What is the reference number of this defibrilator?

[peternewman]

General

Affected tag(s) to be modified/added: ref=* on emergency=defibrillator Question asked: What is the reference number of this defibrilator?

Checklist

Checklist for quest suggestions (see guidelines):

• 🚧 To be added tag is established and has a useful purpose - @rjw62 's https://osm.mathmos.net/defib/progress/S/ notes "Note that this matching is not perfect, and if there are multiple nearby AEDs, or an EAD is a long way from its postcode centriod, they may not be matched correctly.". A lot of these have refs, so could be matched perfectly if this data was gathered.
• 🤔 Any answer the user can give must have an equivalent tagging (Quest should not reappear to other users when solved by one)
• 🐿️ Easily answerable by everyone from the outside but a survey is necessary
• 💤 Not an overwhelming percentage of elements have the same answer (No spam)
• 🕓 Applies to a reasonable number of elements (Worth the effort)

Ideas for implementation

Element selection:

emergency=defibrillator nodes in countries where defibrilators usually have numbers (certainly includes UK), which have no ref=* tag, and ref:signed=* (if present) is not equal to "no".

Proposed GUI:

Q: What is the reference number of this defibrilator?

• The reference number is [ ________________ ]
• The number is not shown or is unreadable.
• This defibrilator does not have a number (perhaps it is a private defibrilator).

In the first case ref=* is set to the value given. In the second case ref:signed=no is set. In the third case ref=no is set.

(There has been some discussion about ref=* more generally in #81.) This quest suggestion was stolen from the post box one, #1556.

[matkoniecz]

No idea about quest itself and is it a good idea but for

Ideas for implementation

https://github.com/streetcomplete/StreetComplete/tree/master/app/src/main/java/de/westnordost/streetcomplete/quests/bus_stop_ref
https://github.com/streetcomplete/StreetComplete/tree/master/app/src/main/java/de/westnordost/streetcomplete/quests/postbox_ref
are likely to be useful

[rjw62]

I'm not convinced that this quest would be a good idea. In the UK at least, not all ambulance services assign refs (or actually publish them) to the AEDs in their areas. From the data in my tool (which covers most ambulance services) only 50% of AEDs have a reference number assigned by the Ambulance service. (Where I live, there are no refs in the Ambulance service data. Some AEDs show a reference number from the installer/manufacturer, some show a postcode, but many don't show anything.)

Also note that these refs are being assigned by the Ambulance service and not by the operator of the AED. Depending on the region, it may or may not be compulsory or recommended for the operator to display the reference number on the box. I'm also not sure how mappers would distinguish, on the ground, between the last two options: not having a ref and having a ref but it not being displayed.

Finally, it should be noted that in the UK, there is a move towards a central register of AEDs: https://www.thecircuit.uk/ . It's not clear whether the data will continue to be made available, or if there will be any standard reference number system.

[matkoniecz]

not having a ref and having a ref but it not being displayed.

With SC both would be tagged as ref:signed=no as mapper has no way to distinguish it while surbeying

[peternewman]

Ideas for implementation

Ah yeah sorry, I'd meant to say I imagined it being a complete clone of the postbox quest.

I'm not convinced that this quest would be a good idea. In the UK at least, not all ambulance services assign refs (or actually publish them) to the AEDs in their areas.

I did see there were only refs for some areas via your tool @rjw62 . I don't know if we'd be able to add further localisation to SC using some of the pre-existing geojson's and only ask for certain counties or whatever, to avoid asking where its not provided.

Although I know of somewhere remote with two defibrillators within about 100m of each other, I think only one is in the ambulance data so I imagine your system might struggle to match them.

Finally, it should be noted that in the UK, there is a move towards a central register of AEDs: https://www.thecircuit.uk/ . It's not clear whether the data will continue to be made available, or if there will be any standard reference number system.

Ah, I had no idea about that. They suggest in their FAQ when everyone is onboarded they might have developed a system to search for a local AED from a PC/smartphone. Can you just FOI them at that point? Although it looks like it doesn't apply to a charity.

I guess for the UK at least, we should certainly be guided by @rjw62 . If you don't think it makes sense or don't want to implement it in your tool, there's not much point asking it. You mentioned someone in France doing a similar postbox thing (or wanting refs for that), do you know if they do AEDs too?

[mnalis]

To be added tag is established and has a useful purpose

Pardon my ignorance, but what is the useful purpose of ref of AED to OSM users?

It's existence and exact position is surely valuable info for general population (for saving lives), but only persons who come to my mind that would care about its ref are the folks who maintain them, and I certainly would hope they have they own database of their AEDs instead of depending of OSM.
And even if they want to have that database of refs synced with OSM for convenience, I would hope they use some (semi-)automated tools for that purpose, instead of sending workers with StreetComplete to roam around the city and fill in the refs ?

[peternewman]

To be added tag is established and has a useful purpose

Pardon my ignorance, but what is the useful purpose of ref of AED to OSM users?

As I said in the quest suggestion template (although I'd tagged it against the wrong question initially:

@rjw62 's https://osm.mathmos.net/defib/progress/S/ notes "Note that this matching is not perfect, and if there are multiple nearby AEDs, or an EAD is a long way from its postcode centriod, they may not be matched correctly.". A lot of these have refs, so could be matched perfectly if this data was gathered.

For example given these two nodes, and one listing in Robert's thing, which is the correct defibrilator:
https://www.openstreetmap.org/node/9192012306
https://www.openstreetmap.org/node/9192012301

More generally, it's the same purpose as any ref, to tie OSM to real data:
https://wiki.openstreetmap.org/wiki/Key:ref

Like all of these:
https://taginfo.openstreetmap.org/projects/tag2link#tags

Robert takes it a step further by using an external data source to effectively QA OSM and flag where the gaps are.

I certainly would hope they have they own database of their AEDs instead of depending of OSM.

You could say the same about post box refs, fire hydrant diameters, even the power pole materials etc, unfortunately it's often not the case, or not compatible for some licence reason.

And even if they want to have that database of refs synced with OSM for convenience, I would hope they use some (semi-)automated tools for that purpose, instead of sending workers with StreetComplete to roam around the city and fill in the refs ?

Matching two separate geographically positions accurately and correctly is unfortunately very hard, again see Robert's pages for some demos of why and what he tries to do to work around it.

[mnalis]

Thanks for clarification, @peternewman !

One additional thing to take into consideration is that SC currently lacks something like Layers #2461 (although I think I've seen it mentioned there might be some movement there in future), so is not ideally suited for tasks which would benefit from being able to see other POIs of same type (like in two nearby nodes in your example above), as users could easily be tagging the wrong one (because the other one was not shown, for example). See similar issue in for example #3174 (comment) (there were a few others with similar issues).

Leaving notes with pictures might often be of more use than filling-in ref directly on the node in such cases IMHO. (for example if position is not totally exact or there are indeed two or more AEDs nearby etc)

[peternewman]

One additional thing to take into consideration is that SC currently lacks something like Layers #2461 (although I think I've seen it mentioned there might be some movement there in future), so is not ideally suited for tasks which would benefit from being able to see other POIs of same type (like in two nearby nodes in your example above), as users could easily be tagging the wrong one (because the other one was not shown, for example). See similar issue in for example #3174 (comment) (there were a few others with similar issues).

It's a question of relative closeness @mnalis . Those two are 50-100m apart; I've never had a major issue with postboxes either, where they can sometimes be a metre apart (as long as both show at the same time).

The issue is more that the external systems often just have a postcode in the UK (~40-50 houses or a small street roughly) which could easily encompass those larger distances.

Leaving notes with pictures might often be of more use than filling-in ref directly on the node in such cases IMHO. (for example if position is not totally exact or there are indeed two or more AEDs nearby etc)

It's two different use cases @mnalis . The ref is the only sane, reliable and accurate/guaranteed way to tie a thing in the physical world (a postbox, an AED, maybe even a place serving food) or more accurately an OSM node representing one of those to a row in another database.

Separately you've got the normal mapping challenge of accurately positioning something in the real world on the OSM map, you could use GPS, or your photo, or some measurements, but the GPS may not be hugely accurate, but if the photo or measurement is used to position it relative to a building, was the building mapped from an aerial image? Does the image have an offset and was it considered or is everything technically offset a bit. It probably doesn't generally matter, although a postbox or AED could be hidden out of sight behind a door or just round a corner if you're not careful.

[matkoniecz]

@peternewman In which countries/areas it would be useful?

[Lee-Carre]

For Jersey, defibrillators are supposed to have an ID, but not all of them do. So, the utility of enabling this quest is unclear, for the Bailiwick / island. I don't know about the other Channel Islands.

[peternewman]

@peternewman In which countries/areas it would be useful?

Well probably not the UK for a while from what @rjw62 has said (and probably until that new database comes in).

Broadly it's useful in any countries where the reference number data is available as open data (even if the locations aren't) and someone is willing to pull together a site to compare the two.

For Jersey, defibrillators are supposed to have an ID, but not all of them do.

Can you FoI them or equivalent @Lee-Carre to get the list of IDs to know whcih ones you're missing?

So, the utility of enabling this quest is unclear, for the Bailiwick / island. I don't know about the other Channel Islands.

Same question, if you can get the data you can use the list of refs to get a list of probably missing AEDs.

[matkoniecz]

"In which countries/areas it would be useful?" remains open.

Is there some area where

• substantial amount of AED carry some clearly identifiable ref
• there are no AED with other numbers easy to be mistaken for ref
• collected ref data would be useful

?

[matkoniecz]

Closed as no area where this quest is worth asking was identified. If anyone is aware - please share such info.