Era Eriksson had written,
| > | Anyhow, one way or the other, this is obviously something that could
| > | rather easily be decided by some sort of majority vote or whatever.
I replied,
| > I must disagree. Where to send replies very rarely is something to
| > set one way for all posts to a given list. It should be determined
| > on a response-by-response basis, primarily by the first poster's
| > preference if stated and secondarily by the respondent's decision
Era rejoined,
| You are addressing this only from the private recipient's point of view.
I don't know how you got that impression. You cut off my next line: "...
and secondarily by the respondent's decision in light of knowing the first
poster's preference." It's a case-by-case determination depending on what
the poster wants and what the respondent, knowing what the poster prefers and
knowing what he or she wants to say in reply, decides is the most appropriate
thing to do.
For example, if a poster points replies to the list but I want to say some-
thing about his or her .sig quote that is unrelated to the content of the
post or to the topic of the list, then, given the content of my reply, I have
a good reason to overlook the other person's preference for public replies
and to send a private one instead (with no copy to the list). It is a case-
by-case determination.
| A reasonable consensus could, for instance, be to agree to add
| "list-cc" to the subject line of such messages, to make it easier to
| filter them out if you're not interested. (This particular tag is a
| bad choice because it will sound to people like they ought to add it
| to all messages they respond to with a Cc: to the list, when you would
| only want them to so tag stuff that is more or less mundane.)
I must be reading that wrong, Era. If there is a "[list-cc]" tag in the
subject of the copy sent directly to the previous poster, then the carbon
sent to the list will have that tag as well. (Nobody, but nobody, is about
to mail the same text once to the previous poster and separately to the list
to be able use two differing subject lines.) Then *everyone* on the list
will get a copy that says "[list-cc]" and we'll each still have to take a
peek at it to see if it matches something that was a "[list-cc]" for us or
not. Unless I thoroughly misunderstood you, that can work only if the list
software removes the tag from the publicly distributed copy, so that it ap-
pears only on the copy that was delivered directly to the previous poster.
Or, here's an idea: if a piece of mail is ^TO_you, also ^TO_(a list to which \
you subscribe), but not ^(From |(Resent-)?Sender:(.*\<)?)(that list), then
gosh, it must be a directy mailed copy of something also sent to the list,
and you're welcome to handle it as you pleaase; if a piece of mail meets not
only the first two conditions but also the third, it must be a publicly
posted copy of a list article that was also mailed directly to you and again
you are welcome to handle it as you please. There's no need for the sender
to do the extra work of tagging the subject. No, it won't catch doubly
addressed responses that are Bcc'ed to the previous poster or Bcc'ed to the
list, but how often does that happen? The doubly addressed responses are
almost always To: the previous poster and Cc: the list (or also Cc: earlier
posters in the thread); in rare cases, they are To: the list and Cc: the
previous poster. If getting two copies annoys a person so much that he or
she must allow for the possibility of Bcc's, formail -D can catch those
[unless the list clobbers Message-Id:'s, in which case formail -D won't work
even for articles whose double addressing is visible but my suggestion at the
beginning of this paragraph will].