Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[NKA mapper]

As requested, here is my code for identifying potential groups of charge
points:
https://github.com/NKAmapper/charging_sites

The script only provides an indication of groups. It is based on proximity,
and the generated tags of each site are based on the most common tags of
the underlying charge points. I am sure the code could be improved.
Each site should be inspected manually before retagging.

Here is also link to the generated file:
https://drive.google.com/file/d/19pBT9zDDIVWt8Eu0gSXXc15BRQLtrknL/view?usp=share_link

Search for example for "GROUP>10", which will find groups with more than 10
charge points.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[NKA mapper]

Please see comments below.

ons. 3. mai 2023 kl. 09:32 skrev Marc_marc :

>
> can you share our algorithmic ?
>

I can share the script in a few days. It needs to be tided up. In the
meantime, the generated file is available for browsing. Search for the
GROUP tag.


> I don't see how an algo will detect that 2 objects
> amenity=charging_station represent 2 sites on the same parking
> (and are therefore valid in the new schema) and not 2 charging
> stations on the same site (which would have to be changed to
> man_made=charging_station).
>

I suppose you mean man_made=charge_point. The script identifies clusters of
amenity=charging_station elements which are closer together than 20 meters
- they are then candidates for being a site. If the elements in the
identified group have common name, brand, operator or similar tags then
they likely belong to the same site. Some elements will be different and
should not be grouped. A few similar tags for the potential site have
already been created in the file to speed up the process, but to be
verified by a human. I do not think it is a problem that for example two
adjacent charging stations continues to stay separate even if they could
have been grouped. Sooner or later someone will come across them and
consolidate into one element.


> I don't see either how an algo will detect that an object
> amenity=charging_station isolated represents a site of
> several charging stations (and is thus valid in the new schema)
> and not an isolated charging station (which it would thus
> be necessary to modify by adding man_made=charging_station)
>

The script leaves isolated amenity=charging_station element untouched.
There is no need to change them. They represent more than 80% of all
charging stations in OSM. The proposal is designed to not require any
changes to those. They already denote a place to charge an electric vehicle
before the proposal and will continue to do so afterwards. Many of them did
not have any information about the number of charge points before the
proposal and they will continue to not have that afterwards (like many
other things with missing details in OSM). Please note that you do not need
to map man_made=charge_point at all, just one node for
amenity=charging_station for the site. If you are interested in
contributing, I would rather focus on the identified groups of charge
points above.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Andy Townsend]

On 03/05/2023 08:28, Marc_marc wrote:
I don't see how an algo will detect that 2 objects 
amenity=charging_station represent 2 sites on the same parking

(and are therefore valid in the new schema) and not 2 charging
stations on the same site (which would have to be changed to 
man_made=charging_station).


Agreed - i tried to explain the problem at 
https://community.openstreetmap.org/t/charging-stations-sites-or-individual-chargers/96810/198 
.  That resulted in this change 
https://wiki.openstreetmap.org/w/index.php?title=Proposal%3AEV_Charging_Station_Mapping=revision=2504680=2498183 
, but beyond that communication didn't really occur (the people 
promoting the proposal thought that there wasn't a problem).


It's less of an issue here than it might me with other proposals because 
the infrastructure being mapped is getting expanded rapidly, so the "old 
ones already mapped" will be a small, shrinking subset of the total.


What might become an issue is that I doubt that the paradigm chosen here 
that mirrors existing petrol / diesel infrastructure actually matches 
what we will end up with - I suspect that we'll get "almost every 
parking space has a charge point" instead, although that's obviously 
some way off currently.


Best Regards,

Andy



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Marc_marc]

Le 02.05.23 à 22:58, NKA mapper a écrit :
tir. 2. mai 2023 kl. 20:01 skrev Justin Tracey >:



There's no algorithmic way to tell


this is not correct.


can you share our algorithmic ?

I don't see how an algo will detect that 2 objects 
amenity=charging_station represent 2 sites on the same parking

(and are therefore valid in the new schema) and not 2 charging
stations on the same site (which would have to be changed to 
man_made=charging_station).


I don't see either how an algo will detect that an object 
amenity=charging_station isolated represents a site of

several charging stations (and is thus valid in the new schema)
and not an isolated charging station (which it would thus
be necessary to modify by adding man_made=charging_station)




___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Marc_marc]

Le 02.05.23 à 16:14, Illia Marchenko a écrit :

What is the impossibility of migration?


you can't select all object with the previous schéma
if you select amenity=charging_station,
you have both objet with the previosu schéma (for ex
an object representing a charging station site)
and the objects in the new scheme (the object
representing a site composed of several terminals)
if you don't know how to select only the objects of the old schema,
then you can't easily go through them one by one and you are limited
to having to select them a bit by chance, without avoiding that each 
contributor checks again and again each time the objects having the

tag amenity=charging_station common to the old and the new schema



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[NKA mapper]

tir. 2. mai 2023 kl. 20:01 skrev Justin Tracey :

>
> Their point is there's no way to tell when that's necessary without
> on-the-ground insight, and there is no way to tell when such insight is
> needed. Had the proposal broken it out into two new tags, there would be
> an indication of which objects need to be updated. With the approved
> tagging scheme, is this node tagged correctly, or has it just not been
> updated yet? There's no algorithmic way to tell, meaning it will be many
> years before everything that needs updating will get updated.
>

It has already been pointed out by several people that this is not correct.
The only elements which will need conversion are roughly 6000 groups of
charge points, but they are easy to identify. Most of them are already
identified in the linked file.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Justin Tracey]

On 2023-05-02 10:14, Illia Marchenko wrote:



2 May 2023, 16:52 Marc_marc >:


Le 02.05.23 à 14:21, NKA mapper a écrit :
 > The EV Charging Station Mapping
 >
> 
proposal is now approved


I really wonder how we can approve a proposal that makes it impossible
to migrate from the old scheme to the new one when we see that even
those that allow it (power=sub_station) take over 10 years to migrate.
I wonder if a vote should not be planned in two stages:
- the idea: everyone was aware of the problem
- the implementation... that we should not vote as long as there is a
concrete problem (and here there were 2: redefinition of existing tag
and impossibility to select the objects of the old schema since the new
schema uses a tag identical to the previous one).

bad day...


What is the impossibility of migration? Just use man_made=charge_point 
instead of amenity=charging_station for charging points. If the charging 
point is already marked as amenity=charging_station, just change it to 
man_made=charge_point.

Regards,
Illia.



Their point is there's no way to tell when that's necessary without 
on-the-ground insight, and there is no way to tell when such insight is 
needed. Had the proposal broken it out into two new tags, there would be 
an indication of which objects need to be updated. With the approved 
tagging scheme, is this node tagged correctly, or has it just not been 
updated yet? There's no algorithmic way to tell, meaning it will be many 
years before everything that needs updating will get updated.


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Illia Marchenko]

2 May 2023, 16:52 Marc_marc :

> Le 02.05.23 à 14:21, NKA mapper a écrit :
> > The EV Charging Station Mapping
> > <
> https://wiki.openstreetmap.org/wiki/Proposal:EV_Charging_Station_Mapping>
> proposal is now approved
>
>
> I really wonder how we can approve a proposal that makes it impossible
> to migrate from the old scheme to the new one when we see that even
> those that allow it (power=sub_station) take over 10 years to migrate.
> I wonder if a vote should not be planned in two stages:
> - the idea: everyone was aware of the problem
> - the implementation... that we should not vote as long as there is a
> concrete problem (and here there were 2: redefinition of existing tag
> and impossibility to select the objects of the old schema since the new
> schema uses a tag identical to the previous one).
>
> bad day...
>

What is the impossibility of migration? Just use man_made=charge_point
instead of amenity=charging_station for charging points. If the charging
point is already marked as amenity=charging_station, just change it to
man_made=charge_point.
Regards,
Illia.

>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - Approved - EV Charging Station Mapping

[Marc_marc]

Le 02.05.23 à 14:21, NKA mapper a écrit :
The EV Charging Station Mapping 
 proposal is now approved



I really wonder how we can approve a proposal that makes it impossible 
to migrate from the old scheme to the new one when we see that even 
those that allow it (power=sub_station) take over 10 years to migrate.

I wonder if a vote should not be planned in two stages:
- the idea: everyone was aware of the problem
- the implementation... that we should not vote as long as there is a 
concrete problem (and here there were 2: redefinition of existing tag 
and impossibility to select the objects of the old schema since the new 
schema uses a tag identical to the previous one).


bad day...



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging