From: Per Abrahamsen (abraham@dina.kvl.dk)
Date: Thu Oct 05 2000 - 06:28:06 CDT
Brad Templeton <brad@templetons.com> writes:
> On Wed, Oct 04, 2000 at 02:03:23PM +0200, Per Abrahamsen wrote:
> > I will send a CC when I believe a CC might be appreciated, no
> > matter what the standard says.
>
> Wait a minute, you want to send a CC if the poster says "don't send one"
> or you just plan to send them when the poster says nothing but the
> spec says to not send one in this case?
>
> If the former, you must be crushed like a bug. :-)
If the poster explicitly asks not to receive CC's, it is unlikely that
I would believe a CC would be appreciated.
> For the latter, I presume you are doing it because you want to call
> special attention to the message? (You will, you'll get a nasty note
> from me if you do it to me.)
Other people might do it for that reason. I wouldn't. I would do it
if I had reason to believe the poster does not read the group (for
example if the specific group culture doesn't require posters to read
the group, or if it clearly is a newbie who doesn't know Usenet
culture). Yet other people will send CC's because they like to
receive CC's, and assume everybody else does as well.
> The spec does allow me to put in the explicit declaration to deal with
> the likes of you, but you're saying we should add a whole extra header
> so that people who
> a) Don't express a preference, but
No, just the people who does not explictly express a preference not to
receive CC's. Which would be everybody who does not both hate CC's in
all circumstances, and know how to customize their posting agent to
express their dislike.
> b) Are worried about people who will ignore the spec and mail them
> anyway can
Or follow the spec, if the original poster has not explicitly
expressed a wish never to receive CC's.
> c) Get a fancy mail or newsreader that knows to discard one of
> the messages if it knows I will see the other.
I don't want to discard it. I just want to know whether it is a
public or private reply. A mail reading agent doesn't have to be
fancy to support that, in Gnus it would be a one-line change (just add
PAM to the default value for the list of important headers).
> However, why not just use the comments of the in-reply-to header?
> Seems the natural place. Or throw in a "References:" header if you are
> adding things to mail anyway.
Well, basically I like the PAM solution because it is existing
practice. A comment would not help, as comments are not structured.
A references header would not help, as many reply agents already
genererate references headers. However, including a Path header in
the mail might do the trick, and could also be argued as existing
practice (Pine uses the path header to distinguish between mail and
news). It will be less obvious, though.