other wording) because this is well prepared; we can luckily re-use all the efforts Hungerburg and others invested into highway=scramble. Main reasons why I see highway=demanding_path superior:
  1. It also reaches most (all?) mentioned main targets of highways=scramble, e.g. it segments off one end of highway=path's spectrum and make this 1st level key-value-pair more meaningful by allowing to tell apart by looking at only one single key instead several dozen keys whether "everyone" can walk a path or not, it bears the same potential to make rendering easier and routing more accurate (see discussion topic, especially lower part), it even fulfills minor goals like "security by design" even for quick & dirty implementations (see community post).
  2. It's much better concerning verifiability in the sense map what's on the ground but also less subjective, less depending on whether etc.
  3. It allows tagging to better reflect the reality because in contrast to highway=scramble, a certain way can not only be tagged as scramble XOR via_ferrata XOR climbing route while all three usage styles are possible, but all three can be tagged at the same object so there is no reason/base for many discussions, edit wars etc. at all or at least their implications are considerably reduced because it's not about 1st level key but some lower level tag/attribute.
  4. It is broader. Hence,
    1. it solves the issues that led to proposal of highway=scrambling not only for scrambles but for a much bigger amount of ways, e.g. paths over ice, crawls like in a tunnel or due to vegetation or on a trail leading through a narrow C-shaped carve-out in a rock face (example pic)
    2. it will have much more uses, which increases chances that mappers, software developers etc will consider it.
  5. I assume it makes our lifes easier in the future because it has a more holistic view and thus avoids a flood of follow-up proposals of new values of a major key, leading to repeated compatibility problems in less maintained tools – while changes to secondary keys/values has less impact. Why do I expect a flood? As I don't see scramble duplicating sac_scale because it's more specific – hand usage yes/no is one aspect of sac_scale – I do anticipate the wish to tag other aspects of sac_scale and/or to have ways with other qualities as dedicated highway values – think e.g. "glacier_traverse" or others.
  6. Some adoption issues named in Peter's and Yvecai's post can be avoided and transition effort could be reduced because an automated edit for a considerable share of ways (all SAC T4-T6 see this tagging mailing list post) seems possible – we must of course discuss whether we want to do it or not – which is not possible due for highway=scramble because of lacking clear information base.
  7. Just like highway=scramble, it can be realized quite quickly compared to a general analysis & clean-up of all path-related tags described in Path_controversy. And IMHO we shall improve the situation quickly!
  8. It's not describing a human activity but a type of ways like other values of "highway" do, so it results in a more consistent tagging scheme.
--Schoschi (talk) 03:05, 21 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Naomap (talk) 08:01, 21 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I prefer path + attribute like sac_scale --chris66 (talk) 15:22, 21 November 2022 (UTC)
  • I approve this proposal I approve this proposal. This hopefully reduces the number of incidents where pedestrians are sent to unsuitable routes. --drolbr (talk)
  • I approve this proposal I approve this proposal. --Nielkrokodil (talk) 17:05, 21 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --clay_c (talk) 02:39, 22 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. This is yet another attempt to hide supposedly difficult and/or dangerous paths from the map. We've already seen this with highway=via_ferrata. It makes no sense at all to invent new highway tags for various shades of mountain paths when they obviously overlap (a path secion can have rungs and require free climbing as well, example: Wildenauersteig) and there are already plenty of subtags to define their properties, e.g. sac_scale=*, via_ferrata_scale=*, safety_rope=*, rungs=*, ladder=*, incline=*, obstacle=*, uiaa_scale=* (or climbing:grade:uiaa=* etc., see climbing), trail_visibility=*, surface=*, smoothness=* and hazard=*. You compare highway=scramble to highway=steps which, as we all know, is a burden from the past and we would nowadays prefer highway=footway/path + steps=* instead but we need to stick to highway=steps for compatibility reasons because millions of highway=steps exist and a change would break all the applications and editors and lead to a mess in the data with millons of objects tagged this way and millions tagged the other way. You want to create the same mess with mountain paths but going from the modern approach with subtags to the obsolete approach with different main tags. To make it even worse, the proposed tag is ill-defined: it's meant for "paths ... where use of hands is required", and "a scramble ends where climbing starts", but climbing by definition starts where the hands are used. Also, there is no clear line between paths and climbing routes. Some people use their hands where others won't. What experienced climbers call "walking terrain" is considered impassable by ordinary people. Different climbers assign different UIAA grades (+/- 1) to routes or sections. A route can technically be easier but more dangerous because of exposure or brittle rock or unstable ground, or the other way round. Every rainfall or rockfall or fallen tree or even a newly grown bush can make a difference whether you use your hands or not. I doubt that any of the "yes" voters have any experience with mountaineering at all and it's a serious flaw in the proposal process that people vote on subjects they know nothing about. Also, a voting on changing millions of OSM objects shouldn't be possible a few weeks after the RFC. If the proposed tag were useful, people would start to use it during the RFC period and we'd see what problems arise. These issues would then be discussed, the proposal would be adjusted, the definitions and subtags refined. This process may take years before the proposal gets ready for voting. --Fkv (talk) 05:13, 22 November 2022 (UTC)
Interesting idea, I may prolong voting for another year :) Mapping a few scrambles in my local area indeed brought up some attributes, that are useful in combination already. I really like the term, it is so much closer to the subject matter. That way I also learned, that some routers supported this out of the box too, at least when the scramble is part of a hiking route relation, etc. --Hungerburg (talk) 11:07, 22 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. I am a retired mountain hiker, blew my knee in 'The Trough' of Longs Peak, so I understand the concept of a scramble. However, I have three issues with this proposal
  1. What most mountain hikers subjectively consider a 'scramble' encompasses what sac_scale=* already covers but separates into two more specific categories namely 'sac_scale=demainding_mountain_hiking' and 'sac_scale=alpine_hiking'. The fact that the sac_scale is 1) a scale and 2) is already more precise is not lost on me especially when I realized that in your proposal you combine the photos seen from sac_scale#Values=* for these two classifications (hiking with mostly footwork only vs hiking with feet and hands) to describe this one binary yes/no condition. Actually looking at the Oxford Dictionary definition for 'scramble' "using your hands to help you" it could even be argued that only 'sac_scale=alpine_hiking' is classified as a true "scramble" meaning there is a direct 1-to-1 analog for this proposed tagging that already exists.
  2. From a routing and rendering standpoint I think adding a new highway type will make both of these worse. The hiking trail is fundamentally a highway=path first and then additional tags are added to describe the path better and more clearly. I disagree with highway=scramble and highway=steps being comparable. Steps have clearly defined starts and stop a scramble does not as demonstrated that two levels of sac_scale=* encompass it. Also steps have clear impacts on bikes and disability while scramble doesn't even necessarily clearly change the mode from foot only in nature and tends to scale more based on athletic ability.
  3. I don't feel 'scramble' as a tag name is specific enough to encompass the one and only meaning within OSM tagging. I'd like to point out that crossing=scramble exist and even without officially being approved has 100+ tagging uses with OSM but NOT to describe a climbing use case but to describe a wikipedia:en:Pedestrian scramble. See usage stats below from TagInfo. I don't agree with this tagging convention but understand its use case more if we are going to have two tag usages both trying to claim 'scramble' then I think there is a good argument to be made that in order to avoid future tagging conflict that 'hiking_scramble' and 'predestrian_scramble' should be used instead of just 'scramble'. (I'm sorry that I didn't see this proposal when it hit RFC to make this call out but I believe it to be a very reasonable request.)
In summary, I think the ideal outcome is not to abandon this line of tagging entirely but I strongly believe sac_scale=* to already cover this ask with more specificity and without trying to replace highway=path when no clear delineation exists. I would encourage that the existing more specific tagging be embraced but I would find it advisable for you to add to sac_scale=* for both its 'sac_scale=demainding_mountain_hiking' and 'sac_scale=alpine_hiking' levels that these two levels are commonly referred to within the hiking vernacular as a "scramble". --JPinAR (talk) 19:02, 22 November 2022 (UTC)
For the avoidance of doubt, that is for and is somethingelse entirely. SomeoneElse (talk) 13:59, 24 November 2022 (UTC)
  • I oppose this proposal I oppose this proposal. --NKA (talk) 07:07, 24 Novembe r 2022 (UTC)
  • I oppose this proposal I oppose this proposal. Sorry bro, but I‘m against another highway= tag for something this insignificant, that‘s really just making everything worse. If you truly see it as this neccessary still, please just Propose something entirely new.Emilius123 (talk) 09:14, 24 November 2022 (UTC)
  • I approve this proposal I approve this proposal. --Vincentius (talk) 21:18, 24 November 2022 (UTC)
  • I approve this proposal I approve this proposal. -- Agreed. A scramble is not a footpath. It is not desirable for routing to regard a scramble as identical to a footpath. There have been other proposals. Something aligned with internationally recognised hiking standards (if such exists) would be preferred, but this is what is on the table today. Thank you for building the highway tagging system. CRaIgalLAn (talk) 01:17, 27 November 2022 (UTC)

Post suspension

A handful of insightful comments here and in the community forum, from supporters as well as opponents and abstainers, pointed out, that the proposal was not yet ripe for vote, resembled an RfC too much, may get superseded by ideas floating. From the proponents point of view, this especially concerns how the relation with sac_scale (a.k.a. state of the art schemes) might have just been poorly communicated, but rationale and delineation certainly can be improved. So vote is suspended for now, and the proposal reset to draft state. Don't hold your breath for new content to appear, work on this is done in spare time. Please use the talk page, if there is anything meaningful to contribute. -- Hungerburg (talk) 23:01, 27 November 2022 (UTC)

Why not set the status to rejected since that's clearly the outcome? Then a version 2 of the proposal can be created at some point if need be, but just re-drafting the proposal makes it seem like there wasn't a clear outcome and the proposal needs minor adjustments when there was an outcome and this needs more then minor fiddling to be workable. For instance, if the "new" proposal is for scramble=* (which seems likely) then your probably going to create a whole new proposal for it whatever the case with this one is. Adamant1 (talk) 15:58, 28 November 2022 (UTC)
+1 --chris66 (talk)

  • I oppose this proposal I oppose this proposal.This would be like having highway=bridge instead of and adittional bridge=*. I do not see an improvement over sac_scale=* or a new scramble=*. The multitude of differing voting opinions shows this topic is still controversial. Please don't push it through. A proposal should meet not too much heavy opossition to be a real lasting benefit. --Trapicki (talk) 16:03, 4 December 2022 (fixed 19 December 2022, sorry)
@Trapicki: This makes me curious indeed: What kind of highways are you wanting to tag with key scramble: Residential, Trunk, Track? The emotional appeal makes me even more curious: Would you share some of what actually makes you oppose this?
Not to tell either of you what to do, but it would help if you both included signatures with your comments so people reading along, including myself, can tell exactly who is saying what. Otherwise, it can get confusing. Thanks! Adamant1 (talk) 01:26, 11 December 2022 (UTC)
@Adamant1:I'd like to invite you on a UIAA II scramble, you might be in for a whole new experience, as you said, you ventured into T3 at most, but unfortunately, I cannot accost you for travel expenses. I can invite you though to enlighten router developers on the clarity of sac_scale, that you praise so much, here - perhaps informing them about what is documented at first. I will remain silent on the lamest ever opposing vote by System-users-3.svgTrapicki (on osm, edits, contrib, heatmap, chngset com.), cast after suspension when the vote template was no longer available. --Hungerburg (talk) 23:57, 16 December 2022 (UTC)
I appreciate the offer, but unfortunately will have to decline it. As I don't live in Europe and the cost of travel is way to expensive right now :( I've always wanted to visit my European counterparts though. It would be cool to meet them in person. Maybe someday. Adamant1 (talk) 04:22, 17 December 2022 (UTC)
@Hungerburg:@Adamant1: I truely apologize for adding a vote after Vote Suspension. I just overlooked the Suspension and overshot heavily by several days. I really meant that the vote should be suspended, well... . Moreover I did not want to offend anyone in any way. With the tone of word of this thread it seems prudent to not continue on this path. While we all put a lot of work and emotion into OSM - and of course everyone tries to improve it with the best ideas (which is understandably everyones own ideas) - , it is not *so* important to get angry at each other. Hungerburg, next time we meet we discuss this over a drink/snack of your choice which is on me. --Trapicki (talk) 21:25, 19 December 2022 (UTC)
No worries on my end. For me the issue is more the fact that this was clearly rejected and probably can't be re-purposed under the same proposal anyway, then it has to do with someone accidently voting after it was drafted. Neither one is that big of an issue in the grand scheme of things. Although, if there's ever a chance of something like this ever passing a vote in the future it has to be sufficiently different from the first proposal. Which making superficial changes to the wording of this proposal while using the same exact tag wouldn't qualify as IMO. But then it's not my show either and I trust Hungerburg knows that without me harping on him about it :) Adamant1 (talk) 03:01, 20 December 2022 (UTC)