We cut a 1.0 release of the Mozilla Stumbler finally.
Have at it. File the bugs. Complain about battery life.
Help us make this thing not suck and build out a proper open location service.
The privacy policy[1] could be clarified for less technical readers, and even for others. I infer that collected data is anonymous because you write,
1) We receive publicly observable data about WiFi access points and cell towers around you, your estimated latitude and longitude, and the date -- Not associated with anything else, that may be anonymous data -- though you could guess my home network or home location by the most common/strongest wifi signals. If you track data by submitter, you also would have a good idea of their travels.
2) we may receive certain temporary data such as your IP address. This data is deleted after being used as follows ... -- You seem to be implying that you do receive non-anonymous data, and delete it after innocuous uses.
3) You can send us data anonymously or under a nickname -- Which implies anonymity is possible.
If what I infer is correct, why not restate it directly and unequivocally with something like the following:
Unless you choose otherwise, the data you send will be anonymous and not associated with you in any way. We will not record who you are or what phone sent the data. We do receive some non-anonymous data, but we delete it within X hours/days after using it as follows ...
And add more detail after that.
[1] https://location.services.mozilla.com/privacy
EDIT: Clarify a bit, and a correction to #1
The IP addresses are just a fact of life of web server logging. They are not stored in the location or leaderboard databases.
edit after downvote: also, mozilla engineering PMs will intimate on hackernews that it won't internally correlate and potentially sell any of the location and other information it most obviously could correlate about people, even though it has already announced its intention to advertise.
Does Stumbler also support this? If not, why not?
Ref: https://support.google.com/maps/answer/1725632?hl=en
edit: Found the answer to my own question. Yes, they do support "nomap": https://github.com/mozilla/MozStumbler/issues/149
I feel very ashamed, as someone who works in IT, everytime this happens. I mean, people can opt-out, of course - but, in order to do that, they need to know what an SSID is, and how to change it. What about people who don't? Will we just assume that they don't care or that their opinion doesn't matter?
Also, why do you need access to my photos?
Mozilla Stumbler does not need to access your photos; it just needs to read/write to your sdcard (to cache map tile graphics and export KML data). "Access your photots" is Google's unfortunate explanation of accessing the sdcard.
One question though. Will ordinary users be able to directly query the database via the API? i.e. if you want to geolocate a set of Access points and you have their MAC addresses will this service provide a direct supported API for retrieving their location?
The documentation on the API page implies that this isn't really a supported application?
Specifically
"If you are developing a native application or library for a desktop operating system or Android, you can in principle use this service via its HTTPS API. Please refer to the development documentation for the details. On most other operating systems, you cannot access the required cell and WiFi network information required to call the service API. "
and
"At this stage the service is open to anyone, who wants to contribute back to the service and applications supporting the Mozilla mission. "
seem a bit vague.
https://location.services.mozilla.com/v1/geolocate?key=test
I'd love to play with the raw data too, but I understand their concern. So far many "anonymous" large data dumps ended up exposing too much (e.g. recently NY taxi data).
You linked to the download page where there's only the Cell Networks database.
Owners of Wi-Fi access points, including mobile phones with sharing Wi-Fi, may not want their unique BSSID/MAC address and location published. This is not exactly comparable to the Google Wi-Fi case in Germany. In addition to recording BSSID/MAC addresses, Google was (inadvertently?) logging Wi-Fi payload data that included cleartext user data.
btw, here's a zoomable map of the Mozilla Location Service's data coverage. Please help fill in the blanks! :)
Stumbler seems to anonymize the data on your device before sending it, which is the way it should be done. There is no reason to transmit any sort of wifi data except the AP's MAC address.
We caught that bug just before release. :)
I remember reading how Wigle Wardriving calculated fixes and the method seemed unsophisticated and lame.
For example, if I bike down a street then fixes would be detected 100m ahead of me and always pinned to the road at my current location.
If I bike down the road in the other direction the next time, will be fix become more accurate?
Normally higher SNR fixes should have more weight than weak fixes. Do they?
https://github.com/mozilla/ichnaea
We store signal strength for possible future use, but our current location algorithms don't use it. Wireless signal strength is notoriously flaky. From some reports I've read, signal strength is more highly correlated with the user's orientation (i.e. is their body blocking the signal to the source) than with distance to the source.
Kudos for giving back what people gave you.
The first check is "if we don't get a SSID, we don't collect". That's what happens with hidden networks.