Timeshape of the xlist community

Eva Ekeblad (eva.ekeblad who-is-at ped.gu.se)
Sun, 2 Nov 1997 17:27:08 +0100

This was sitting in my outgoing mailbox waiting to be sent to David Dirlam,
who is my consultant and collaborator in quantitative matters. I read it
through once more and decided it was not too esoteric to send to you all,
in spite of some references to previous backstage conversation:

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
//////////////////////////////////////////////////////////////////////////
Another quantitative overview of the xlist data:

unique
year msgs contributors

1988 367 80
1989 690 112
1990 952 140
1991 1759 214
1992 1719 255
1993 1797 275
1994 1991 260
1995 2227 271
1996 2294 247
1997 1976 276

1997 obviously isn't complete yet: it's data from the first three quarters.
1996 suffers frome some data loss due to procmail expiry dates.

Aggregating the data into years instead of months or quarters also produces
a quite interesting picture, as this aggregation just adds messages
together but "telescopes" the number of unique contributors -- you can see
it's been quite stable for a number of years. A pity, really that Guastello
& Philippe don't have data on contributorship but only on number of
subscribers. There is far from total coverage of subscribers in the
x-archived materials, and I have also not yet processed all of the
masterlists given in the material, but I HAVE looked at the lists given for
=46eb 15, 1994 and Apr 4, 1995, i.e shortly before the May 94 restructuring
of the lists and almost a year after (here, too, "telescoping" the lists
into unique subscribers):

year subscribers
1994 631
1995 422

Compared to the figures above it looks as if the changeover didn't disturb
the active community a lot, but reduced the number of lurkers quite
substantially. I guess that may have been among the goals?

Then this is how many contributors also posted something in the immediately
preceding year (again, I have aggregated further from the continuity
between quarters):

year cont. stayons %
1989 112 50 45%
1990 140 66 47%
1991 214 83 39%
1992 255 131 51%
1993 275 133 48%
1994 260 127 49%
1995 271 117 43%
1996 247 116 47%
1997 276 113 41%

And I have also looked at the continuity between the two complete years the
farthest apart I could get, 1989 and 1996 (if now 1996 can be considered
complete). Between the 112 contributors in 1989 and the 247 contributors in
1996 there's an overlap of only 23 individuals! (And then one of them is
Mailer Daemon!!)

I think all this is very interesting as quantitative measures of what it
takes to sustain a cultural continuity such as on the xlists -- I mean
there's evidently a balance between stability and change here. One of my
original concerns in this has been to describe the conditions for learning
from the xlists and applying it to distance education (or other course
related CMC). In a sense the quantitative description of the xlist
community is a bit sobering here. UNLESS distance education is re-thought
quite drastically in its temporal structure, to accommodate to a community
of learners NOT all at the same sequential step in some given course, but
nevertheless in some sense taking the same program. Actually I suspect that
this is what places like the British OU already do, but I haven't checked
the facts. I also wonder about how the xlists stand among other scholarly
mailinglists: there is a vast number, I realize, although it's hard to tell
how many are active (not sleeping) etc. The Web, where I've been looking,
is for obvious reasons not very well updated, old stuff hangs on,
proclaiming to be current. There is probably research done in some corner
of much the same kind as what I have been doing, although I haven't found
it (what I do find tends to be "juicier", like stuff on the spatiality of
cyberspace and subjectivity in MOOs etc.) Which may be because I haven't
been struck by the best search terms for quantitative data on CMC.

Eva