Re: [100attendees] Incorrect GeoIP for IETF network addresses

[Michal Krsek <[e-mail]>]

Hi Alex,


> For BSSID: the MAC-to-coordinates mappings dont deserve much 
> consideration IMHO, as it's not IP.

Sorry for my shortcut:

BSSID is not on IP level, on the other hand, APs are being (most likely) 
connected to the IP network. And the IP network operator may signal to 
the devices that this is a "mobile" network. Not proposing any 
particular protocol here for now.

> Wireshark setting 'GeoIP' fields is something non-standardised that 
> should not be trusted.  IT may say "Alaska" but it should not be taken 
> seriously.

There are other non-standard ways, how to handle it (like call to your 
friend that works for MaxMind).

Michal

>
> Alex
>
>>                 Michal
>>
>>
>> On 13/11/2017 10:52, Mike Bishop wrote:
>>>
>>> I’ve also pursued this with another providers which used the 
>>> “sightings” of BSSIDs and cell signals and how they handle updates, 
>>> when I moved a portable cell spot.  The issue turned out to be that 
>>> it requires:
>>>
>>>   * GPS-enabled clients to report the new location in sufficient
>>>     volume to cause the service to reconsider the location
>>>   * The service-side database to be updated
>>>   * Clients to age out their cache of the previous location of the
>>>     BSSID/cell.
>>>
>>> As a result, the expected time for clients to see the new location 
>>> was 10-15 days, by which point we won’t be here any more.
>>>
>>> *From:* 100attendees [mailto:[e-mail]] *On 
>>> Behalf Of *Jim Martin
>>> *Sent:* Monday, November 13, 2017 10:42 AM
>>> *To:* Darshak Thakore <[e-mail]>
>>> *Cc:* [e-mail]
>>> *Subject:* Re: [100attendees] Incorrect GeoIP for IETF network addresses
>>>
>>> Darshak,
>>>
>>> This is a sadly complicated issue. There is no geolocation standard, 
>>> so it’s a matter of trying to update the various organizations that 
>>> provide geolocation services. We work closely with Google, and  try 
>>> to work with SkyHook (the geolocation that Apple uses for BSSID 
>>> based location) and MaxMind (a common location provider). 
>>> Unfortunately, these are a bit hit or miss, as they’re not really 
>>> setup for a non-trivial network to move around the world. We 
>>> continue to work on it, and are always open to help, if anyone has 
>>> suggestions.
>>>
>>> BTW: In terms of google maps, it “works for me” on my Mac using 
>>> Safari right now. Swing by the NOC in Indiana if you get the chance 
>>> so we can try to understand the differences in results. Thanks!
>>>
>>> -Jim
>>>
>>>
>>>
>>>     On Nov 13, 2017, at 10:34 AM, Darshak Thakore
>>>     <[e-mail] <mailto:[e-mail]>> wrote:
>>>
>>>     Hi all,
>>>
>>>     Is there a way to update the geoIP that the IETF IP addresses
>>>     resolve to? I’ve been seeing this behavior for the past few
>>>     IETF’s where on google maps (or on bing maps for that matter) in
>>>     a browser, when I allow geo-location (while connected to the
>>>     IETF* networks) it always resolves to the previous IETF venue.
>>>     So today when I click “Your location” button on google maps, it
>>>     shows that I’m at Hilton Prague and when I was in Prague, it
>>>     showed me Chicago (and so on)
>>>
>>>     Just wanted to know if there’s something I/we can do to make it
>>>     resolve to the right venue
>>>
>>>     Darshak
>>>
>>>     _______________________________________________
>>>     100attendees mailing list
>>>     [e-mail] <mailto:[e-mail]>
>>>     https://www.ietf.org/mailman/listinfo/100attendees
>>>
>>>
>>>
>>> _______________________________________________
>>> 100attendees mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/100attendees
>>
>>
>>
>> _______________________________________________
>> 100attendees mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/100attendees
>
>
>
> _______________________________________________
> 100attendees mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/100attendees