Presence and IntComm WG

[Jiri Kuthan <>]

Actually some while ago we did put a feature in serweb which was
displaying geo-location of users in the webportal based on SIP
contacts + NetGeo. Whereas it was a really nice feature, the
NetGeo information was not very accurate. I remember it was
showing my Berlin-Germany IP as located in Amsterdam-the Netherlands.

-jiri

At 11:08 PM 9/27/2005, Ben Teitelbaum wrote:
>Candace,
>
>Thanks for the pointers.  As far as I know, these sort of heuristic-
>based location schemes began with CAIDA's NetGeo, which I played with
>years ago while the project was still active.  Now, there are all
>sorts of dot-coms doing it (as [1] and [2] mention).  NetGeo used
>latency only as a sanity check (e.g. does ping take at least as long as
>direct path / (0.66 * speed-of-light)).  Mostly it relied on whois
>records and understanding various ISP's conventions about naming
>routers (many embed airport codes).  To get a sense of the heuristics
>that NetGeo used (which are probably similar to what the commercial
>services are doing today), see:
>
>http://www.caida.org/outreach/papers/2000/inet_netgeo/inet_netgeo.html
>
>Best,
>
>-- ben
>
>Candace Holman 
><>
> writes:
>
>> Hi all,
>>
>> I'd like to tell you about two news articles regarding location
>> services.   There was a great suggestion during the member meeting
>> conference call,  to try to expand user agents to provide location
>> services for rich presence.  Perhaps we can start with geolocation? Or
>> maybe eliminate it as an option?  Here are some ideas:
>>
>> The first article is about a new patent that was issued to the NSA for
>> location tracking [1].  It appears that the location tracking patent
>> is based on creating a "network latency topology map" based on IP
>> addresses that are already located and known to be static.  Then an
>> unknown computer location can be estimated based on the latency of
>> connection attempts with it.  There are fears that this could be used
>> for spying, but it seems to me that it would be pretty simple to
>> confuse any program that is attempting to locate you based on round
>> trip time variations to your IP address.  Not to mention that these
>> latencies would have to be constantly monitored, and the topology map
>> would have to be updated in real-time.  But the basic idea is
>> interesting.
>>
>> A related story [2] has some great Use Cases, social implications, and
>> real-life examples of geolocation but also lacks a good technical
>> explanation.  I found some details on IP address geolocation online in
>> the Linux Journal [3], and came across the July '05 internet draft on
>> SIP location conveyance [4] which describes the use of  PIDF-LO and
>> geopriv.
>>
>> [1] NSA granted Net location-tracking patent
>> http://news.com.com/NSA+granted+Net+location-tracking+patent/2100-7348_3-5875953.html
>>
>> [2] Geolocation: Don't Fence Web In
>> http://www.wired.com/news/infostructure/0,1377,64178,00.html?tw=rss.TOP
>>
>> [3] Geolocation by IP Address, October 2004
>> http://www.linuxjournal.com/node/7856/print
>>
>> [4] SIP Location Conveyance
>> http://www.ietf.org/internet-drafts/draft-ietf-sip-location-conveyance-01.txt
>>
>> Candace
>>
>>
>>
>
>-- 
>Ben Teitelbaum   http://people.internet2.edu/~ben/

--
Jiri Kuthan            http://iptel.org/~jiri/