This is the kind of feedback all such projects get. A million "if you just implement this one critical feature...".
It must be really hard to balance giving all those users what they want, building quality code, keeping the project focused, and having low costs. Is it possible to enable this through extensions, or leveraging a product like OpenCloud?
Yeah, and it's a huge trap, because the authors might believe the users. The truth is, if the app gave you that much value, you would have been using it, and would have found a workaround for the bookmarks export.
Let's be honest, the average user will not do this. In fact, this comment reminds me of the infamous Dropbox one which suggested a similar failure of UX.
As comments in the first link suggest, the project emphasizes privacy, and it would have to be done in a privacy-respecting way.
The second link is asking for iCloud synchronization.
A project that emphasizes privacy and open platforms might want to just deprioritize requests coming from people on privacy-violating, closed platforms, since those people hurt their own credibility in the approaches they suggest.
If privacy advocates want to persuade people outside their bubble, this is surely not the right attitude. Imagine telling someone who wants iCloud sync, "oh you're hurting your own credibility"? Uh yeah (backing off slowly), guess I'll go back to Apple/Google maps and miss out on the ineffable benefit of your non-functional, but deeply virtuous, software.
I’m a contributor to OpenStreetMap, but what I find very lacking is the POI data
It’s only updated if there’s someone like me in the local area, or if it’s a tourist destination. But even then it can be sparse (eg just visited Merida, Bacalar and Valladolid in Mexico), and it’s worse in “non-western” countries
Tellingly, Overture Maps that use OSM for the street data to not use the POI data.
A lot of people pointed out that the POI data of Overture had mediocre quality. While that is true, they do at least have coverage in places OSM do not. If you know the name of the place you can at least find it
I’m not sure how this can become better, I’d think that a prerequisite would be that corporate users like Grab, DiDi, Uber start contributing
But that may not happen as they are not allowed to combine datasets [1]. Hence you have a chicken and egg problem: The dataset is not usable until it is complete…
The street data fundementally have the same problem, but that is easier to edit remotely and may be more stable in contrast to POIs
So I’m starting to question if it’s really worth it to continue update POIs in OSM, but I also would not know how to contribute to Overture as it does have quite some errors :)
What do you think about Street Complete [0][1][2]?
I know it doesn't let you provide all and any data, but I find it super friendly.
It lets you add details to objects that already exist in OSM. It looks very polished, has nice map, icons, and this default style of sticking to your actual location like some location based games for crazy people (I'm a former Ingress player, Enlightened FTW).
My neighborhood has all details about building size, roofs, street lights, you know if there are markings on the ground on street crossings (for people with poor vision), each path has clear info of what's it's made of, and stores has open/close hours provided. Super cool, easy to use and has real impact.
StreetComplete is great for newbies, especially those that need the gamification approach to be more motivated. But if you are already an experienced OSM editor, you likely have developed muscle-memory in Vespucci or EveryDoor that lets you quickly add additional tags while barely looking at the screen.
Also, as I mentioned in my other post, adding POIs and additional tags in some regions’ urban areas is difficult because you don’t want to walk around with your phone out. And the USA makes it challenging to add and update POIs because you often can’t just walk down a street, the sprawl sets everything too far apart for that. The existing mobile editing apps, alas, cannot make up for political failures.
An idea might be to have an "Adopt Your Own Neighbourhood" campaign. I don't know if this is seen as wanted in the OSM community, it might attract a lot of low-effort contributions which might wear out longtime volunteers.
The way I started contributing to OSM is when I was tired of seeing old or no POI data and thought "enough is enough". I first started editing the shopping mall in my neighbourhood, then the schools, and so forth. It is now 5 years later and I can quite keep up locally with POI.
Same, 95% of my edits are POI data in the past two years and I fixed two of the main shipping streets in my neighborhood
But yeah, not sure how scalable even such a campaign would be. I’m the only user I know that use EveryDoor to confirm/update POI in a city of 500k people…
You’d probably need at least 1 out of 1000 people globally to actively contribute. That’s almost 10 million people, quite a bit more than the 1 million Organic Maps users and 250k OSM contributors in 2022 :)
> So I’m starting to question if it’s really worth it to continue update POIs in OSM, but I also would not know how to contribute to Overture as it does have quite some errors :)
It's not a chicken-and-egg situation. Here, the egg is clearly OpenStreetMap data, and the chicken are other maps (such as Overture) that benefit from OSM.
Contribute to OSM for better eggs, or contribute to Overture to make your chicken look better.
In my OSM career, bike-traveling the world, I have added POIs in all kinds of unlikely places in the developing world. But a real challenge is adding more than just the indication that there is, for example, a shop there. It’s hard to add a name= tag when the shop has not hung up a sign with a name. It’s hard to add an opening_hours= tag when the shop has not posted opening hours, and even if I ask the owners, they might say hours are totally fuzzy. Such clear information is largely a feature of the developed world.
In some parts of urban Latin America, the challenge in adding POIs is that it is ill-advised to slowly, aimlessly walk down the street with your phone, because it could get snatched.
> “It’s hard to add a name= tag when the shop has not hung up a sign with a name. It’s hard to add an opening_hours= tag when the shop has not posted opening hours”
Even Google Maps has this problem. There’s a lot of “great noodle shop second on left” type business names in some parts of the world. Owners often live above/behind/in the shop/restaurant and they open more or less on demand. Hours will change seasonally - if there’s customers they’re open and if there isn’t they close.
The real trick with POI data is not just the data itself but the metadata that tells you how important it is. If you know how many people are searching for a place, visiting a place, reviewing it etc then you can boost its prominence in map rendering and search results, greatly improving UX.
An ideal solution would be for all the travel, mapping apps, etc that compete against Google (TripAdvisor, Yelp, Apple Maps, Facebook, all the in-car apps, etc) to share some kind of global OpenPOI database on top of OSM that aggregates reviews, photos, metadata, etc. And also gives owners a single source of truth to update their business data, keep it updated in real time, and have it disseminated everywhere…
Organic Maps is so great. I don't need offline maps as often as I used to but I'm very glad to have this app when I do. And thanks to the folks for rescuing Maps.Me after that product got corrupted.
They absolutely deserve this! App is light weight and works smoothly. Every once in a while I do get into unmarked building numbers, to get around that I search location on google maps (in browser) and get the coordinates and then plug them into organic maps - not a big deal for me.
For day to day use I use Apple Maps, but when hiking Organic Maps is my absolute favorite. A lot more useful than Apple Maps or Google Maps (as it includes routes that are missing in the other two) and it allows to add custom tracks which is super useful for navigation.
Organic Maps is great, especially for planning hikes, but I'd be careful about relying on the OSM topos for wilderness travel. I got burned pretty hard in the Alps by OSM when a trail didn't actually exist. Since then, I like to use Organic Maps for a digital "reconnaissance" of OSM points of interest and then a country-specific topo map like USGS or IGN for backcountry navigation or route finding.
Note: AlpineQuest is a great app for this and is one of the few apps that doesn't charge a subscription to access our taxpayer-funded maps.
I had a similar experience using Organic Maps for hiking. I was hiking on a forested hill near an urban center. Both Google and Apple Maps only had a couple of the biggest trails while organic maps showed an extensive network. This was really helpful for exploring but I ended up having to take a pretty long detour to get back since the trail I had planned to take didn’t exist. I also encountered a trail marked on the map that had obviously been closed for years and was extremely overgrown. I really like the app but I wouldn’t trust the data for backcountry navigation at all.
One of the things I love most about OSM, and has kept me contributing to it for 19 years, is that it isn’t car-centric like the rest of the mapping industry. The world would be better with much less reliance on cars and OSM is a small part of that.
But if you want to have an open source freeway navigator, there’s nothing stopping you: Organic Maps and OSM data are both open and welcome contributions.
If you have an iPhone, Apple Maps is amazing for driving.
When in driving mode, the UI simplifies completely, only showing you the minimum you need to drive. The spoken directions are super clear, telling you things like "skip this light and go into that lane and then take a right at the next intersection". I let Siri guide me just by voice and it works brilliantly, even when navigating spaghetti intersections.
Had the same experience, and ended up being satisfied with Magic Earth[1]. I tried more or less every major navigation service (paid and "free"), and I ended up sticking with it.
And no, I'm not affiliated in any way; just wanted to share a recommendation because I was in the same boat.
On GrapheneOS, I switched from OSMAnd~ to Organic Maps, and have been mostly happy with it.
Organic Maps user interface could still use a little refinement (such as to give an unfamiliar user a sense of what mode they're in and how to get out of it). But it's much more approachable for basic use cases like "From where I am in the city, I want to get to address X."
Readit News
https://github.com/organicmaps/organicmaps/issues/622
https://github.com/organicmaps/organicmaps/issues/1694
... right now your bookmarks aren't synchronised anywhere so if you lose your phone all your bookmarks are gone.
When that's working I'll change over in a second!
It must be really hard to balance giving all those users what they want, building quality code, keeping the project focused, and having low costs. Is it possible to enable this through extensions, or leveraging a product like OpenCloud?
If it regularly exported automatically, it could get picked up by a syncing app, but otherwise...
The second link is asking for iCloud synchronization.
A project that emphasizes privacy and open platforms might want to just deprioritize requests coming from people on privacy-violating, closed platforms, since those people hurt their own credibility in the approaches they suggest.
It’s only updated if there’s someone like me in the local area, or if it’s a tourist destination. But even then it can be sparse (eg just visited Merida, Bacalar and Valladolid in Mexico), and it’s worse in “non-western” countries
Tellingly, Overture Maps that use OSM for the street data to not use the POI data.
A lot of people pointed out that the POI data of Overture had mediocre quality. While that is true, they do at least have coverage in places OSM do not. If you know the name of the place you can at least find it
I’m not sure how this can become better, I’d think that a prerequisite would be that corporate users like Grab, DiDi, Uber start contributing
But that may not happen as they are not allowed to combine datasets [1]. Hence you have a chicken and egg problem: The dataset is not usable until it is complete…
The street data fundementally have the same problem, but that is easier to edit remotely and may be more stable in contrast to POIs
So I’m starting to question if it’s really worth it to continue update POIs in OSM, but I also would not know how to contribute to Overture as it does have quite some errors :)
[1] https://github.com/OvertureMaps/data/discussions/102
edit: Or maybe it is possible to conflate them, but that is not planned - https://github.com/OvertureMaps/data/issues/96
It lets you add details to objects that already exist in OSM. It looks very polished, has nice map, icons, and this default style of sticking to your actual location like some location based games for crazy people (I'm a former Ingress player, Enlightened FTW).
My neighborhood has all details about building size, roofs, street lights, you know if there are markings on the ground on street crossings (for people with poor vision), each path has clear info of what's it's made of, and stores has open/close hours provided. Super cool, easy to use and has real impact.
[0] https://streetcomplete.app/ [1] https://f-droid.org/packages/de.westnordost.streetcomplete/ [2] https://play.google.com/store/search?q=street%20complete&c=a...
Also, as I mentioned in my other post, adding POIs and additional tags in some regions’ urban areas is difficult because you don’t want to walk around with your phone out. And the USA makes it challenging to add and update POIs because you often can’t just walk down a street, the sprawl sets everything too far apart for that. The existing mobile editing apps, alas, cannot make up for political failures.
But shops can be added (using shop overlay).
(I am one of people developing StreetComplete)
The way I started contributing to OSM is when I was tired of seeing old or no POI data and thought "enough is enough". I first started editing the shopping mall in my neighbourhood, then the schools, and so forth. It is now 5 years later and I can quite keep up locally with POI.
But yeah, not sure how scalable even such a campaign would be. I’m the only user I know that use EveryDoor to confirm/update POI in a city of 500k people…
You’d probably need at least 1 out of 1000 people globally to actively contribute. That’s almost 10 million people, quite a bit more than the 1 million Organic Maps users and 250k OSM contributors in 2022 :)
if anyone here is interested and unsure how to start, feel free to ask at https://community.openstreetmap.org/ (official forum, requires OSM account) or at one of other contact channels ( https://wiki.openstreetmap.org/wiki/Contact_channels ) - there is Slack, IRC, Discord and so on.
https://wiki.openstreetmap.org/ may help but it is sadly suboptimal for someone completely new.
If you just click edit you will get tutorial for default in-browser editor.
It's not a chicken-and-egg situation. Here, the egg is clearly OpenStreetMap data, and the chicken are other maps (such as Overture) that benefit from OSM.
Contribute to OSM for better eggs, or contribute to Overture to make your chicken look better.
In some parts of urban Latin America, the challenge in adding POIs is that it is ill-advised to slowly, aimlessly walk down the street with your phone, because it could get snatched.
Even Google Maps has this problem. There’s a lot of “great noodle shop second on left” type business names in some parts of the world. Owners often live above/behind/in the shop/restaurant and they open more or less on demand. Hours will change seasonally - if there’s customers they’re open and if there isn’t they close.
The real trick with POI data is not just the data itself but the metadata that tells you how important it is. If you know how many people are searching for a place, visiting a place, reviewing it etc then you can boost its prominence in map rendering and search results, greatly improving UX.
An ideal solution would be for all the travel, mapping apps, etc that compete against Google (TripAdvisor, Yelp, Apple Maps, Facebook, all the in-car apps, etc) to share some kind of global OpenPOI database on top of OSM that aggregates reviews, photos, metadata, etc. And also gives owners a single source of truth to update their business data, keep it updated in real time, and have it disseminated everywhere…
Note: AlpineQuest is a great app for this and is one of the few apps that doesn't charge a subscription to access our taxpayer-funded maps.
While I am thankful for their time, it's as if all the devs don't drive. Or don't drive on freeways.
It never announces an exit soon enough to take action, at 130km/hr, when you also have to move over 5 lanes.
I've had it announce an exit when I'm beside it, making it impossible to take any action.
Well, anyhow, hopefully rough spots like this get ironed out, we need a Google competitor.
It's sad, because this one thing keeps so many on google maps.
But if you want to have an open source freeway navigator, there’s nothing stopping you: Organic Maps and OSM data are both open and welcome contributions.
https://github.com/organicmaps/organicmaps/issues
When in driving mode, the UI simplifies completely, only showing you the minimum you need to drive. The spoken directions are super clear, telling you things like "skip this light and go into that lane and then take a right at the next intersection". I let Siri guide me just by voice and it works brilliantly, even when navigating spaghetti intersections.
And no, I'm not affiliated in any way; just wanted to share a recommendation because I was in the same boat.
[1]https://magicearth.com
Organic Maps user interface could still use a little refinement (such as to give an unfamiliar user a sense of what mode they're in and how to get out of it). But it's much more approachable for basic use cases like "From where I am in the city, I want to get to address X."