Hi Elad,

> Not everyone will add the "geofeed" value and the geolocations in the csv will not validated by a 3rd party and may also be not up to date.

It is a fair point that not everyone will use this, but I think a significant enough number will to warrant addition. On the point of validation it would be perhaps advisable for the RIPE DB to verify the reachability of the file (as with rDNS setup), but I don't think further validation is within scope.

> I suggest to make the csv creation (of each INETNUM object) automatic and hosted by RIPE NCC

Ggood suggestion, not everyone will want to use their own webserver for this, and a simple service provided by RIPE would help improve adoption, however I'm not sure that people who can't host a CSV file should really be dealing in this anyway.

> RIPE NCC can use all the current ATLAS Probes in order to check the physical location of each ip

This feels like a guess at best, and providing guesswork as authoritative information is hardly a sensible idea.

Overall this proposal gets a +1 from me, and due to this purposefully light nature of the geofeed standard itself I don't think RIPE NCC should add too much to it.