Proposal: convention for signing posts, encourage signing posts, retroactively sign...

diff --git a/doc/forum/convention_for_signing_posts_to_ikiwiki.info.mdwn b/doc/forum/convention_for_signing_posts_to_ikiwiki.info.mdwn
new file mode 100644 (file)
index 0000000..ad99625
--- /dev/null
+++ b/doc/forum/convention_for_signing_posts_to_ikiwiki.info.mdwn
@@ -0,0 +1,24 @@
+I habitually sign my messages (comments, forum posts, discussion pages, or replies to same) on [ikiwiki.info](https://ikiwiki.info),
+but not everybody does; especially newcomers might not realise it can be helpful. I think it should be generally encouraged that
+we do, and  think it would be useful for others to add
+signatures to unsigned comments/forum posts/discussion pages/replies in those situations, perhaps at the point of replying to
+them.
+
+But what should a standardised signature look like? I propose the format I am using, namely
+
+    *— (user-name), YYYY-MM-DD*
+
+Where "user-name" should be the logged-in user or git commit author.name field, if either are defined; that "user-name"
+should be a wiki-link to the relevant User page if such is known to exist; and failing that, use the string "unsigned comment";
+and "YYYY-MM-DD" is substituted with the year-month-day at the time of the comment. I don't see the need for a higher
+resolution timestamp (always available in the git metadata anyway).
+
+The signature should be suffixed to the last line of the relevant comment-block (as in this example).
+
+If there is general support for this idea, we could develop it further and have machine-assisted signatures, a bit like e.g.
+Mediawiki with `~~~~`.
+
+There are probably other similar conventions we should canonicalize, such as how to handle replying inline to someone else's
+comment/forum post/discussion pages/reply and what to do with respect to attribution/signature lines in those situations.
+
+Thoughts? *— [[Jon]], 2021-09-29*