OpenStreetMap

OSM in many areas has a complete street network and has for quite some time. What it is lacking compared to many map sources is a comprehensive POI index. Stores could be there but the data likely is outdated, even then you could have a major chain business that isn’t even shown on the map. So what do we do?

We don’t need some complex editor, just a stripped down editor that can leave a poi and has a search for presets and chains that makes adding one as simple as possible. From there you can add more information but by default it should just show the basics. No distractions and dead simple so anyone with a phone or on the web can easily add a tone of POIs for even a simple quick walk.

Secondly, POIs need to expire and autodeleted; to keep the information from getting stale POIs need to be checked into or deleted automatically after a year if there is no new activity. Mappers should be able to subscribe to an area and get emails warning of pending expiration. The checking API could be open to app developers.

Thirdly, we need a TEMP layer for maps a temp data layer that can be deleted daily that basically is the dumping ground for temporary data such as road obstructions, traffic and events. Basically a meta database that allows apps to share and make use of more realtime data. This can allow for more uses of the OSM map and by contributing to this could reveal closed bridges and other obstructions and improved the map. The more data that is shared between apps the more that can be improved.

Comment from DeBigC on 13 January 2022 at 23:36

I don’t mean to hijack your thread but on the first part, what is it that makes so many of the useful apps collapse? This was one of my favourites and isn’t maintained anymore. Maps.me is turning to mulch too, and StreetComplete, while fun will only ever be android.

I don’t like POIs because sometimes you have POI measles, which is when a building changes use, and nobody takes responsibility for cleaning up the item that should be retired. I prefer to place businesses into the available building tags. I even convert the POI measles into building tags. I think we should regard POIs as a temporary thing, with the business details placed into the building where it belongs.

Comment from redsteakraw on 14 January 2022 at 01:26

DeBigC well POI measles is what the expiration and checkins would prevent. Checkins would be a simple confirmation it is still there or someone visited it and expiring / flagging for autodelete would get rid of old POIs with no interaction. thus your POI measles would be handled automatically. I know that there are a lot of businesses that simply aren’t there and the turnover rate of restaurants are pretty high. Covid lockdowns and restrictions killed a whole bunch more so there has been a big turnover all over the place. As for building it into buildings, well things can get messy especially when you have more than one level. Adding to the building takes more time effort and care I would rather have the POIs there than not.

I think it also comes down to how everything is mapped locally. Most places have the POIs in the buildings. Which technically would be true the business is located within the bounds of the building. I know some people prefer placing it on the entrance and some want even more than that. I think we all don’t want tones of old POIs or stale data this was some ways to solve the problem.

Comment from Zverik on 14 January 2022 at 07:03

First suggestion is what I’ve almost finished in my app :) Outlined the idea in this SotM talk.

Comment from SimonPoole on 14 January 2022 at 09:20

@DeBigC in general it is technically preferable to map POIs as individual nodes, in fact putting tags on buildings is only acceptable if it is the only POI associated with the building which most of the time it isn’t. Definitely you shouldn’t be basing your mapping on how a specific map rendering looks.

@redsteakraw the collection problem has been addressed, if not solved many times, going back to the mapzen POI collector (not related to the defunct mapzen company). The problem is more that people continue to ask for better hammers, but don’t actually want to do the hammering.

In practical terms Vespucci has shown outdated POIs for ages, functionality that was added to StreetComplete financed by the OSMF in the first microgrant round roughly a year ago. Not to mention that there has been numerous incarnations of outdated POI maps over the years.

And while an unpopular solution, in tag only mode in Vespucci (a mode which has existed over a decade), the effort for adding or editing a POI is minimal without having the complexity of being able to change geometry, but still being able to add new objects, and most important you can still see all existing POIs (which is were navigation apps tend to fail) .

But naturally both OsmAnd and organic maps / maps.me are valid solutions for POI collection if you are careful.

In summary there are plenty of tools available and if they were actually used, discussing refinements would be a valid point. As is, you are just asking for somebody to build you another hammer that will rust away in a corner.

PS: you can actually add POIs with voice commands in Vespucci, a experimental feature that has been revamped in version 17 to make it easier to use.

Comment from SimonPoole on 14 January 2022 at 09:39

PPS: in the end what we really want is for operators of POIs to add and maintain them themselves, leaving just the remaining gardening to OSM mappers. While the association with tech giants (Apple, FB, MS and so on) has its pros and cons, it definitively has led to more awareness in business communities and here we are seeing a clear interest in being present on OSM (yes I know FB et al don’t actually use OSM for POIs, but that can change).

Self maintenance is easy for example with https://osmybiz.osm.ch Obviously this could be even further simplified, but again what about using what we already have and improving that.

Comment from ᚛ᚐᚋᚐᚅᚇᚐ᚜ 🏳️‍🌈 on 14 January 2022 at 10:05

Agreed about an easy & fast way to add PoIs when walking around a city. It’s something we’re lacking in. OSMAnd has a… clunky approach to do this. I think we “just” need the apps to do it, rather than changing the OSM data model to delete old data.

StreetComplete is awesome. Alas, it doesn’t support adding new PoIs. They have a good way of checking if data needs to be updated (that fits well within existing OSM conventions). I dunno why “android only” is a problem, Android is by far the most popular type of phone.

Comment from Zverik on 14 January 2022 at 10:47

Simon, I agree that we all want for the map to update itself: for shop owners to add their shops, for road builders to map new roads, for forestry officials to update forests on OpenStreetMap. But, living in Eastern Europe, I see hundreds of shops and amenities that I’m pretty sure won’t bother themselves adding to any map. And if to look at less developed countries, even having time or opportunity to internet there are small. Yeah, I want other people to do my “job”, but I don’t see that happening like ever.

So tools like osmybiz perpetuate this privilege that e.g. Frederic mentioned numerous times: when you, as an owner, have enough time and money, you can get your business much more visible and popular through marketing companies and by doing some things by youself. But we want not only successful businesses on the map, but every shop and every amenity.

Hence we need tools to add these. Maps.me helped with diversity, adding amenities in distant countries. My editor will help with completeness, adding all shops and amenities where there are experienced mappers (mostly USA and Europe, that means). But for a definite solution to this issue — I still don’t know how it could be done.

Comment from o_andras on 14 January 2022 at 11:11

Re 1st: that’s a cool idea. What I currently do is I keep a list of things I commonly add to the map, things that usually have mostly the same tags (e.g. recycling containers, trash containers, business chains, &c). I just copy&paste, make any required changes, et voila. Works well enough for me.

Re 2nd: I don’t think that’s a good direction to go… I’ll assume you’re talking only of business-POIs (retaurants, shops, cafes, &c). First, adding POIs requires much more effort than removing them. Then, some businesses really don’t change details often, if ever, and they keep on going. I have examples of POIs that have been added 9~10 years ago that are still open today (and I know they’re even older than that) - for examples of this just look at big chains, though they’re not the only ones. Plus, if you’d delete POI after some time what you’d get is worse coverage. It requires even more hands and time, and that’s not something you can just go to some shop down the street get more of. Something more useful is to be warned of outdated POIs (and @SimonPoole already mentioned that).

Re 3rd: I also really like this one! It would open up a lot of possibilitied, like “real-time” parking info, and opening hours (sometimes businesses make exceptions on their open schedule). OsmAnd has an “avoid road” feature, but it’s offline.

Comment from SimonPoole on 14 January 2022 at 11:30

Further point that @DeBigC raises “… what is it that makes so many of the useful apps collapse?”.

It is the “many” in your question.

Conservatively estimated a conventional general purpose OSM editor will require at least 2 FTEs to maintain long term, something simpler more (providing a “simple” UI is -more- complex, not less). 1 FTE for a limited functionality app (though that is probably too low).

For the sake of this argument, let’s assume that a FTE is $200’000 annual total cost to their employer, or themselves, if not paid for the work.

Have a look at editor stats, annual costs per user using above assumptions for the top couple of entries are

General purpose editors

iD + Rapid $2
JOSM $19
Vespucci $88
Go Map ! $106
OSM Go $576
Gnome Maps $680
Potlatch 3 $775

Restricted functionality editors

SC $9
maps.me + om $9
osmand $25
OsmHydrant $144
MapComplete $278
Pic4review $451

To be clear this is what you would have to assume in cost for continued maintenance to keep things current once the initial enthusiasm has worn off. Obviously most apps do not get so much attention and therefore slowly bit rot and fade away. Further dividing both user base and developer resources just makes matters worse.

Try thinking about the numbers as a yearly subscription you would have to pay to be able to continue to use the app.

Comment from Mateusz Konieczny on 14 January 2022 at 11:53

Maps.me is turning to mulch too

Use Organic Maps - made by some of original authors, with nasty parts like trackers, data stealing and ads eliminated.

https://github.com/organicmaps/organicmaps

Comment from ᚛ᚐᚋᚐᚅᚇᚐ᚜ 🏳️‍🌈 on 14 January 2022 at 12:59

@Simon You’ve often mentions how “Vespucci can do that!”. Perhaps there is issues in informing (potential) users about these features. Have you considered making how-to videos?

Comment from SimonPoole on 14 January 2022 at 14:52

@᚛ᚐᚋᚐᚅᚇᚐ᚜ 🏳️‍🌈 sure I have. The problem with that should be clear.

Comment from redsteakraw on 14 January 2022 at 15:13

SimonPoole don’t get me wrong Vespucci is a very powerful app, however the UI is pretty complex and is mostly geared towards power mappers. What I am talking about is simply a dead simple UI that makes adding POIs either on the go or from home dead simple. It would have local chains and presets built in and a simple search would get you the store type or preset. With many apps the UI gets in the way, we know what the problem is and just need enough of the UI to get the job done, no editing polygons, nor ways nor relations just adding a POI. It should be simple enough for your parents to do it. Vespucci will still be there for the power mappers it has it’s place for sure.

Now on showing old POIs on the editor well that is good but having a Bot that would auto delete them may be the path forward if paired with some sort of check-in or it’s still here API. The problem is most people won’t see the staleness of the POI and without a check-in we have no idea if a POI is still there especially if there is no reason to edit the POI. OSM is used by a bunch of mobile apps and this would be a good thing that could give back to OSM by just confirming yes it is still there or I just navigated there and I am there. No editor needed. The question is what is worse possibly deleting good data or keeping old outdated data on the map. Covid has really changed the business landscape worldwide there are tones of closed locations or new locations this is needed now more than ever.

Comment from redsteakraw on 14 January 2022 at 15:24

Zverik I saw your Video and the UI looks great and the idea is right on track. In order to have a Map-less UI you need complete and accurate data. Your app seems to make this easier however it would be even more accessible if it was a WebApp instead of relying on Telegram and at least the tagging / data side used OSM / OSM schemes. The pictures yeah that could be hosted anywhere. IPFS, LBRY, Telegram, openstreetcam, ECT.. Any friction or extra step from gathering data to pushing it to OSM will lead to a dropoff in the follow-through. At least adding something on the spot would be better than not. I like what I see and it looks very interesting.

Comment from redsteakraw on 14 January 2022 at 15:33

@ andre-sa Yes I know deleting something could be bad which is why it would have to be paired with some sort of Check in or It’s still here API. So that popular old restaurant could easily be checked and passed over for any deletion. Mobile apps and simple editors could expose this and make the whole thing transparent. The problem is what is worse removing a small amount of good data or having a lot of unreliable data. The thing is if you have unreliable data it taints the good data, as the system as a whole can’t be relied on. You go to a new city lets say and go to look for a local pub, and get routed to a closed one or maybe it’s a bakery now. There is no point to having the data if you can’t rely on it. You can only rely on it if there are some mechanisms that make sure that the data says fresh. Covid has upturned a lot of businesses and the problem is worse now than ever.

Comment from SimonPoole on 14 January 2022 at 16:59

@redsteakraw my point was more that you are asking for an extremely complex editor, far far more complex than any other one, just alone handling all the edge cases will be a royal PITA.

I only mentioned Vespucci because you don’t have to have geometry editing enabled in Vespucci, you can even directly start it in tag editing mode and that is from a functional pov what you were proposing, but it doesn’t hide complexities of OSM modelling from the user and doesn’t want to.

The other point though is that there is only a very limited pool of mappers, and every additional app segments that small pool further, making only the absolute largest ones sustainable once the hype has died down and the only thing left is work. So my expectation that we will end up just with iD and maybe one nav app with some add on editing in the longer term.

Comment from redsteakraw on 14 January 2022 at 18:19

@ SimonPoole not asking for a complicated editor it could be a stripped down iD or a gloriffied script it just needs to find and make it easy to add preset POIs no UI for editing ways or polygons or complex map rendering. Just I am in front of a POI and add it where I am. I will check out Vespucci some more to see what you are talking about but this needs to be so simple a casual person could trivially add it. This is something that OSM apps can link to and possibly allow for new mappers that aren’t all that advanced but can literally just add that restaurant or local place. This wouldn’t be a full blown editor just a specialized POI add or POI check. OSM for the most part has a complete road network. Maps.Me started this but it never quite got there.

Comment from Mateusz Konieczny on 14 January 2022 at 18:40

not asking for a complicated editor it could be a stripped down iD or a gloriffied script it just needs to find and make it easy to add preset POIs

making it properly requires significant effort

just hiding some features from iD is not enough

Comment from Mateusz Konieczny on 14 January 2022 at 18:48

So my expectation that we will end up just with iD and maybe one nav app with some add on editing in the longer term.

At least JOSM, level0, StreetComplete serve different needs and would stay.

Even if iD would need to be actually usable on mobile. And it seems to be very far away from becoming a real replacement for Vespucci and not becoming closer to that. And it seems fundamentally lacking offline support at all, which is a critical Vespucci feature.

While with - very significant investment - iD can be made better than Vespucci in all relevant aspects and therefore replace it, and on timescale of decades/centuries it will happen or both will be replaced by a new editor. If OSM will exist for so long time and it is a very significant if.

But I cannot really imagine it being still relatively newbie friendly (while trapping them with some leaky abstractions) and having JOSM features and performance.

And merging JOSM, iD, level0, StreetComplete into one editor seems not really feasible.

Comment from Mateusz Konieczny on 14 January 2022 at 18:58

You’ve often mentions how “Vespucci can do that!”. Perhaps there is issues in informing (potential) users about these features.

I think that problem is that many users are scared away by overly complex interface. I like Vespucci, I edit with Vespucci a lot, I tried to convince people to use Vespucci but all that tried basically run away.

While I had some success with iD, StreetComplete and JOSM.

Initial experience really scares away people and there are other parts where I would consider removing functionality to keep interface more usable. For example “keep changeset open” checkbox.

https://github.com/MarcusWolschon/osmeditor4android/issues/899#issuecomment-886530450

Initial startup after the first install has very confusing screen with space for manual coordinate input, ability to search for location by name (with dropdown for Photon/Nominatim).

Why not just zoom to current location?

Or have a big button “edit in my current location” that would zoom in and download data in one click.

What is the difference between “load” and “go to map”?

Why selecting “current location” is not doing anything (I need to also click one of the buttons).

Why “current location” and “last known location” even show lat/lon?

Why it would be clear for a new user?

I would consider seriously simplifying this form, I am not surprised if many people would fail to ever download data after installing.

Comment from redsteakraw on 14 January 2022 at 21:00

Mateusz Konieczny maybe it would require a new codebase but some sharing like iD’s preset scheme could be reused. This wouldn’t even need to display a map and if it does a single OSM tile would suffice. No vectors no complexities just a little db of presets and search bar to search between them. current location puts the POI and a bunch of preset text boxes / ui for standard input from input based on the presets.

Comment from o_andras on 14 January 2022 at 21:31

Re @redsteakraw

The problem is what is worse removing a small amount of good data or having a lot of unreliable data.

I’d say easily (as I said on my previous comment) that removing good data just because it may be bad is much worse than having bad data.

The thing is if you have unreliable data it taints the good data, as the system as a whole can’t be relied on.

Yeah, I understand that, but removing good data doesn’t take you anywhere closer to your (our) goals – in the end, you’re throwing away what you’re looking for.

(…) which is why it would have to be paired with some sort of Check in or It’s still here API. So that popular old restaurant could easily be checked and passed over for any deletion. Mobile apps and simple editors could expose this and make the whole thing transparent.

That makes much more sense. StreetComplete already does that for opening hours. Maybe they’re open to adding another quest type for outdated POIs and other data.

Comment from Mateusz Konieczny on 14 January 2022 at 22:46

Maybe they’re open to adding another quest type for outdated POIs and other data.

“Is this still here?” quests exists for some nodes ( https://wiki.openstreetmap.org/wiki/StreetComplete/Quests )

For areas it gets trickier and require a full scale editor to handle things.

Shops that are discovered to be gone during opening hours quest can be retagged to another POI or marked as vacant property.

Notes can be always left for all quests, typically used when object being asked about is gone.

Comment from Mateusz Konieczny on 14 January 2022 at 22:47

This wouldn’t even need to display a map and if it does a single OSM tile would suffice.

This would end in creating duplicates and would be therefore a poor idea.

Comment from SimonPoole on 14 January 2022 at 23:03

@Mateuz I was pointing out that economic basis for other editors will vanish and that they are not really sustainable even now. That has nothing to do with if they do something slightly different or target a different audience.

As to your bit about beginners using Vespucci, thats akin to complaining that chainsaws are not designed for children. No they aren’t and it even says so on the box.

Comment from mmd on 15 January 2022 at 15:59

In case this temp layer thing should be part of the osm.org stack, this has been suggested by a several people in the past. Frequently mentioned use cases are facilitating imports, QA, approval workflows, conflation. All the usual suspects.

None of this is really feasible, as the entire stack (including all data consumers) is not prepared for such things. And it would certainly be a huge effort to add it.

Comment from SimonPoole on 15 January 2022 at 16:14

@mmd not only that, it would be weirdly at odds with the global trend to use map matching to transmit and store such information (that is OpenLR and similar, some actually using OSM data).

Now something that could actually make sense is build our own infrastructure and data collection around such a concept, including enabling providing an API for clients to access the data. It is unlikely that the OSMF would ever take something like that on, but a separate project could easily do that.

Comment from H@mlet on 17 January 2022 at 15:34

About the third part, you might want to check out OpenEventDatabase.

I don’t think it’s actively maintained anymore, but has a lot of ideas implemented.

Regards

Comment from DeBigC on 17 January 2022 at 23:44

@SimonPoole “Definitely you shouldn’t be basing your mapping on how a specific map rendering looks.” That’s not at all what I am doing. I am mapping based on the confluence of contradictory pieces of information, old businesses/services that are gone or renamed where every new POI placed in the same location is placed there without review of what is also there.

Comment from Pieter Vander Vennet on 17 January 2022 at 23:56

The mapcomplete shopping theme is pretty close to what all of you want, I think? MapComplete is built around the concept of a ‘theme’, which can contain various layers and allows to add a new Point. Some care has to be given as sometimes a ‘closely related’ concept might get added with wrong tags, so questions and all have to be crafted very carefully.

@SimonPoole: cool statistics there! Right now, MapComplete is being developed part-time as freelancer. For the very first incarnation, I was around $5/changeset - but your figures won’t be too far of.

Comment from redsteakraw on 18 January 2022 at 15:00

@Pieter Vander Vennet the Mapcomplete shopping theme has a good UI, just wondering where they host the images? This seems good for adding more information to POIs. Something like this would be good for beefing up existing valid POIs. It also seems to just ask for a phone number but I am guessing it would be trivial to add any field. If I see a shop is empty or closed there doesn’t seem to be anything to mark that and quickly strip the POI tags. It seems like it is very close to being the mobile POI editor. With some preset searches and a bit of tweaking this could be a real game changer. I might try it out with the bench view at a park just to get the workflow down.

Comment from Pieter Vander Vennet on 18 January 2022 at 15:07

Images are hosted on IMGur

There is a delete button, but you have to be logged-in to see it. Points are deleted, ways are stripped of their tags.

A ‘what is the phone number’ is included, but you have to go through the other questions first (click the ‘skip question’-button).

Comment from Dimitar155 on 23 January 2022 at 11:15

https://osmybiz.osm.ch is the editor that your are looking for.

Comment from PlayzinhoAgro on 25 January 2022 at 22:46

the best way is the editors in the navigation apps like Organic Maps because they are easily accessible by everyone who already uses the app and the OSM data, I would also like to see an OSM layer that shows temporary events like Waze


Login to leave a comment