Following this discussion: Highway=busway on non-BRTs
I wrote this proposal:
https://wiki.openstreetmap.org/wiki/Proposal:Add_brt%3Dyes/no_to_route%3Dbus_to_mark_BRT_lines
Can I ask for it to be sent to tagging list, please?
I hope this could be a simple nad non-controversial thing that would help with the busway mess.
Also, is there a way the proposal page can be renamed from
Add brt=yes/no route=bus to mark BRT lines
to
Add brt=yes/no to route=bus to mark BRT lines
Simply use the Move function (located under “More”).
With strong personal opinions, I must say “BRT” is not “non-controversial”. You will end up with people misleadingly tagging it for systems that call themselves a “BRT”, but isn’t one in any definition. Or there will be many arguments over them. Then it’s not really that useful. Shouldn't we make rules that suit each country's road system? - #45 by oruk09
Even if you use brt=yes to mean anything advertised as “BRT” somehow, using it on both qualified and unqualified services dilutes the meaning. It does a disservice to both.
My perspective is to describe verifiable criteria Highway=busway on non-BRTs - #37 by Kovoschiz
Eg stopping_pattern= is something specific. passenger= is already used on railway= and route=train etc, instead of the generic scope= proposed before. Proposal:Differentiation for routes of public transport - OpenStreetMap Wiki
If the name= contains BRT, it can already be searchable in theorey. To describe the functionality, I love the historically accurate Japanese, and existing Korean terminology “trunk” bus. So we can describe whether the route=bus is performing a backword, main, competing alternative to rail, auxiliary, or rail feeder role, etc.
Recently, there’s a question on articulated buses as well Talk:Key:bus class:length - OpenStreetMap Wiki
I’ll be honest: I’m not happy with using brt=yes. It feels rather bare bone and doesn’t conform to typical OSM tagging.
My preferred solution would be to use route=brt but that has the downsides of
a) being invisible to not-up-to-date renderers and
b) you can’t specify whether it’s a trolleybus route or not (though that alone gets a bit difficult muddy with the increased use of dual-mode trolleybuses which can run off-wire for longer).
The alternative is to follow the practices of train routes and use service=brt which might be an even better fit akin to the difference between service=regional and service=commuter.
Well, I do not care about the particular name of the tag, I just strongly feel there should be a simple way to indicate a route is BRT. BRT is something that exists out there, is pretty recognizable and potentially very important for people using maps (for those of us not having a car).
Are there actually any trolleybus BRTs? I always thought trolleybuses are a very niche Eastern Europe thing :-).
@Kovoschiz The proposal explicitly says one should map according to reality, not what something is called (or certified as).The motivation is not for it to be searchable by endusers but for renderers to be able to display such routes more prominently. THe proposal you linked definitely sounds overengineered to me.
Thank you @supsup for starting this. I think the idea to have an easy way to highlight BRT/Express bus routes could help a lot in some areas of the world.
I have two comments:
In the inteoduction you refer to a wikipedia page for the definition. Please put the definition directly at the proposal page, so that a) we dont have to click the link and b) the proposal stays the same when wikipedia gets changed.
as there are many objectives to qualify as BRT, there ist not just yes/no but something in between. The proposal must be clear on how to handle these routes which fit some, but not (almost?) all brt qualities.
I was thinking of the wheelchair=limited tagging as middle between yes and no. Maybe brt=full/partial/no or yes/partial/no could solve this.
Who is the authority on whether a line qualifies as BRT service or not? It is a global standard, or do we leave some leeway for local pecularities? Is the purpose of this tag for users to merely identify self-styled BRT services, or for data consumers to automatically treat these services differently?
ITDP publishes a Bus Rapid Transit Standard and scores existing systems on a 100-point scale. This is probably as rigorous as it gets. However, sort of like the various walkability scores out there, I wonder whether this information would belong in OSM. It’s not as though the lack of a tag has stopped anyone from compiling these scores externally.
@Nate_Wessel once created this transit map of my hometown, which focuses on frequency over branding – hard facts, classically OSM. Even if the city were to introduce a “BRT” service, this map would treat it like any other, by design. The various characteristics associated with BRT are mostly intended to improve reliability and frequency, so if those characteristics prove to be a success, that success would be apparent on this kind of map.
On the other hand, maybe there is a desire to record different relative classes of service as designated by each operator. service=* seems like a good place for that, even as we lament the continued overloading of that key. Just don’t tag it on ways, because then we’d have to call it service=busway all over again. 
Personally, those service= is too German-centric. =regional is unsuitable, and =commuter is a poorly translated term. They are different aspects, one for the extent overlapping with passenger= , and one for the service pattern. S-Bahn means “city”, and RER is “regional”. In Japan, “regional” rail is commonly “commuter”. North American “commuter rail” is peak hour only.
They could be changed to eg =rapid_transit and =metro (“metro” service) somehow. Then this can be used on both =train and =bus .
We can use more differentiation than service= . Mixing different aspect doesn’t work.
=night : would appear that it can be used for any late-night service, which is better distinguished as =overnight and =sleeper somewhere.=high_speed : there is already highspeed=yes , and there are high-speed sleeper trains.=tourism : There are daytime tourist trains, and overnight sleeper cruise trains. There were high-speed tourist trains as well.Transit frequency mapping is location-dependent. 15min in USA may be considered “frequent”, but only so-so in Eurasia. The style needs to be rendered per-region.
On the other end of the spectrum, brt= or =brt is oversimplified, and too subjective. How to make those local judgements? Discarding those “certified as” reference standard makes it worse.
You will get complains and abuse from people trying to add that to what they see as being advertised as “BRT”. That won’t work nicely.
service= at least attempts to focus on the “service”, perhaps rather than being decided by the infrastructure. Although, this still has many possible meanings.
Fundamentally, a bus service could be high-quality and providing trunk service without being “BRT”. That’s what I suggest adding directly, the functional classification, not whether it’s some “BRT” concept. That would show significance and prominence accurately and reliably.
Done, the definition was actually there all along after the parenthesis.
brt=yes/limited/no sounds like a good improvement to me.
@Kovoschiz I think the main criteria is if it is almost as important for the transport in a city as a subway or commuter rail would be. The BRT standard “leaderboard” is here: BRT Scores - Institute for Transportation and Development Policy
I think there probably should be an implication that anything that is “silver” and “gold” should be classified as BRT (and BRT certified and bronze can be “limited”). However, I am not sure they certify everything, I can imagine one needs to pay them and there were only several added rankings between 2018 and 2023, plus I am not sure if a ranking from ten years ago still holds.
I am for human-intelligible tags, so service is kind of meaningless to me (it can mean too many things). importance or quality would make more sense to me.
I guess there are two competing visions here - devise a very(?) detailes classification that would closely map onto a complicated reality vs. have a somewhat crude classification that would allow a simple distinction between very important routes, somewhat important routes and not so important routes.
Another way to put it is that that there is a phenomenon that is called BRT, it has physical manifestations, is kind of distincti and is unmappable in OSM currently. The question if if it should be mapped by its own thing or through some very generalized scheme.
For one, I think this should only apply to mass transit routes, no intracity routes.
Given how varied mass transit systems are across the world, I am not sure we can have a one-fits-all approach. Where I live, we have an excellent fixed-rail infrastructure so no need for BRT. Other cities are not as lucky, some are in between (Istanbul comes to mind).
It’s still conflating 2 aspects. Functional importance and physical standards are different, similar to the logic in highway= roads. A route=bus could be as important as heavy rail, while not being up to the level of Basic BRT. The aforementioned misleadingly named Tokyo BRT might ironically be an example of this, and Keisei also runs the Shuttle Seven on Kan-Nana-doori (Circumferential Route 7 Avenue) . There are even “certified” “Not BRT”, while eg NYC MTA SBS may be seen as similarly important. Bus rapid transit creep - Wikipedia
For one thing, the BRT Standards only scores center-running, or one-side-running two-way along development edges (obstacles such as coasts), with turns banned. A right-side bus lane system along the sidewalk kerb may be important, but isn’t considered a “BRT”.
Besides, we need to consider routes vs infrastructure. Open BRT systems can travel on a BRT corridor in the core section only. How long does the traveled length, or the BRT corridor needs to be? These are all questions that needs to be considered. So BRT seems a distraction away from the inherent need of distinguishing importance. We can take it into account, but that doesn’t need to be the centerpiece and decisive factor.
Ok, so something like
quality=backbone with a note saying, if this is good quality BRT, it is this, but can also be non BRT if it is important enough.quality=medium important links that are not good enough for backbone, some preference and special infrastructure, low graded BRT can often be thisor maybe
For fixed rail routes the route= tag is used to distinguish between rail, subway, light rail, and tram. The dividing line between all those isn’t always clear but from what I’ve seen the OSM classification seems broadly sensible. E.g. the two nearest cities to me have systems branded “Metro” which might suggest a subway tag, but are more like trams running partly underground, and are tagged as light rail in OSM.
Is the distinction between BRT and “normal” bus routes analogous to those fixed rail distinctions, or are we trying to achieve something different?
I don’t see this as competing. They can coexist. See tracktype and surface/smoothness. I am glad we have tracktype, but I think if it was to be approved today it would get rejected.
@supsup Could you draft something like a checklist of the most important brt criteria and how many of those should be fulfilled for “yes” and how many for “limited”? Maybe there are some “must haves” and some “nice to haves”. I think the list should be short with only the most important points.
I think the mapper can judge if overall the tagging fits or not. It’s a bit like railway=light_rail on a line being partly tram. This is somewhat fuzzy, but works. As long as within a city/transport network it is applied consistently I would be happy.
Partially. To me it is about “having dedicated infrastructure” some “dedicated infrastructure” and no special infrastructure. The first two somehow map on lightrail and tram.
@Nielkrokodil
Checklist would be (inspired by the BRT standard but not quite)
Personally I would require all for “yes” and at least three for partial (the first one being obligatory for partial too), with dedicated infrastructure and preference on intersection needed for most of the route.
Does anybody have any other idea? I do agree that requiring them to be in the middle of the road is not necessarily that important.
That is a good list!
Maybe your “most important” point should be mandatory even for partial?
And I have a question to make sure I understand correctly (I am no native english speaker): What is preference of intersections? Is it priority at intersections? Is it getting green at the traffic light without waiting time?
Also: There are more tagging schemes for bus lanes, see Bus lanes - OpenStreetMap Wiki. Either list all methods or none, so noone gets confused if the not mentioned taggings are ok.
I think this is the point that we need to reach consensus about: do we care more about global consistency and comparability, or do we care more about surfacing relative distinctions that might be important locally?
What specific use cases do we envision for this tag? Is it about rendering certain lines more prominently, or preferring certain lines when the user requests a stress-free route, or accounting for a BRT line’s effect in a road traffic simulation, or analyzing buildout of BRT concepts around the world, or giving mappers a BRT route preset in an editor because the existing presets are unsatisfactory? Each of these use cases may require more or less flexibility about different aspects of BRT operation.
@Nielkrokodil yeah, preference AT intersections, corrected. Also corrected the link to bus lanes, good point. In practice, ti can be eigther preferential traffic lights (sensors in the road that turn green for the buses) or brdiges underpasses specifically for buses or interssections made in such a way the traffic in opposing directino cannot turn to block the oncoming bus.
For me, it would be enabling of “rendering certain lines more prominently”
The last two would go with it I think.
Nope, we have an adorable one here in Rhode Island:
This is my thinking also. With a brt=yes, we would be describing a boolean when the reality is that there are shades of gray in the BRTness of a particular transit system. This is equivalent to wheelchair=yes, which sounds useful but ultimately conveys no useful information to someone wheelchair-bound – wheelchair-accessible is entirely dependent on the degree of mobility of the wheelchair-bound person! What is actually useful is to tag the specific details.
Likewise, the actual useful thing for a bus route are things like:
fee=yes/no if you have to pay a feehighway=busway versus (traffic-prone) public roadsIn that vein, I don’t think it’s useful to try to categorize bus routes in this way. Many will defy categorization based on their unique particulars.
For example, this week I rode the free Silver Line in Boston. Wikipedia twists itself into pretzels trying to define this thing:
The Silver Line is a system of bus routes in Boston and Chelsea, Massachusetts, operated by the Massachusetts Bay Transportation Authority (MBTA). It is operated as part of the MBTA bus system, but branded as bus rapid transit (BRT) as part of the MBTA subway system.
Is it a bus line, BRT, or part of the subway system? All three? Only a small portion of the route runs on dedicated bus roads; otherwise it’s a bus driving on normal city streets.
brt=yes would provide no assistance here in describing the system.
Haha, the U.S. has both trackless trolleys and trolley-replica buses. The former still operate in a handful of cities like San Francisco and Dayton, Ohio, while the latter are common in cities throughout the country as either a tourism gimmick or as an ironic cost-saving measure.
This one is network=TANK, not to be confused with tank=yes: