While I like the concept of the open maps idea and I'm sure it has a bright future, what about aerial views, street view, etc? Is this not a functionality step backwards for FourSquare hopping on such a new service at this point?
The biggest problem I see with MapBox right now is that their map labels don't seem to be dynamic. There are a lot of places around my house I can zoom in and I have to pan to see the street name because the label is out of view. There is one location nearby where I can scroll 3/4 the way across town and not see a label on one particularly major street. This could quickly make for a frustrating user experience.
Don't get me wrong, I applaud what MapBox is putting together and I'm sure it'll improve rapidly. At this point though it does seem like a significant drop in functionality for a major service to move onto it.
In some ways it's better, in some ways it's worse. In the long-term though, we feel that this is the right move. OpenStreetMap and MapBox will be improving continuously, and with open data we can drive those improvements in ways that we couldn't with Google data.
I've been following OpenStreetMap since it launched, and its really a testament to how far its come in the last 7 years that it's good enough to be used in major consumer websites. In many ways, this is an effort that dwarfs Wikipedia in size, complexity, and ambition.
We're excited to be making use of open data, and will be even more excited to start contributing back in the near future.
> The biggest problem I see with MapBox right now is that their map labels don't seem to be dynamic.
As far as I know, there's only one map where labels _are_ dynamic - Google MapsGL, and it's available only on 10% of browsers.
> While I like the concept of the open maps idea and I'm sure it has a bright future, what about aerial views, street view, etc? Is this not a functionality step backwards for FourSquare hopping on such a new service at this point?
FourSquare wasn't using street view, but the satellite imagery void is definitely something to address.
Based on the system requirements MapsGL should be available much more than 10% of browser. WCIU... pegs "partial" WebGL support at 46%. Evan if you remove a generous % for lack of hardware support you are still significantly over 10%.
What about contributing the Venues to an open places database? AFAIK Foursquare is one of the only companies in the Factual Crosswalk API that is included but not an active cooperative partner.
I noticed that all comments on HN and on the blog post that addressed the issue of venue data were conveniently dodged.
This is unfortunate and reflects badly on FourSquare since it is highly indicative that this decision is almost certainly about the money being paid to Google for its API and not at all about supporting open-source.
It appears that FourSquare is happy taking what it wants from the commons generated by the community, but when it has an asset that would greatly benefit the commons, and that they don't necessarily need to hold close to their chest to maintain an advantage, that they don't donate it.
To me, that's a shame.
The whole world would clearly be better off if one of the Place API providers with a solid collection of POI internationally (Google and Foursquare) were to donate that set to the commons.
It's great that the commons gains another major supporter, but near as I can tell Foursquare is just using information in the commons to its advantage here without truly participating in the commons.
[Disclaimer: These are my personal views and do not represent those of my employer, foursquare.]
Really though, the new Google Maps pricing scheme is just totally bonkers. They're trying to charge a higher CPM than typical AdSense rates. It's just nonsense.
The biggest problem I see with MapBox right now is that their map labels don't seem to be dynamic. There are a lot of places around my house I can zoom in and I have to pan to see the street name because the label is out of view. There is one location nearby where I can scroll 3/4 the way across town and not see a label on one particularly major street. This could quickly make for a frustrating user experience.
Don't get me wrong, I applaud what MapBox is putting together and I'm sure it'll improve rapidly. At this point though it does seem like a significant drop in functionality for a major service to move onto it.