At 11.17 -0700 01-08-12, Bill Barowy scrobe:
>the mathematical recreation of the pattern (i.e.
>derivation) assumes, as I mentioned before, that events in any category are
>independent of prior events, i.e. postings by any person are independent of
>their prior postings.
Oh, is that it? what the assumption is: independent of their OWN prior
postings? Was I wrong to think each posting event should be independent of
ALL prior events for Zipf to be applicable?
But... perhaps it doesn't matter. Some categories of postings - notably
announcement type stuff would be independent in any of the senses. But most
other postings are, as you say, acts of remembering = related to one or
more preceding postings. (No math necessary beyond the few-many
distinction, for making that observation.)
And I'd say that for a lot of the subscribers who post a very small number
of messages in a given time, it is quite typical that their later postings
depend on their first posting in the sense that they are responding to the
responses they have received. (Although another mailinglist phenomenon is
the little cluster of "please unsubscribe me" postings sent within a few
minutes by someone in a panic, who will then never appear again - in a case
like that the later postings don't seem to remember, or believe in the
existence of, the first one.)
I don't know about their first posting... although it DOES happen that
people post to say THAT and/or explain WHY they have been silent
participants. This is common on any list that brings up the topic of
"lurking" once in a while. So those first postings depend on remembering
that you have NOT posted before ;-)
Eva
This archive was generated by hypermail 2b29 : Mon Oct 01 2001 - 01:02:07 PDT