[Tagging] Was public_transport=platform intended to always be combined with highway=bus_stop?
winfixit at gmail.com
Tue Jul 30 09:31:46 UTC 2019
A bus stop, a place where a bus halts to pick up and drop off passengers is
both real and current. Tying it to a geographic object can be done in
various ways, as we've shown over the past years.
I read the wiki a few times over the past years and then I started looking
for something that works, both for mapping the stops and for adding them to
the route relations, WITHOUT
On Tue, Jul 30, 2019 at 3:39 AM Joseph Eisenberg <joseph.eisenberg at gmail.com>
> I think you might be referring to this proposal from Zverik last
> summer, which suggests stopping using
> public_transport=stop_position/platform/station, but keeps the
> - =stop_position is not really needed for routing; doubles work for mappers
> =platform is ambiguous; use =bus_stop, highway=platform (if the bus
> stop has a physical platform) railway=platform etc
> = public_transport=station is not specific, use amenity=bus_station,
> railway=station etc).
> re: > "It's hard to ascertain precisely why PT was originally created"
> According to this comment, it started with imports, and a dispute
> between English mappers and a few Central European mappers (or just
> one?): back in 2010 there were 200,000 highway=bus_stop mapped beside
> the road in England, at the location of the bus stop sign. But there
> was data available in Switzerland that could be imported that had the
> node in the center of the highway, probably for bus operator purposes,
> and a mapper started importing these and changed the wiki to say this
> was better. There was a dispute about this. To resolve it, the
> proposal that was eventually approved created specific tags for
> stop_position (on the highway) which could use access tags like
> "bus=yes" to specify the vehicle and "platform" (for the bus stop),
> and a stop_area relation.
> This wasn't sufficient information to render bus stops differently
> than tram / light rail platforms, so the original tagging method
> remained more common up until now.
> This hasn't stopped some mappers from claiming that there is a
> "version 2" of mapping for transit which should replace the "old"
> tags, and editing the wiki pages to put this view at the forefront,
> going so far as to suggest public_transport=platform and =station for
> ferry terminals and taxi stops, where this tagging is very rare.
> I've made to various wiki pages to describe the current situation. I'm
> also working on making specific wiki pages for tags like bus=yes (used
> for both access restrictions and to specify the type of public
> transport vehicle at a feature).
> On 7/30/19, Dave F via Tagging <tagging at openstreetmap.org> wrote:
> > Hi
> > This is not a criticism of Joseph.
> > This post confirms what I've been saying for the past year - PT tags add
> > nothing but confusion to OSM, which directly leads to errors.
> > highway=bus_stop is a completely separate tag to any in the PT schema.
> > It was created long before the invention of the PT schema and is the
> > original & the most popular, accurate way to map a bus stop.
> > The PT schema purely duplicates existing, well used, clearly defined
> > tags. It adds no extra detail or information.
> > A platform is a raised physical object compared to the surrounding area
> > to aid vehicle boarding. Popular tags such as railway=platform,
> > tram=platform are used for such entities.
> > Public_transport=platform has been hijacked by PT to falsely represent a
> > bus stop as an imaginary area on a pavement. As defined in OSM's welcome
> > page: "OpenStreetMap is a place for mapping things that are both /real
> > and current/ ".
> > It's hard to ascertain precisely why PT was originally created but it
> > appears to be that the existing tags weren't complete. However instead
> > of adding that missing data, somebody though it would be a great idea to
> > start from the very beginning with a completely new set of tags & try to
> > paper over the gaps. The irony is that, after a lot of discussion over
> > tag names & locations & quickly waning enthusiasm for adding them to the
> > database, PT is *less* complete than the original data. What a waste of
> > time. It's a mess.
> > Over on the transit forum the PT schema has become so convoluted even
> > those who helped create it are baffled. At least one is advocating its
> > removal.
> > It's time for the PT schema to be redacted.
> > DaveF
> > On 29/07/2019 15:29, Joseph Eisenberg wrote:
> >> I read up on the rather exhausting history of public transport tagging.
> >> The strange thing is that the approved proposal which introduced
> >> public_transport=* and the current public_transport pages suggest
> >> using bus=yes only for public_transport=stop_position. In contrast,
> >> public_transport=platform doesn't have bus=yes added.
> >> Does this mean that tagging highway=bus_stop on the same node as
> >> public_transport=platform is the approved way to specify that a
> >> certain "platform" is a bus stop?
> >> It certainly looks that way. Does this mean that tagging
> >> public_transport=platform + highway=bus_stop was the tagging that was
> >> intended by the proposal to specify bus stops, and
> >> public_transport=platform + railway=platform for train platforms, etc?
> >> It appears that the proposal specifically said that the existing tags
> >> like highway=bus_stop were not deprecated. Current usage confirms
> >> this: in the past year just as many (or perhaps slightly more?)
> >> highway=bus_stop have been added as a public_transport=platform -
> >> about 350,000 each - though the latter tag also can be used for
> >> railways, trams, etc.
> >> Joseph
> >> _______________________________________________
> >> Tagging mailing list
> >> Tagging at openstreetmap.org
> >> https://lists.openstreetmap.org/listinfo/tagging
> Tagging mailing list
> Tagging at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Tagging