I believe that these are often pelican/puffin/toucan crossings, but not always.
Yes, it is possible. We have a few of these in London where the junction is controlled by traffic signals and the crossing has dropped kerbs, tactile paving and dots across the carriageway. There is no button and there are no pedestrian signals, so you have to look at the traffic lights to determine when it is safe to cross. This is where the proposed crossing:signals=shared might be useful.
The nearest example to me is Fairfield Road E3 at its junction with Bow Road Bing street side imagery
It not possible for a junction to have pedestrian green man signals, and no call button. The call button âboxâ also provides crossing information for visually impaired pedestrians by means of the spinning cone under the box. (Though Local Highways Authorities will routinely break rules)
If a junction has several phases of traffic, is is possible to have call button box where the button has no effect on the traffic phases.
It is common in the UK for there to be junctions, especially T-junctions, where road traffic is controlled by signals, but there are no signals for pedestrians.
This junction is a
âSignal-controlled junctionâ for vehicles, and
"Uncontrolled crossing" for pedestrians.
Many (most?) drivers misunderstand the rules regarding green/red lights. The law creates a prohibition regarding crossing the White Line. You MUST NOT cross the White Line unless the light is green AND it safe to do so. Emphasis is given that you must not cross the white line if pedestrians are crossing within the junction. This means pedestrians have right of way at all times when crossing at junctions with light for vehicles, but no lights for pedestrians, but in practice drivers believe the green light gives them priority over pedestrians.
This crossing for pedestrians is an âUncontrolled crossingâ, which Iâve talked about in my previous post. Itâs a good example of a junction where pedestrians always have priority due to the presence of a solid stop line for vehicles. And as I said there is a historical problem with drivers incorrectly believing a Green Light gives them priority through the junction.
The traffic planners (probably TfL in this case) agree with you, as the tactile paving is buff rather than pink for the crossing I gave as an example.
That leads to another question: how do we tag this to tell pedestrian/VI/wheelchair routers that traffic flow through the junction is controlled by the traffic lights, but that the crossing itself has no traffic signals?
The historical problem with driver understanding is because the Driving Theory Test is based on rigorous application of Caucus Race standards.
Yes. Theyâre something different legally, but practically speaking there are a lot of commonalities.
There were never equivalents to pelican crossings because signal-controlled junctions in the UK have never had flashing amber phases. However the resemblance to puffin and toucan crossings is extremely strong.
The signal-controlled junctions donât have the white zig-zag lines on their approaches. However in every other practical way theyâre identical to puffin/toucan crossings. Same call boxes. Same pedestrian signals.
Near me thereâs a crossroads with three effective puffin crossings and because a cycle track crosses the fourth arm of the junction an effective toucan crossing there. Theyâre tagged accordingly.
Oh, joy. Weâve got a new âupgradeâ being blindly implemented by Rapid users.
Theyâre trying to synchronise the tagging between crossing nodes and crossing ways, but unfortunately giving precedence to the tags on the way over those on the node. The result in London is that armchair mappers on other continents who will never actually use the data are unthinkingly degrading crossing=traffic_signals nodes to crossing=marked.
The changes tend to be made in combination with added sidewalks etc., so itâs more something to find and fix rather than revert at the changset level.
A highway=crossing + crossing=marked node may be suspicious if itâs combined with any of: button_operated=yes, crossing_ref=pelican, crossing_ref=puffin, crossing_ref=toucan, crossing_ref=pegasus, crossing:signals=yes, crossing:markings=dots, traffic_signals:sound=yes, or traffic_signals:vibration=yes.
A minority of these may really be the type of uncontrolled crossings at traffic lights discussed earlier in this topic, so aerial and sometimes street-side imagery is needed to confirm.
@rskedgellâs most recent edits morphed the table into something highly specific to the UK. That may serve this quarterly project in the short term. However, in the long term, it risks repeating the sordid history of mappers abroad diluting the meaning of UK legal terms to the point of meaninglessness.
For the semantic protection of Britainâs street fauna, I split out a UK-specific table at âCrossings in the United Kingdomâ based on these edits, then reverted the original table to serve the rest of the world more generically. I also took the opportunity to distinguish between the 2018/2019 proposal and 2022 proposal, trying to show how the latter skirted around the controversy over the former. Hopefully the revised tables will be a little easier to digest and avoid the confusion thatâs apparent in the revision history. Feel free to correct and expand the new UK page as needed.
Fine, if it spares us from silly iD-isms like crossing:markings=yes. If a mapper can see what the markings are, they should tag something useful. If they canât see the markings, they should ignore what iD and Rapid tell them and leave the tag unset (unfortunately, they wonât).,
A table which had, and still has, a column for UK-specific crossings is UK-specific.
I agree that that should be the case, but unfortunately people keep seeing UK-specific tagging guidelines and thinking they can wantonly apply them elsewhere in a completely different context where different distinctions are needed, just because all the tags are supposed to sound like British English. Thatâs how both zebra and uncontrolled came to have different meanings globally than in the UK. It didnât help that the table was located on a global page, so I spun it out to avoid confusion.
Out of curiosity, are the crossing:markings=yes occurrences making it more difficult for you to query for underdetailed crossings? Just trying to understand the practical implications so we can potentially come up with a mitigation strategy.
I think perhaps the problem with the table now at Crossings in the United Kingdom - OpenStreetMap Wiki is that it appears to classify crossing types based on marking and signal configurations in the first four columns, rather than the well-known UK crossing types.
I think it would make more sense for there to be a new column on the left called something like âCrossing Typeâ, that gives the animal name (where it has one) and possibly a second new column with a description of how to identify it. Then the later columns become attributes of the type, and instructions on how to tag it.
Iâll make this the topic of the OSM UK Online Chat next week. I welcome examples and discussion on how to find, tag and update (and what not to do?) on crossings.
Could use an Overpass query to find potential Pegasuses. Assume thereâd be a bridleway leading up to some of them at least
Put on the right Wiki page, these photos photos would help identify the many things on a crossing. Think ahead for them being included in StreetComplete/SCEE too.
One thing thatâs bugging me right now is that Vespucciâs presets currently have this confusing thing: a âTactile mapâ checkbox under âIn case of traffic signalsâ that results in a traffic_signals:minimap=yes.
Iâve never seen any of those in the UK. The wording, however, always makes me want to select it for the vibrating dangly cone thingy (traffic_signals:vibration=yes), which we have on almost every button-operated crossing installed in the past 20 years, if not longer.
Iâve had a go at doing this at https://wiki.openstreetmap.org/wiki/User:Rjw62/UK_Crossings . It may need some more rows adding, and the tagging columns probably arenât complete or 100% accurate, but what do people think of the idea?
