Re: IETF mailing list

[Frank Kastenholz <kasten%europa.interlan.com at RELAY.CS.NET>]

The exploration is probably useful. Other lists
do have problems - but usually they problems
get stomped on pretty quickly and the discussion
returns to the "topic" of the list.

As to our list - have these complaints actually
been voiced on the list? Have the offenders
been stomped on as necessary? The only recent
discussion thread that I recall that was "questionable"
was the pothole discussion before the Pittsburgh(?)
IETF. I'm sure Phil Gross remembers it.

As to mail list policies in general - I think that
they should be "exclusionary" in nature. They
ought only to list what types of discussion are not
allowed instead of listing only the types of
discussion that are allowed. The reasons are:

1) If the discussion of list X is narrowly limited,
   then there will be an explosion of lists. E.g.
   one for ietf meeting announcements, one for IAB
   minutes, one for driving conditions in the IETF city,
   one for mail list policy discussions, etc, etc. 
   Remembering which list I am on and which list
   to send what to would be a pain.
   
   It also means
   that there will be alot of cross postings, which
   means that I will get more mail. The SNMP
   work has been divided into at least 5 lists.
   QUite often discussion topics migrate from specific
   problems, suited to a highly focused list, to
   the more general form of the problem, which has
   wider scope. I then get 2 or 3 or 4 copies of the
   same EMAIL.....
   
   To be safe, people may cross post to all of the 
   sort of relevant lists anyway.....
   
2) By narrowly limiting discussion to a specific
   topic, the general discussion on the list can
   not evolve over time. We can not forsee, today,
   all of the topics that may be relevant to a list
   in the future.
   
3) A certain amount of not-strictly-relevant
   chatter on a list is, I think, beneficial.
   It tends to turn the folks on the list into
   people instead of just names at the end of
   an email. I think that this builds a sense
   of fraternity in the community at large, meaning
   that we will be more effective at dealing
   with the technical issues.
   
   In other words, we tend to become friends
   and it is easier to work with people who
   are friends and not just faceless,
   personality-less names.


Cheers
Frank Kastenholz
Racal Interlan