[plug] Re: Noise!

[Christian]

At 13:20 7/08/98 +0800, you wrote:
>I suppose that it also depends on what purpose you see for a list like
>this. I feel that it's more useful being informal and chatty, more like a
>group of friends/colleagues than a formal meeting of some kind.

I've got no problem with an informal list but "chatty" just will not work
as the number of list subscribers grows.

I'm all in favour of changing the Reply-to field to the originator of the
email and not the list.  Damion points out that this will prevent the rest
of the people on the list learning the result of the discussion but I think
the loss of information through changing the reply-to address will be
minimal.  Certainly not as much as will be lost eventually when everyone is
throwing emails left, right and centre and no one can obtain the signal
from the noise.  All changing the address field will do is force people to
consciously think "Does the rest of the list need to know about this or is
it possible they would benefit?" and this is definitely a good thing.  And
if you read my original email, it's one of the things I said then too.

>Personally, the number of messages gives me no problem, neither does the
>'signal to noise ratio'. Anyway, compared with Usenet the ratio is pretty
>good.   :-)

Of course it's pretty good - Usenet has the whole wired world "subscribed"
- that's why the signal to noise ratio perhaps isn't as important as the
signal to noise ratio PER PERSON SUBSCRIBED.  I'd say if you were able to
calculate some sort of semi-objective values for this of both this list and
Usenet, Usenet would put us to shame.

Here are my suggestions for guidelines for this list:

1.  Before posting to the list to *answer* a question or make a comment,
ask yourself if everyone (90%+) on the list wants to hear what you've got
to say.  If it can be directed to the originator of the post, do so.  (As
I've noted above, changing the reply-to address will help make this
guideline effective)

2.  If you've got what you think is a really basic question, especially if
you're a complete newbie and the question feels quite involved/complex, see
what information you can find about the topic first.  Doing a web search is
quick and easy.  Checking to see if there's a howto is also easy.  If it's
a specific program, checking if there's a man page is, once again, easy.
Especially on a mailing list, asking a specific question (eg, "Can someone
give me a sample chat script?") is better than a general one (eg, "How do I
connect to the internet in Linux?").  I still think IRC is a very good
first port of call in a lot of situations and it's worth thinking about.
Newbies should be encouraged to ask questions but they should also be
encouraged to look for answers to their own questions.  That's why people
bothered to spend their time writing howto's and man pages - to establish a
central and standard source of reference information - and to prevent the
same topics being discussed over and over again on thousands of mailing
lists around the world.  I'd say a key difference between Windows users and
Linux users is that Linux users can be bothered to think for themselves...
if they couldn't, they'd still be using Windows.  I think the mailing list
should encourage this, without discouraging people asking good, intelligent
questions.

3.  Don't respond to a response with a minor correction or fairly
insignificant extra piece of infomration unless it's really important.  And
then *consider* just sending it to the person who asked.

4.  If you end up having a private discussion off-list which might be of
use to people on the list, post a summary to the list.  That way no one can
complain that they're missing out on potentially valuable information and
we don't have to see that information being developed with painful detail.


I don't see these as "formal rules" or anything that's particularly going
to stifle the informal and vibrant nature of this list.  But they will save
time and bandwidth and end up with a more concise, useful list.

Can anyone suggest a practical problem with any of these guidelines?