Apple: To map or not to map – That was no question…(ipadwatcher.com)
ipadwatcher.com
Apple: To map or not to map – That was no question…
http://ipadwatcher.com/2012/09/29/apple-to-map-or-not-to-map-that-was-no-question/
2 comments
How does real user usage improve the maps, exactly?
It helps with sorting algorithms. Whenever you do a search the popularity of the potential results can be used to put the most relevant first.
It helps with map updates - people are mostly going to search for things that exist. You can use that to look for anomalies. eg if they keep searching for "statue" in some town and you don't have results, then chances are the map/data is missing a statue.
It can provide feedback for routing. If the users do not follow the directions you give then chances are you have a map problem. Maybe you are trying to send them the wrong way down a one way street, maybe layouts have changed etc. If they do follow the routing then you can use it to gauge/improve the accuracy of your estimates.
Those are some examples where usage gives feedback in an automated/statistical fashion. The app also has a feedback (or in iOS case "Report a problem") mechanism so users can explicitly tell you of the problems.
It helps with map updates - people are mostly going to search for things that exist. You can use that to look for anomalies. eg if they keep searching for "statue" in some town and you don't have results, then chances are the map/data is missing a statue.
It can provide feedback for routing. If the users do not follow the directions you give then chances are you have a map problem. Maybe you are trying to send them the wrong way down a one way street, maybe layouts have changed etc. If they do follow the routing then you can use it to gauge/improve the accuracy of your estimates.
Those are some examples where usage gives feedback in an automated/statistical fashion. The app also has a feedback (or in iOS case "Report a problem") mechanism so users can explicitly tell you of the problems.