Simple answer: Google Maps suddenly costs a lot of dough.
More complex answer: Companies that need mapping services don't want to rely on Google if they are a competitor.
Say you're Foursquare. Would you rather rely on Google, a company that's investing heavily in social networking, or someone like Navteq or OpenStreetMap, organizations with a laser sharp focus on mapping services?
Foursquare went with OpenStreetMap though, and claim "while the new Google Maps API pricing was the reason we initially started looking into other solutions, we ultimately ended up switching because, after all our research and testing, OpenStreetMap and MapBox was simply the best fit for us."
They could be lying to cover a penny-pinching move, but I think they're telling the truth simply because Foursquare doesn't need particularly good geodata. It only needs for locations you visit to be fairly recognizable and locatable, it doesn't need to know that you can't do a u-turn or go that way up a one-way street or the exact location of a point to less than 1m accuracy. The only businesses it cares about it wants in its own dataset, not the map providers. Same story for Apple and its photo display or (going way back now) Yahoo/Flickr using OSM for geolocating photos in Iraq and China, Afghanistan etc. despite Yahoo having their own map team:
This process has been going on for a while now, it's the traditional story of the open alternative eating the market from the bottom-upwards and accelerating as it gets wider use and improves in quality in a virtuous circle.
Quite right. I knew Foursquare went with OSM, I was just putting a hypothetical out there as a thought exercise. But I did change my post to be a bit more accurate.
Also, you make a great point about the lower level of fidelity Foursquare, iPhoto and Flickr need. Higher map accuracy wouldn't be worth the cost in those use cases.
It seems Chris Anderson the author of Free was right. Free is a fundamental principle of the Internet because of its super efficient way to distribute information. If you try to mess with that too much, even something as huge as Google Maps could start losing market share if there's a good enough free/cheaper alternative instead.
Anyone who is considering making the switch should take a look at http://switch2osm.org
Also, if you want to use Google (or any other provider) but not get locked in I can recommend writing your code using Mapstraction - as the name tries to suggest it's a mapping provider abstraction layer http://mapstraction.com/
Is this really an issue for you? Assuming it is, simplest solution I see (besides the obvious appealing to good will of others to make the updates you need) is you pay someone to do the update you need and then submit it to the project. That person could be in a lower cost location than you are, thus much cheaper than you doing it yourself. A much better problem to have than vendor lock in.
The design you can achieve on top of OSM is beautiful. Map design can and should be part of your service's brand image, if it forms a big part of your service. Not everyone wants to look like Google Maps. The new 4sq maps go nicely together with their branding now.
And although Google now offers map styling, it is still pretty limited and brings your free quota down to 2,5k requests if I remember correctly.
This is true, but other parts of the map data may well be of higher quality too.
I find OSM data in and around London to be more complete than Google Maps, or rather the speed of updates is faster making it more accurate and I also find data that is more granular (such as where in a fence a gate may be).
Also, OSM maps for specific purposes work really well. Nearly all of the cyclists I know who have Garmin units have replaced the maps with OSM ones just to get contours and other cartographic features that assist riding in the countryside (where fields or woods dominate most maps and more detail is needed).
How much would Google have to drop the price to reverse the trend?
The plain old JS API would come to about $1000 a month for a site with 1 million map views. Mapbox (http://mapbox.com/plans/) meters by bandwidth but by my estimate (4gb map, low for a global map I know, and 1,000,000 views) it's still about $600.
Then again tilemill, tilestache, and leaflet are all free and work great once you get it going.
Google Maps does have a quality problem. Within the UK the search results are appalling. More often than not you get an odd town in the USA, or something completely irrelevant to your actual query.
The quality of the data simply isn't good enough to give good results. Being Google they try and fix this with a clever algorithm. When people are searching a map they are usually after one correct answer, rather than hundreds of possibly right answers.
Google Maps is a generic product that is highly funded and good in the USA. Everywhere else it is seriously lacking.
Note: Not sure if the app or the service is to blame. But - disappointing:
From this afternoon. Wife and me are sitting in the sun at the Kikar Rabin/Rabin Square in (more or less) center Tel Aviv. She decides she'd like to have sushi. I pull up the Google Maps app, type sushi <enter>
Map zooms to the US and gives me recommendations in god-knows-where. I laugh it off, go back, search 'sushi bars'. Guess what it does? Riiight.
So this company _constantly_ throws Hebrew at me. After updating maps it showed the TOS in HEBREW (I .. cannot read that) on first launch. It certainly knows (and abuses. Certainly abuses. Abuses all the time..) geolocation. I have Latitude enabled the last weeks. GPS and Wifi were turned on and the map showed my current location quite accurate.
Why in the world would I want to know about Sushi places in the US?
That's certainly fixable (I could get somewhat better results with 'Sushi, Tel Aviv') but I think the gp hit similar idiotic behavior.
Unfortunately that just shows flaws in the Maps search algorithm and is unrelated to Google Maps vs. OSM for example.
Mostly my problem is that I don't use the UI properly. I hit return and don't select an option when they pop down. However, it does tend to come up with strange results:
Worbarrow Tout (has wikipedia page). If you are zoomed on the UK you get a selection of B&B's in the right general area.
Oxwich Burrows gives me a caravan park and a hall.
Both of these results have only one unambiguous meaning. Google gives me several results that are not correct.
Other than more street view locations on Google Maps, Bing Maps is far, far superior (mostly in terms of speed, but after using Bing it's painful to go back to Google).
Pretty poor analysis to mention Apple switching, but not mention the fact that Android is the primary competitor to Apple's biggest platform, and that maps are one area where Android has been consistently ahead of iOS.
Apple could afford to pay a lot for its maps, but with Google there's a direct conflict of interest, and Google will always drag their feet in helping Apple create a great maps experience.
Why? Because Google maps is no longer in a position of inertia. Now if you require mapping functionality you'll perform some kind of due diligence based on requirements and constraints.
I don't think you can get even simple street directions from OSM. There probably are or will be third-party services that enable doing this using OSM data. Encourage your transit agency to make its GTFS feeds public if they aren't already.
What it actually says is that whilst the cost is an issue, the primary reason is that these companies don't want Google ads seeping into their products.
Data's only half the equation, building up a service with all of the features to what you need costs $. From the Switch to OSM site "Serving your own maps is a fairly intensive task. Depending on the size of the area you’re interested in serving and the traffic you expect the system requirements will vary. In general, requirements will range from 10-20GB of storage, 4GB of memory, and a modern dual-core processor for a city-sized region to 300GB+ of fast storage, 24GB of memory, and a quad-core processor for the entire planet."
Don't get me wrong OSM is a fantastic initiative. It will be very interesting to see if a slew of competitive services spring up in the map 'service' arena, using OSM data.
For a while, Google was far from transparent about potential costs and some interpretations were indicating that the cost would be enough to destroy certain services.
Whilst Google eventually opened up, despite the prices being reasonable, most affected services had already finished (or nearly finished) their conversion to OpenMaps.
Given the cost of finishing the conversion or sticking with GMaps for a reasonable fee, many took the decision to convert.
Google kinda dropped the ball here and got a lot more damage than they needed to.
If you were a heavy user, it's likely the costs were pretty clear early on, even if they weren't public.
We got costings sent to us in the middle of 2011. However we chose to reduce our map impressions as the prices quoted were nowhere near what we could afford.
More complex answer: Companies that need mapping services don't want to rely on Google if they are a competitor.
Say you're Foursquare. Would you rather rely on Google, a company that's investing heavily in social networking, or someone like Navteq or OpenStreetMap, organizations with a laser sharp focus on mapping services?