Talk:Protocol ossification

Further expansion potential
Didn't find these in the sources I found, which I'd've liked to have included: Polyphemus Goode (talk) 09:46, 20 November 2022 (UTC)
 * More history of the problem. When did people really start paying attention to ossification? What are the major landmark results? When was the term coined? Is there prehistory?
 * What rate of intolerant paths/middleboxes makes a protocol undeployable on the Internet?
 * Linkage to Moxie Marlinspike's essay titled 'The Ecosystem is Moving', and more broadly the topic of protocol evolution & deployability. That might make for an article in its own right if the right sources can be found.

Wire Image
The term "wire image" is used repeatedly - but is nowhere defined. 2601:600:C880:34E0:E5B1:44C3:64DA:2BCE (talk) 06:17, 25 January 2023 (UTC)


 * I took the term from RFC 8546 as a useful abstraction that basically means all the information that can be gleaned about the communication by participants other than the endpoints. I was thinking it might make sense to turn the redlink blue with a redirect to a new section of communication protocol describing the idea and implications of different design points, viz cleartext wire images are bad for privacy, unauthenticated wire images allow middlebox meddling, but encrypted wire images knock out some presently-useful operational tools and research opportunities. Polyphemus Goode (talk) 10:39, 25 January 2023 (UTC)