Privacy thoughts on DNS over QUIC ?

[emikaadeo-git]

So DNS over QUIC (DoQ) has become a proposed standard https://datatracker.ietf.org/doc/rfc9250/
In addition to the performance advantages, DoQ reguires Server Name Indication (SNI) to be encrypted.
Now, let's asume a typical scenario: user don't want ISP to see what website he visit.
It's possible to achieve this goal with DNS over TLS (using a public DNS resolver) even if ISP is doing a Deep Packet Inspection (DPI) but there's is one requirement -> server hosting a website must support TLS v1.3
Is there any difference with DoQ ?
AdGuard, which was one of the first public DNS resolver to support DoQ states:
In short, here are the main advantages of DNS-over-QUIC: It encrypts DNS traffic. Nobody except you can see what websites you visit (...) https://adguard.com/en/blog/dns-over-quic-official-standard.html
Is DoQ really makes this possible ?

[jonaharagon]

DoQ reguires Server Name Indication (SNI) to be encrypted.

Source? SNI has nothing to do with DNS, so I don't see how this could be the case. If you want eSNI, the webserver you connect to must support it, not your DNS resolver. RFC 9250 which you linked to appears to answer your overall question in its introduction, emphasis mine:

The goals of the DoQ mapping are:

1. Provide the same DNS privacy protection as DoT [RFC7858]. This
   includes an option for the client to authenticate the server by
   means of an authentication domain name as specified in "Usage
   Profiles for DNS over TLS and DNS over DTLS" [RFC8310].

[...]