Shortly after I published my Cartography Comparison last June, I noticed Google updating some of the areas we had focused on.
Coincidence or not, it was interesting. And it made me wonder what else would change, if we kept watching. Would Google keep adding detail? And would Apple, like Google, also start making changes?
So I wrote a script that takes monthly screenshots of Google and Apple Maps. And thirteen months later, we now have a year’s worth of images.
This is a fascinating article. Google is changing the look of the actual maps in Google Maps a lot, and improving its data all the time – whereas Apple seems to lag behind, and contains far less places of interest, stores, and so on.
Whenever I search for something on Apple Maps it gives me results from another country (I live close to the border). The few times I tried to use it I had to switch to Google Maps anyway. What a waste of time.
Apple maps is also a non-starter if you don’t have an iDevice at hand, severely limiting its usefulness.
Although not has featured rich compared to Google Maps, OpenStreet Maps is more accessible to the user. Reproducing the scenario showed how up to date it is in term of location.
Yes, running a “Cartography comparison” article with GM and AM but without OSM looks really strange to me. Different worldviews, I guess.
Speaking of update velocity, OSM currently has about 300 daily active contributors for the USA (about 4500 worldwide), some doing just a single change, some doing thousands, all available immediately. The main OSM style (out of many many different styles) gets a big update every 2-3 months.
While contribution statistics for GM and other providers are hard/impossible to come by (I’ve tried), it seems pretty safe to say that OSM is unmatched.
https://mc.bbbike.org/mc/?lon=-122.424661&lat=37.776233&zoom=18&num=…
http://osmstats.neis-one.org/?item=countries&country=United%20S…
https://github.com/gravitystorm/openstreetmap-carto/releases
Unless I’m misreading what you wrote, there’s no way OSM’s 300 daily contributors are unmatched compared to GM. I see those Google mapping cars everywhere, all the time, and that’s just here (Seattle).
Seattle is a very small part of the world. Around here (Soth-Eastern Europe) their cars are rarely seen (I can’t say I saw any in the last couple of years). But to be fair, there are not many OSM contributors around…
That’s the point. If there’s plenty of Google mapping cars here, think of what the number is when you include all metropolis cities in the US alone. You will blow past the `unmatched OSM 300` no problem.
are the cars doing something else than street view imagery?
They are probably doing both imagery and LiDAR to contribute data to the automated driving project.
In most parts of the world, StreetView is years out of date. And while getting a fresh StreetView image is great, it doesn’t yet transform into map data without human intervention (that’s not entirely true, but for most of your daily map usage it is).
You’d be surprised in how much directly useful data a measly 300 daily contributors (who aren’t necessarily focused on big cities) can map. I’d take 300 human mappers over 300 streetview cars (again, it seems impossible to get stats on the streetview fleet) any day.
OSM’s contribution velocity is unmatched when you look at the worldwide stats. Given that the USA is OSM’s comparatively weakest country and GM’s comparatively strongest one, you really need to attach a YMMV disclaimer when using the USA as a starting point for comparison. This is even more true if your reference is Seatle.
Yet look at the comparison website from my previous comment: OSM typically has more details than GM in the discussed area. The only thing that GM has going for it is the 3d-styles buildings. OSM may only have 300 daily contributors in the USA, but they are prolific. The tools make a big difference. You’ll get a lot more done in an hour of OSM mapping than an hour of GM mapping. The OSM tools (QA, editors, query…) are available to the whole community, whereas most of GM’s tools are only available to Google employees. A sizable portion of OSM contributions come from fulltime professional mappers, so it’s not a case of “hobbyist vs professionals”.
Taking a fresh picture of an area and translating that to map data are two very different things (I should know, given the number of survey photos I’ve yet to act upon). StreetView is absolutely great, but isn’t a replacement for good map data (to some extent, GM’s detail-poor map *needs* to be complemented by StreetView while OSM’s detail-rich map doesn’t).
Again YMMV, but living in Europe I’m more likely to find the POI I searched in OSM than in GM. And I’ll get the answer faster, without using any mobile data. If you’re spending any time contributing map data, you really should spend it with OSM, not GM.
Does anyone know if Apple is still relying on OSM data, or did they move everything in-house?
Apple maps is a classic “me to” business decision. Management sees something and says we can do that too, then sinks millions into it, then wonders why they can’t compete with the establishment. Apple simply doesn’t have the resources to compete here, same with Bing. Management can only see facts on spreadsheets, and not why things are and how they would do it better. They can’t beat Google’s products just by sinking more money into their own.
Edited 2017-05-29 18:46 UTC
Apple has or can buy the resources to compete anywhere it wishes. Any lack of competition with Google Maps is a decision Apple made, for good or ill. You can blame them for a poor decision, but not lack of resources ($250 billion + in cash on the balance sheet).
1) Both Apple & Google Maps are getting uglier and it is getting harder to see anything when looking at the maps. The more they try to show you the less I can find without verbally or typing searches into them.
2) Apple’s big problem with their maps is TomTom. As noted in the review of both programs, Apple maps looks almost identical to TomTom which isn’t a good thing.
3) Apple “could” use the technology they bought from a Swedish company that used data they captured by flying over cities. Couldn’t that be used to create better maps?
Note: Apple doesn’t (at least currently) do Street View. So they really don’t need to drive around on all the streets to get the streets correctly down to inches. The only issue would be with overhanging objects such as trees that block the view of certain streets. But with everything else, all you need is a couple of passes from different angles with GPS and they could create their own maps from scratch. Look for the demos the Swedish company had before Apple bought them. It wasn’t all about just the buildings but the roads too.
I’m probably the only odd one out, but I’ve had better luck around here with Apple maps, particularly when getting directions. I was rather surprised to find that out. It really depends on where you are and, no, I am not in the bay area nor anywhere close to it
.