It seems legit: the pool is a virtual concept, less visible than for fuel facilities.
No cashier, grocery, shops, toilets…
Then you actually refer to the actual station: the device you use to pay and plug your car.
But the station hasn’t got any unique ID, only the pool and the “charge point” (service equipment, EVSE) actually have.
No cashier, grocery, shops, toilets…
More and more stations do actually have cashiers, shops, toilets, etc.
It seems legit: the pool is a virtual concept, less visible than for fuel facilities.
I would argue that a “pool” is about as virtual as a conventional gas/petrol station. They are often connected through systems, and displayed as one object in apps and such.
I hope I’m not sounding rude or aggressive. I appreciate your feedback.
And that’s great, but it’s side businesses operated by different companies than the charging operator.
Fuel stations use to be one big block with shops operated by the fuel company (at least in my surroundings).
It’s true we need to display pools and not stations at lower zoom levels.
Same here, no rudeness intended and I appreciate to discuss about this topic.
What tags for which objects would you propose, @InfosReseaux?
I would like to point out that Google Maps (which is very popular) refers to the whole charging location as an “Electric vehicle charging station”, at least in North America.
And we could find a better tagging as to merge charging_station and charging_point when the station only has got one charging_point.
I currently miss a tag for stalls, it could be amenity=charging_point for kiss sake.
It is popular but not forced to be consistent.
Nothing actually prevent to label every pool a “station” but it would be awkward.
I think we have to be cautious here. To say there is a “standard” is a stretch at the moment. We’re still in the early(ish) days of EVs. This was the first time I came across eMI3 but I’d heard of OCPI before. The latter is more recent which suggests to me that there was something the industry didn’t fully support in eMI3.
In the absence of a unified standard, use of common words makes most sense to me. I’ve never heard people refer to a charging pool.
OCPI is based upon eMI3, at least for IDs.
eMI3 is not very visible but obvious.
I agree with this. Maybe we can’t follow “standards”, as they do not necessarily exist and are complicated. Using common terms that most or all people understand with minimal extra research would be better, in my opinion.
Am I missing something here? amenity=charging_station has already been used nearly 90,000 times. We are long past the point where we can make the decision what a “charging station” should be without looking at how it’s already used.
NKA’s analysis shows that in some cases (about 6,000) people clearly used it to mean a single charger within a larger site (we know this because we can see multiple charging_stations close to each other). In twice as many cases (about 11,000) people probably meant a larger site (we know this because they are tagged with capacity > 2). In the overwhelming majority of cases (tens of thousands), the mapper never had to make the decision, because there was only a single charger to map (serving 1-2 vehicles). Either way, the existing tag is already ambiguous. Unless you want to start a massive re-tagging programme, data consumers will have to deal with both situations.
There are two options
Neither is perfect. Which one, practically speaking, is the lesser of two evils?
I would say #1
I agree that neither is perfect but that’s the reality of mapping something that relates to a new and evolving industry.
While it means re-visiting a lot of OSM, by country it is manageable and we’ll be doing it anyway to add official reference IDs when open data on live availability becomes available.
Like @jmarchon I feel that #1 is the better choice here.
A station is not an individual charger, despite it’s a device.
Look at example Charging stations (sites or individual chargers?) - #2 by InfosReseaux
There could be 2 or 3 individual charger on the same station.
As proposed upside, we need new tags for individual chargers AND pools.
A station is between them both.
So for me it’s #1 + #2 !
Personally, I find this very confusing. Many others will likely also. I think it is better if we use a system that makes sense to the layman who is not experienced in this topic.
Oh well, the terminology is confusing. In my post I used “charger” to mean what you call a station. The same way you might have a USB charger that can charge four devices at the same time.
I was in the same position a few years ago and now it makes sense.
We shouldn’t confuse making sense to the layman and distort reality to match our own interpretation.
Understood, your USB dock is an actual station indeed (because every port is independent).
I find it confusing due to this “EVSE” concept. If I have understood it correctly, each physical asset (let’s call it a “pump”) can have one or more EVSEs. And each EVSEs can have one or multiple sockets/cables but only one of those can be used at a time.
It’s a bit like a fuel pump which may have two sides to service two lanes, and each side can have multiple fuels (petrol, supreme petrol, diesel etc). But on each side only one fuel can be used at a time. We don’t map that complexity in OSM at the moment!
And as the EVSE seems to be a concept not a physical feature, we should focus on mapping the physical feature with the main tag, and the conceptual feature (the EVSE with sub tags as necessary, e.g. capacity and socket info).
I agree with @RobJN, currently we almost don’t map EVSE, except with capacity=*
We could map EVSE on each corresponding individual park spaces, but we currently don’t.
We should focus on pools and actual stations.
Given problem is stations don’t have unique IDs.
It makes sense to me, I just don’t believe it is the correct way of tagging in OSM. I believe we should map the physical reality as ways and nodes and add the “technical reality” that you describe as tags. Which to me means station (for the site) and chargepoint for the physical thing. Tags on that for the EVSE concept.
Signing off for today. We seem to have a lot of agreement here. Just stuck on choice of words (“station” or “pool”, and “charge point” or “station”). Not sure how to break that impasse at the moment but sure we’ll get there.
I think we should not be too hung up on which terms are being used in technical, conceptual standards. What matters, I think, is to get a tagging scheme which is intuitive for OSM users so that charging features will be mapped consistently in the future.
I believe tags are intuitive when they reflect how physcial objects are being described in everyday language, and even better if they are well understood by non-native english speakers. In the Scandinavian languages, for example, it is helpful that the two objects are called “ladestasjon” (site) and “ladepunkt” (charge point). I have never seen “pool” being used.
Also, it will be helpful if the tags resembles other tagging schemes already widely used in OSM. As already discussed, there is a clear analogy to amenity=fuel_station for the site. Also, we have amenity=bus_station, which is a site that can have a collection of bus stops/platforms.
Finally, amenity=charging_station is already used in OSM 90.000 times, and the numbers strongly indicate that a large majority of that mapping is for a site (more than one charger).
In conclusion, I think we will be well served with the two proposed features, amenity=charging_station for a site plus the optional man_made=charge_point.
I agree that technical details about EVSEs and connectors should be documented using tags on the two proposed objects (which we already are doing), rather than introducing even more objects.