[90305] in North American Network Operators' Group
Re: Geo location to IP mapping
daemon@ATHENA.MIT.EDU (Alain Hebert)
Mon May 15 17:37:00 2006
Date: Mon, 15 May 2006 17:36:33 -0400
From: Alain Hebert <ahebert@pubnix.net>
Reply-To: ahebert@pubnix.net
To: davids@webmaster.com
Cc: nanog@nanog.org
In-Reply-To: <MDEHLPKNGKAHNMBLJOLKIEMFLOAB.davids@webmaster.com>
Errors-To: owner-nanog@merit.edu
Yeap,
I'm moron. You didn't know it yet?
-----
Come on...
"The way we disperse static IP" ain't imagination, its fact... We
spread a /20 on dynamic dialup and dsl over 2 provinces and since most
of the residential services is build like this you cannot get a read of
where that ip user is located, unless you have also access to our
customers db and authentication account db too.
The best you'll get is country. Even then we have some that are
LanEx'ed in europe and usa.
(FYI: English is not my first language btw... so dont expect too much)
David Schwartz wrote:
>> I'm sure that everybodies here understand that the city databases
>>cannot be accurate more than 50%.
>>
>>
>
> They *cannot* be?
>
>
>
>> The way we disperse static IP on commercial accounts there is not
>>way they can figure out where the destination is.
>>
>> The last best guest will be the peer router before my routers.
>>
>> For me the country db is good enought for basic webalizer report for
>>the customers websites.
>> (This way my customers dont waste queries to countries.nerd.dk on
>>non-spam related things)
>>
>>
>
> This is a pure argument from lack of imagine. They most certainly can be.
>
> Hypothetically, consider a company that had access to sales and account
>databases from sites like eBay, Amazon, and the like. It extracts from this
>database IP/city pairs. From this, it could do much better than 50%.
>
> You are basing your conclusions on your own lack of imagination.
>
> DS
>
>
>
>
--
Alain Hebert ahebert@pubnix.net
PubNIX Inc.
P.O. Box 175 Beaconsfield, Quebec H9W 5T7
tel 514-990-5911 http://www.pubnix.net fax 514-990-9443
