[BALUG-Admin] how to (and not to) post to lists, etc.

[Rick Moen]

Quoting Michael Paoli (Michael.Paoli@cal.berkeley.edu):

> You've emailed/posted/etc. many many times about how to, and how not to
> post to lists, etc. (e.g. plain text, don't do attachments, etc.) ... and
> we thank you very much for that!  :-) ...
> 
> To hopefully be a bit more proactive, do you have recommended web page(s)
> (most any suitable URL will do - but probably best if it's a web page or
> the like that at least somebody somewhere can tweak/update as/when suitable,
> rather than a read-only archived message that's web accessible) somewhere
> on that topic (if not, we could potentially create one, if we wanted).


Let me take another swing at this one, and my apologies for having been
rushed.  (I'll still be rushed, but have some thoughts I'd like to
articulate, from a slightly different angle.)

There are indeed pages out there that give tips about how to best
participate on mailing lists as to posting style, things to do and not
do, etc., especially as a newcomer to the online
technical community.  I can hunt them down and cite them -- but so can
you.  ;->

The thing is, most of the time those don't work very well, and, if you
think about it from the viewpoint of _process_ (sequence of events), the
reasons for this (general) failure become evident -- especially when
one's primary aim is to reduce the incidence of misbehaviour.  One is
that someone about to post essentially never stops to think "Oh, maybe I
should stop before posting and go read this group's tips on how to
participate first, so I don't mess up."

There's a necessary distinction between lists of tips and lists of
allegedly enforced mailing list rules -- but neither of them tends to
work well (in the above-described sense).  The people who would most
benefit from them, i.e., those who genuinely don't know how to post
cluefully and why, are least likely to read them (especially in
advance).

And then there are those who _do_ know better:  These will often ignore
any posted rules or guidelines entirely, and if challenged will cite
nonsensical, invented-after-the-fact reasons why they're supposedly an
exception -- or a corner case that they'll assert that you haven't
covered.  (Computerists tend to madly adore babbling about alleged
corner cases.)  A large fraction of computer geeks think that if something
is posted somewhere, and expresses a point of view, that implies that 
the matter's up for debate.  The more you write on the subject, and the
greater the detail you expound, the more they assume it really is up for
debate.  

You couldn't possibly be serious, they think, if you're that verbose.
They're trained to think that a big, red sign saying only "STOP" is
probably serious, while one saying "Stop, except of course as required
for safety, e.g., to get out of the way to avoid collision, and in
certain other cases" isn't serious at all.  That's human nature.  That's
how _real_ people think, which is just not the way computerists assume
they do.

Moreover, most people in all populations operate from a tacit assumption
that, if it's physically possible to do something, it must be OK -- that
any rule worth respecting is enforced in a way that makes it the
path of least resistance, and that any rule that can be ignored is just
noise.  Suggest to them that they should _not_ follow the path of least
resistence just to do the right thing, and they'll want to know why they
should do you a favour.(!)

(Consider the number of technical people on Linux mailing lists who post
from GMail and who both top-post and quote entire prior message threads in
every posting.  Do you think it's because they don't know that's wrong
and rude?  No, they're almost uniformly fully aware of that.  They just
can't be bothered to do anything requiring extra effort.)

And then, a certain number of people will argue with anything:  If you
put up a page saying "Please don't post Lotus 1-2-3 spreadsheets",
someone will figure out how to create one, and post it, just to push
back.  Or invent some alleged corner case, and do likewise.

A great many cultural-norm rules are present in automatically enforced
form, by being embedded into the software.  In many cases, those
magically cease to be perceived as rules:  They become perceived as
merely the base reality to which subscribers accomodate themselves.  

For example, Mailman has a built-in ceiling on individual message size,
which by default is low enough that it snags almost all, e.g., MS-Word
attachments, but high enough that all legitimate postings go through.
Do we see a lot of angry postings protesting the listadmin's tyranny in 
cruelly rejecting Joe User's 40MB PowerPoint attachment?  Nope.  Joe
gets the automatic reject notice, and thinks "Oh, I guess I can't do that."
No debate then ever arises.

Back in Majordomo days, there was an incessant tussle with people who
erroneously sent "Please unsubscribe me" postings to the broadcast
addresses of mailing lists.  They would screw up, get chewed out for
being twits, protest the supposed purity of their motives as making it
perfectly OK to be asshats in public, attempt to blackmail their way
into having the listadmin figuratively wipe their noses for them (i.e.,
unsubscribe them anyway just to shut them up), and so on.  Mailman put
an end to about 99% of that through its automatic "administrivia" filter
that intercepts most such posts and redirects them to the listadmin.

The crossposting is another case in point:  People here may recall how I
kept politely asking Kristian Erik Hermansen and several other people to
please cease crossposting between svlug@lists.svlug.org and
sf-lug@linuxmafia.com -- and his reaction was to scream listadmin
oppression and keep doing it, keeping his roster of recipients down to a
few, to avoid the "Too many recipients" filter.  (People who responded
to his posts, in contrast, often hit that filter, because they added one
or two recipients in replying.)

I tried for many months to just ask people nicely, and the problem kept
getting worse, and eventually just created a couple of filters to
intercept and autoreject a couple of the most common crossposts between
svlug@lists.svlug.org and elsewhere.  And, immediately, no protest
marches or namecalling about the mean listadmin, and no more attempts to
crosspost after just one that got bounced:  That's because it was now
seen as built into the software, instead of being something in words,
which is then seen as the user being asked to do someone a favour.

Now, I am _not_ suggesting hiding into the software various attempts to
manipulate members of mailing lists.  A smart listadmin seeks to cut to
the absolutely bare minimum the amount of hassle that he/she and also
list subscribers need to endure.  A smart listadmin runs the mailing
lists' unmoderated to the extent possible, and tries to automate
everything that can be processed without his/her help.  Less
regimentation and control means less work, less friction, more
pleasantness.