Two lists

[from Joel Judd 950725.1145 CST]

I was inclined to applaud Gary's notice about two lists being formed. I
[erroneously] assumed I would be hearing more from people such as Ed
Ford whose work and results appeal to me more than programming language.

There are upwards of 100 people signed on to the net; lately about seven
of them have accounted for the vast majority of postings. Is this
lurker/poster ratio typical of e-mail lists? A separate list would be
welcome if it invites a wider range of participants.

ยทยทยท

TO: CSG-L INTERNET Any user on the Internet, not at DESE Proj. Box

FROM: JUDDJ DESEINST Joel Judd - DESE - Division of Instruction

DATE: July 25, 1995
SUBJECT: Two lists

There are upwards of 100 people signed on to the net; lately about seven
of them have accounted for the vast majority of postings. Is this
lurker/poster ratio typical of e-mail lists? A separate list would be
welcome if it invites a wider range of participants.

The KSI monitors large numbers of lists and models the discourse. The
distribution of postings generally follows Ziff's law, and a small
small proportion of members accounts for most of the postings.
In particular, small lists with only 100 or so members are
critically dependent on one primary poster. If that poster leaves
and is not replaced then the list activity dies.

CSG is remarkable among lists for the length of messages, for the continuity
of volume, and for the scientific quality of the discourse -- the list
has become the primary medium for the community -- this is still rare.

General experience is that the effect of splitting lists is erratic. If
most members are interested in most postings it is best to stay with one
list. A split is best managed in terms of getting rid of a high volume
stream of discourse that interests only a small part of the community.

b.

Dr Brian R Gaines Knowledge Science Institute
                                University of Calgary
gaines@cpsc.ucalgary.ca Calgary, Alberta, Canada T2N 1N4
403-220-5901 Fax:403-284-4707 Knowledge Science Institute

Tom Bourbon [950725.1451]

Re: a post by Brian Gaines (with no header):

The KSI monitors large numbers of lists and models the discourse. The
distribution of postings generally follows Ziff's law, and a small
small proportion of members accounts for most of the postings.

. . .

CSG is remarkable among lists for the length of messages, for the continuity
of volume, and for the scientific quality of the discourse -- the list
has become the primary medium for the community -- this is still rare.

Very interesting. Some of us spend all, or nearly all, of our e-mail time
on this list. We have little, or no, knowledge of how it compares with
others. (By the way, for people who mistakenly think PCTers dislike all
statistics, we don't. The statistical data summarized in Brian's post are
legitimate -- descriptions and comparisons of the proper kind.)

General experience is that the effect of splitting lists is erratic. If
most members are interested in most postings it is best to stay with one
list. A split is best managed in terms of getting rid of a high volume
stream of discourse that interests only a small part of the community.

Something to consider. It's too bad you couldn't attend the meeting -- and
not just so you could provide comparative data about traffic on lists.

Tom B.