Interesting project! Is it an open source alternative to streetview, mapillary or KartaView/OpenStreetCam? I guess an "about" page would be nice to have :) (edit: on panoramax.fr one can read more about it it seems)
The content is under a Creative Commons CC-BY-SA 4.0.
And it is stated that it is compatible with ODbl but the wiki here:
It is a good analogy, also in the sense that some areas are not reachable without an e-bike and that you'll need to be prepared differently as you have to plan with charging and bigger weight etc.
> .. the issue of door handles. On Teslas, they retract into the doors while the cars are being driven. The system depends on battery power.
I never will understand this horrible decision. It isn't good design if it kills people. I wonder why this isn't regulated. They could at least implement a "push to pop up" functionality that works without battery power or have a narrow slot under the handle.
Thanks a lot for posting this. I highly recommend having a look into the mentioned flexagons. This is a child toy where Feynman laid the mathematical background and it is very fascinating toy which you can easily build yourself. Try it out - it is really fun. No child required except yourself :)
I didn't realize it until I searched for it, but as a kid I did have one of those. An M.C. Escher one (a quick search found exactly the one [1]). My aunt got it for me when I was little and I was fascinated by it. Fond memories.
I regret buying FP4 too. Unfortunately the hardware is very sturdy and does not break justifying buying a different one. But the software feels half ready in a few critical parts (GPS and phone/sms in my case) and the support is non existent (very bad) for my two issues I had (still have).
We give a deterministic O(m log2/3 n)-time algorithm for single-source shortest paths (SSSP) on directed graphs with real non-negative edge weights in the comparison-addition model.
This is the first result to break the O(m + n log n) time bound of Dijkstra’s algorithm on sparse graphs, showing that Dijkstra’s algorithm is not optimal for SSSP.
OSMAnd and OrganicMaps both have the limitation (and big advantage) of functioning offline by default. The routing will be much more powerful (with alternatives on by default) and faster if you enable an online routing service. For OSMAnd this is possible with e.g. GraphHopper: https://www.graphhopper.com/blog/2024/02/27/osmand-with-grap...
The same is true for address search. If you have an online address search like photon the search can be more user friendly. We've put together photon and GraphHopper routing on GraphHopper Maps: https://graphhopper.com/maps/ which you could self-host on your own (i.e. also use offline): https://github.com/karussell/local-maps
GraphHopper Maps is also available on fdroid store or you can install the website as PWA in iOS.
> (and big advantage) of functioning offline by default.
I don't know about others but that's the main reason I use it. My day to day mapping app is still Google Maps but I always keep a copy of Organic Maps with downloaded maps of wherever I'm going as a backup. While I do not use it often, it's gotten me out of a couple of sticky situations while camping and roadtripping.
Organic Maps (and other offline mapping providers) are far from perfect and the UX is just not the same as it is on Google Maps for example. But with it being a backup app, if I need to open it I don't really care about the limitations, I just need an offline map.
I know this probably doesn't solve the same issue, but google maps has a offline feature. Click your profile picture (on mobile), pick offline maps, ....
It has a huge limitation in that it only allows you to pick a certain area to download and those maps "expire" after a period of time. The key advantage that Organic Maps (and other OSM providers) has is that I can download an entire state, province or even country and that data will never "expire".
I think this is less Organic vs Google than OpenStreetMap's data set vs Google's. I don't know why Google does so much worse with trails than OSM, but it really does.
> I don't know why Google does so much worse with trails than OSM, but it really does.
I expect that Google never saw a market in trail mapping. I also assume no Google employee took an interest in trails as a 10% project. Google Maps doesn't really do much for topography either.
Google Earth can be good for trail mapping, but that has basically atrophied since it was acquired from Keyhole.
Yeah, but the thing is that I don't think any of this data is "self-collected". I suspect that OSM gets most of its US trail data (or at least western state trail data) from the National Forest Service and Bureau of Land Management. I also suspect that incorporating such a data source into Google Maps is relatively trivial, but they just seemed to have done so.
I map many trails on osm from personal site surveys and a combination of sat imagery and my gpx files. No way google is doing that because there is no one to steal the data from. It was me, the enthusiast that put it in osm directly. That’s just me and the trails I load tho. Example - Latest was short one at monkeyface falls. https://www.openstreetmap.org/#map=19/34.098058/-116.955639
I hike on trails in New Mexico, and I find the existing trail data on osmand/OSM to be astoundingly accurate. I had concluded that NSF/BLM must have data.
So few people hike these trails that I do not believe they were entered one by one. The one "trail" I hiked that was entered by someone I deleted later that day, because it should not have been shown as a trail.
I happened to work for a car navigation software development company 15+y ago. Cool stuff, Windows CE / PDAs as devices, android and ios nowhere. These were totally offline devices (map updates through usb / sdcard).
Even then, this offline navigation was super fast, across countries. Today I managed to wait a whole minute for a 5km bike navigation in OsmAnd. Then I uninstalled (after years of hoping for improvement. Yes, I was regularly donating money.)
In my experience, OsmAnd is mostly slow for very long routes.
Maybe it is a matter of quality. Because of course you can find routes fast if they are not the fastest or best routes.
But there is room for improvement.
brouter could be integrated even better. Or a router like that could be used directly in OsmAnd.
And long routes could be handled more flexible.
E.g., when I go from Copenhagen to Barcelona, it is not super important at first to find the optimal way into Barcelona, or shortcuts in France using regional roads. It will take several days, but I would like to start with a reasonable route giving me an estimate of distance and time. At first I just need a good route to the Great Belt Bridge or the Rødby ferry -- Copenhagen is on an island.
When I drive long distances, I sometimes use several devices.
The Xzent system is much faster for longer distances, but the map is not as good, especially it is missing may POI's.
Often they disagree, especially if one is optimizing for distance and other for fuel or time. Then if there is an obstacle or a bad road, I instantly have a good alternative at an intersection.
Just that comparing Google and OSMAnd/OrganicMaps in terms of routing alternatives & speed and powerful address search is not 100% fair (even when they'd use the same data source which they don't)
> The routing will be much more powerful (with alternatives on by default) and faster if you enable an online routing service.
What is the essential reason that online routing has an advantage over local routing, if the data is all available locally anyway? Is it that you need an index, and that index is large and/or very time-consuming to produce, and hence not viable to store/generate on-device after each map data update?
At least for bycicle routing, Brouter also runs offline and is much more performant than both OSMAnd and OrganicMaps (and can be integrated into OSMAnd).
To me it feels like OSMAnd heavily prioritizes feature develompent over performance, which is fair enough but still annoying.
And their choices in features they wish to develop confuses. 50% of the time I still have to translate an address into GPS coordinates before I can find a place, and yet I have an OBD-II plugin that allows me to monitor my car's performance?
Offline navigation is really nice. The fact that I can use maps and find routes regardless of where I am and what connection I have, is great.
It would be nice to have slightly smarter search, though. That definitely requires improvement. Even just the ordering of the results is terrible sometimes.
