← Back to context

Comment by threeseed

4 years ago

Please provide evidence of this because Apple's official documentation says otherwise:

https://support.apple.com/en-us/HT203033

They do send nearly WiFi hotspots for crowd sourcing purposes but it is never in conjunction with your local IP address (which is an identifying piece of information).

Did you read the OP article? The researchers clearly outline what apple is phoning home. They even made a nice clean table showing what apple and google are sending back to themselves.

  • I read the article and it's wrong.

    Apple does not explicitly "send" the user's IP address. It naturally is accessible on their end as a result of the TCP/IP protocol. But Apple has made quite clear that it does not use that information in any way.

    • The linked PDF (direct link for the truly lazy [0]) shows that every few minutes Apple sends binary data to its servers which includes the MAC address of nearby devices. (It's unclear if these are only devices on the same network, e.g. from arp, or any nearby devices that are broadcasting a static MAC address). Here, I'll even quote it for you. It's on page 7 at the bottom left, continuing on the top right:

      > However, the geod process uploads binary messages to gsp85-ssl.ls.apple.com... While it is not clear what information is contained in this binary message, it can be seen to contain the MAC addresses of nearby devices sharing the same WiFi network as the handset e.g. f2:18:98:92:17:5 is the WiFi MAC address of a nearby laptop, 70:4d:7b:95:14:c0 the MAC address of the WiFi access point.

      Idk what they do with this info, and I'd much rather Apple have it than Google, but you can imagine the "God mode" they could create at Apple HQ if they were so inclined. The data is absurd... imagine what you could do if you knew where billions of people were at every second of every day for years.

      [0] https://www.scss.tcd.ie/doug.leith/apple_google.pdf

      4 replies →

    • Local IP would be your 192.168.1.whatever. Apple won't get that unless they capture it on the device. They would otherwise only get your public, nat'd IP with normal tcp/ip.

      Local IP isn't identifying, but it's a weird thing to include. And the paper clearly shows that being sent to Apple.

      > Later during the startup process the local IP address of the handset (i.e. not of the gateway, but of the handset itself) is sent in a POST request to /lcdn-locator.apple.com: POST https://lcdn-locator.apple.com/lcdn/locate Headers User-Agent: AssetCacheLocatorService/111 CFNetwork /1128.0.1 Darwin/19.6.0 POST body {"locator-tag":"#eefc633e","local-addresses":[" 192.168.2.6"],"ranked-results":true,"locator-software":[{" build":"17G80","type":"system","name":"iPhone OS","version ":"13.6.1"},{"id":"com.apple.AssetCacheLocatorService"," executable":"AssetCacheLocatorService",<...>

      So no the article isn't wrong. I suggest you give the paper a read (or at least a skim) if you're going to try and claim they are wrong about something.

Nearby WiFi spots can be easily used to track device location. Here is a neat little project that does exactly that for IOT tracking [1].

[1] https://github.com/dmsl/anyplace

  • Nearby wifi hotspot location is the base for aGPS. Without it it would be slower and eat up more battery.

    • No... aGPS doesn't use wifi at all. aGPS uses specific services built-in to the cellular network to accelerate your initial fix.

      You're right aGPS does speed up your initial time to fix so the GPS receiver doesn't have to be on as long to get an ephemeral location but other than that it doesn't have any impact on battery life such as when you're looking at Google maps with a constant fix.

      1 reply →

The wifi thing is fine/good IMO. It allows everyone to get their location without gps. It’s what let’s devices with no gps like the MacBook and ipad to get their current location. Google does the exact same thing although they used street view cars for the initial dataset.