On 1/26/2024 11:33 AM, Brian E Carpenter wrote:
Hi Ted,
On 26-Jan-24 09:51, Ted Hardie wrote:
Hi Brian,
I don't understand this objection: "None of that works if we leave
any statement at all (either the 6 months text, or a purported expiry
date) in the document itself." The field tells you either when the
document's author thought it expired or when it will expire. This is
not a change from our current practice. Yes, you can have an archival
copy of something that has expired, but you know either that it is
stale or when it will become so.
No, you don't know much at all. Version -07 might have been replaced
after ten minutes by version -08, or it might still be in IESG
processing 13 months after being posted. The expiry date given in the
text is totally unreliable information. If we do nothing else we should
remove it, but I'd much rather replace it by an explicit URL for the
corresponding status page in the tracker. If we do that, I don't really
care whether we abolish notional expiry as Martin proposes or not, but
an explicit active/inactive bit does seem better to me.
I think that Martin's draft is addressing an existing issue, and that
Brian's suggestion to replace the "expires" line by a pointer to the
state in the data tracker is fine.
I did not see in the draft a discussion of the "tombstone" practice,
i.e., publishing a draft-example-nn that just says "this draft was
abandoned, the authors believe it was a bad idea because of X, Y, Z."
Should it be in the scope of the draft?
-- Christian Huitema
