[arch-d] Re: [EXTERNAL] Re: IP Address Geolocation Considered Harmful

[Divyank Katira <[e-mail]>]

On Wednesday, March 25th, 2026 at 11:43 PM, Eric Rescorla <[e-mail]> wrote:

> With that said, I don't think it's particularly helpful for the IETF
> to tell people that in the current environment they shouldn't try to
> extract information from the IP address, as that's not going to have
> any meaningful impact. Instead we should focus on what technical
> measures would improve user privacy by decreasing the usefulness of
> the IP address for surveillance and tracking. Although obviously much
> of our ongoing work (VPNS, MASQUE, O-HTTP, etc.) tries to attack this
> from various angles, this is obviously a hard problem to address at
> scale while also maintaining network efficiency. This approach is
> consistent with our historical practice of favoring technical
> solutions to privacy and security (primarily strong encryption) over
> policy ones (restricting monitoring).

+1, this could be a reasonable way for IETF participants to intervene.

Ongoing work on MASQUE unfortunately does not move the needle on IP geolocation prevention.

Despite being marketed as IP privacy solutions, current MASQUE deployments like iCloud Private Relay [0] and Chrome’s (proposed) Advanced IP Protection [1] retain IP location when routing through relays.

This is currently done by maintaining numerous egress IPs [2] but there is also a proposal to relay this information through an HTTP client hint [3].

[0] https://www.apple.com/icloud/docs/iCloud_Private_Relay_Overview_Dec2021.pdf
[1] https://github.com/GoogleChrome/ip-protection?tab=readme-ov-file#ip-geolocation
[2] https://mask-api.icloud.com/egress-ip-ranges.csv[3] https://datatracker.ietf.org/doc/draft-pauly-httpbis-geoip-hint/

>