Personally, I’ve always thought that IP Geolocation was a bad idea and nothing I’ve seen in the usage of it to date has changed my mind. I get that there’s money to be made in providing it. I get that there’s content geofencing and other consumer abuses that depend on it and are considered desirable by content providers. From a consumer perspective, it’s 99% headache and 1% occasionally helpful in getting more useful search results. For the 1%, asking the consumer where they want search results for is usually more effective than most geolocation solutions, but because geolocation is perceived to be good enough, frequently, there’s no way for the consumer to specify their search area which makes it worse than not having geolocation. My worst experience with Geolocation was delivered by Wells Fargo many years ago. I was in Kigali, Rwanda for the AFRINIC meeting and needed to verify that my mortgage payment had been properly processed. I attempted to log into the Wells Fargo website from Kigali and was informed that there was no problem with my account, but I should log in when I get home as they don’t allow logins from Rwanda. (joy). So… I decided to use VNC to manage my computer back home and log in that way. This set off all the alarms and forced me into a 45 minute VOIP call from Kigali trying to explain the internet to the Wells Fargo technician who couldn’t comprehend VNC despite the fact that they were using RDP on a terminal server to do their job with a thin client as I spoke to them. (Somehow they just couldn’t grasp that a person in Rwanda could control a computer in California to access their web site). Eventually, I was able to convince them that VNC and RDP did similar tings and that the internet didn’t care whether it was on the same LAN or across the world. (The idea that my desktop at home had a public IPv4 address was also a difficult concept for them to grasp, so I left them with the delusion it was a port forward, there’s only so much education one can do over a VOIP call from Africa in 2010). Finally, they reset my account and I was able to log in, but it was an insane amount of effort, all because of misguided abuse of IP geolocation for reasons that made no sense and invalid assumptions about their customers. Owen .
On Feb 15, 2026, at 21:24, Hank Nussbacher via NANOG <nanog@lists.nanog.org> wrote:
On 15/02/2026 20:07, Tony Patti via NANOG wrote:
I compared IPinfo’s answers to MaxMind’s https://www.maxmind.com/en/geoip-demo answers, for a random recent visitor from 44.198.66.156 to my website https://cryptosystemsjournal.com/ and one notable difference is that you specify a larger CIDR block of 44.192.0.0/11 (2,097,152 addresses) whereas MaxMind specifies a much smaller CIDR block 44.198.64.0/19 (8,192 addresses).
Tony Patti
Unfortunately, there is not a quality academic paper that has come out yearly comparing the various geo-location DBs and their accuracy (IPv4 & IPv6) down to the 100m-1km level. So in the interim, we will continue to see geo-location DB wars - just like we have had OS wars and IP vs OSI wars.
Regards,
Hank
_______________________________________________ NANOG mailing list https://lists.nanog.org/archives/list/nanog@lists.nanog.org/message/2UA32SPC...