Martin Thomson wrote: > But I don't think that you can drop TLS 1.2 today without some care and that approach is not really generally applicable.
I don't think this statement is true. For example, one could easily write an iOS or Android app, and only communicate with big cloud providers or CDNs [0]. I agree that TLS 1.2 support can be described, but it is not the case that TLS1.3-only implementations [1] fail to follow best practices in any meaningful way. The handshake and key schedule [2] are different, so it can make sense to drop 1.2. Maybe the question should be flipped. Why is supporting TLS 1.2 a best practice? It's true that dropping support for any protocol version will lead to problems with implementations that aren't actively maintained. Surely the document could just say that? thanks, Rob [0] https://www.f5.com/labs/articles/threat-intelligence/the-2021-tls-telemetry-report [1] https://github.com/facebookincubator/fizz [2] https://github.com/rustls/rustls/blob/52e053e48b44c9c7d50df56f1c23898f4cf26231/rustls/src/tls13/key_schedule.rs#L60
_______________________________________________ Uta mailing list Uta@ietf.org https://www.ietf.org/mailman/listinfo/uta
