Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#1

Post by sethito » Sun, Jan 04, 2015, 8:04 PM UTC

I've been using the geoip db from MaxMind for over a decade now; however, I've noticed the accuracy of both the free and paid dbs have fallen to unacceptable levels. I also recently learned that it is against their aup to build publicly facing systems based on their data (https://www.maxmind.com/en/sitelicense).

Are there any accurate alternatives that don't use MaxMind's free db?

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#2

Post by phoenix69 » Sun, Jan 04, 2015, 9:45 PM UTC

http://www.w3schools.com/html/html5_geolocation.asp

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#3

Post by sethito » Sun, Jan 04, 2015, 10:10 PM UTC

phoenix69 wrote: post #2

http://www.w3schools.com/html/html5_geolocation.asp

That works great until the user pushes "No, I don't want you go geolocate me".

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#4

Post by mavelikara » Mon, Jan 05, 2015, 5:42 AM UTC

I have had good success with NeuStar.

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#5

Post by mulligan » Mon, Jan 05, 2015, 6:53 AM UTC

I can understand the policy and I'm not sure it makes them evil. You could build a publicly facing system that was simply an API into their database, thus providing the public with the data you licensed.

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#6

Post by thenomad » Mon, Jan 05, 2015, 9:22 AM UTC

mavelikara wrote: post #4

I have had good success with NeuStar.

What are the costs associated with that?

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#7

Post by sarciszewski » Mon, Jan 05, 2015, 3:10 PM UTC

sethito wrote: post #3

phoenix69 wrote: post #2

http://www.w3schools.com/html/html5_geolocation.asp

That works great until the user pushes "No, I don't want you go geolocate me".

If the user doesn't want to be geolocated, why are you going to geolocate them?

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#8

Post by mavelikara » Mon, Jan 05, 2015, 4:11 PM UTC

thenomad wrote: post #6

mavelikara wrote: post #4

I have had good success with NeuStar.

What are the costs associated with that?

They were about the same cost or slightly more than the MaxMind paid version.

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#9

Post by dangrossman » Mon, Jan 05, 2015, 4:31 PM UTC

sarciszewski wrote: post #7

sethito wrote: post #3

Earlier quoted context omitted.

That works great until the user pushes "No, I don't want you go geolocate me".

If the user doesn't want to be geolocated, why are you going to geolocate them?

Basic web stats reports. Checking that you're getting the geotargeting you paid for when you advertise. Preventing credit card fraud (billing address in one country, user located in another, detecting open proxies, etc). Lots of reasons geolocation has been used since long before browsers had a mobile-style API for it on the client side.

Re: Ask HN: Less Evil Alternative to MaxMind GeoIP Db?

#10

Post by sarciszewski » Mon, Jan 05, 2015, 7:51 PM UTC

dangrossman wrote: post #9

sarciszewski wrote: post #7

Earlier quoted context omitted.

If the user doesn't want to be geolocated, why are you going to geolocate them?

Basic web stats reports. Checking that you're getting the geotargeting you paid for when you advertise. Preventing credit card fraud (billing address in one country, user located in another, detecting open proxies, etc). Lots of reasons geolocation has been used since long before browsers had a mobile-style API for it on the client side.

And if the user withdraws their consent (clicking "No", DNT headers, etc.) that should be the end of it. Don't locate, don't track, don't subvert user's consent.

Just flag it as "no location information available" if you need to use it for an access control/fraud prevention mechanism and be aware that determined bad guys can spoof almost every signal you receive from them.