Geolocation services: what's good, what's not?

[Simon Wistow]

On Wed, Dec 03, 2014 at 08:18:06AM +0700, Peter Sergeant said:
> The first question I'd ask anyone considering using anything other than
> MaxMind's Geo2IP product is "Why aren't you using MaxMind's Geo2IP product?"

I like and use MaxMind but, by all accounts, they're not as accurate as 
Digital Element (who are much more expensive) - I believe the figures I 
heard were 90-92% accurate for MaxMind vs 96-98% accurate for Digital 
Element but those are just anecdata.

Quova/NeuStar is also supposed to be good.

There are a couple of of papers (albeit from 2011) which go into 
accuracy

http://www.caida.org/publications/papers/2011/geocompare-tr/geocompare-tr.pdf
http://arxiv.org/pdf/1005.5674.pdf
http://www.caida.org/workshops/isma/1102/slides/aims1102_yshavitt.pdf


S