Guidelines for moderating the R-help mailing list

Remember: The main goal is to keep the list open to non-subscribers. However it is not a problem if they have to wait a couple of hours before their posting is approved. As a matter of fact, that will remain an incentive to become subscribers.
Consequently, moderators should really only look at the moderation web page a couple of time times during 'duty day', and not more.

Main rule: You should typically NOT reject anything; rather either approve (non-subscriber postings) or discard (spam).
As with most interesting rules, there will be exceptions, see below.

"Duty" Days: In order that you don't tread on each others' toes, for each day of the week, we'll have one or two (maybe three) moderators assigned.
If there are two (or more) moderators assigned to a specific day, please coordinate on your own how you will typically handle "the day".
If you know in advance that you won't be able to attend a given duty day, please try to make sure another moderator takes over from you.

E-mail notifications: Note that currently all of you will receive e-mails about pending moderations, both

  1. when they happen and
  2. a daily summary of all pendings
Probably, we will turn off the first ('1.') pretty soon, but, moderators, please voice your opinion about this.

Rare Rejections: The following are considerations some of which may well be revised in the future.

  1. do never reject "professional" spam (Viagra, Stocks,...), but rather disgard it; rejecting only helps the spam mafia.
  2. Maybe reject (instead of discard) ``mass e-mailings'' that are only very remotely related to the R software or the R community.
    I would approve "Computational Statistics" conference announcements, but not a general ISI or JSM conference announcement. I might reject that, in the hope the sender will learn.
    This is a matter of taste though, and moderators will slightly differ here.
  3. It will happen that clueless newbies post, then see that the message is hold, and will wrongly conclude that they should just try again. This will lead to multiple postings from the same poster typically with the same or similar subject. In such cases, approve only one of the multiple postings, and reject the others, so the newbie can learn.

  4. ... (more to come ..) ...

Compiled by Martin Maechler, Feb 2008; updated by potentially all R Core members