[R] Trailing on r-help messages

[Martin Maechler]

Thanks to Gabor, Spencer, and Hadley,
for the constructive propositions.

>>>>> "hadley" == hadley wickham <h.wickham at gmail.com>
>>>>>     on Mon, 17 Jul 2006 09:39:36 +0100 writes:

    >> I think that is likely true but it would at least mean that they had
    >> seen it repeatedly and there would really be no excuse for not following it
    >> (unlike the current situation where one needs to take action to follow
    >> the posting guide link and then read a lengthy page).

    hadley> Logically, that makes sense.  However, try asking a group of R users
    hadley> how to cite R (or how to find out how to cite R).

    hadley> I think it's a good idea to make the default list
    hadley> signature as good as possible, I just don't think it
    hadley> will have that much effect.  Supplying a small
    hadley> reproducible example when reporting an error is
    hadley> "common-sense" (how else will the developers/others
    hadley> be able to replicate it?) and yet it is very rare
    hadley> (not just in R, but in any project).

I can well agree to slightly extend the R-help message footer,
e.g., to 4 instead of 3 lines.
Maybe we can move discussion of the final wording "off list".
I also agree that the effect from that change will probably not
be very big.

To other points made:

- Most R-SIG-* lists have their own maintainers (i.e. different
  from me).  It's really the maintainers decision (and "job") to
  change the default footer.
  I cannot change the default footer for all newly generated
  mailing lists here, because some are non-R related.

- The "inscription" page is dynamically generated by mailman, i.e.,
  typically python scripts.
  If any of you are interested in patching Mailman's sources in
  a reasonable way (i.e. easily reproducible for the next
  version of mailman), I'd consider a change there; otherwise not.

Martin Maechler, ETH Zurich
(Maintainer of R-help and a few other R-* mailing lists)