Hi Jen, > So while I have some ideas of what improvements I need from a given protocol, I'd not be writing actual implementations, because I have no places to actually run it in the production. > I'd rather collaborate with people who can write that code much faster and better than I would ever be able so I can run their code in production and provide some feedback from my operational experience. It is important in this discussion to distinguish a couple of things. - First, there is prototyping software that helps to provide input while writing the specification. Ideas that look great on paper may not necessarily work well when implemented. - Second, not everyone in the working group is likely to implement a given protocol. However, someone has do the work. If we write specifications without any implementation input then the quality will be lower. I am arguing that at least the authors of documents should have enough incentives to prototype their own specs. The leadership can help to create a culture in the IETF where implementation work is valued. So, if you are an author of a protocol spec then you should IMHO also verify the spec via an implementation. There are lots of specifications that do not describe protocols and hence they may not need any implementation work. They will benefit from different input, as some of the other folks on this list have pointed out. This may, for example, be operational know-how to configure already implemented protocols and to make observations with the installed base. This is an important area, as Henning Schulzrinne pointed out in his paper "Networking Research - A Reflection in the Middle Years", see https://arxiv.org/abs/1809.00623. I hope this clarifies my statements. Let's look at an example. The drafts you are involved in*, such as draft-ietf-6man-grand-03, look like protocol specifications to me (although I am not participating in 6man). If you care enough about draft-ietf-6man-grand, which I am assuming since you the sole author, you might want to check whether the content is actually sound. This can happen in various ways, such as by doing a prototype or by testing configurations (if that's appropriate). Ciao Hannes *: I hope I searched correctly. IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
