BIP 3: mention posting a dedicated ML thread

[jonatack]

Clarify that the BIP author(s) post a new, dedicated thread on the mail list to present the idea.

Address situations like #2017 (comment).

[dathonohm]

To anyone wanting to avoid the delay I experienced on #2017, make sure you subscribe to the ML before posting the PR. It took a day or so for the mods to approve me to post on the ML, then another day or so for them to approve my actual BIP announcement. During this time I had to repeatedly submit subscription requests and the announcement email, because there is not really clear feedback about whether the request succeeded.

I'm not sure, but just in case, I also recommend not using an @proton.me address, as this appears to trigger Google's defenses.

This was the cause for the delay.

[jonatack]

It is a requirement in BIPs 2 and 3 that BIP draft ideas begin with a mail list discussion. The idea here is to clarify that the required discussion be in its own dedicated thread, rather than a possibly tangential comment in a different thread, to give the idea sufficient exposure and discussion.

(Yes, best to hold off on opening a PR until after a discussion on the mail list.)

[dathonohm]

Now that you mention it, yes, I should have posted perhaps a day or two in advance on the ML before posting on Github.

For future reference, how much discussion do you think would have been necessary? I naively attempted to send the ML post at the same time as the PR on Github, but the first was immediately published while the second took much longer.

My apologies for overlooking that requirement.

[jonatack]

how much discussion do you think would have been necessary?

Quite a bit, normally. Suggest searching on the word "list" in the BIP2 text.

[luke-jr]

Seems fine, though I don't think it would have avoided the example situation. That had a dedicated thread (started by PortlandHodl), even if the top post wasn't the specific rules that got adapted - and that's likely to be the case for many BIPs.