On 20/09/17 11:20, Mike Holmans wrote:
> I wonder if such ratios are still common on what's left of Usenet. In
> days of yore, a lurker:poster ratio of 100:1 would have been entirely
> believable.
In those days, it was not merely "believable", it was [with a
fair number of caveats] measured. It is probably still measurable in
principle, but no-one here will have the resources to do so. I don't
see any reason why the ratio should change dramatically. FWIW, I am
still subscribed to a reasonable number of groups, inc some reasonably
active ones, to which I never contribute.
> In those days, it was normal to see at least one new poster per day in
> any reasonably-patronised ng.
In that case, I think your "yore" is considerably more recent
than mine [mid-80s].
> Lurker-to-poster conversion ws
> definitely a sign that the ng was provoking interest from new people.
Yes, but lack of same is not a sign of anything interesting.
> But if we've seen a new poster in uk.s.c in 2017 (other than
> spambots), I've missed it.
Of the 34 posters previously alluded to, 6 were not known to me
as regular posters, and 3 had something to contribute. I don't know
whether that proves anything more than a fallible memory, nor whether
any of them were, are or will become regular lurkers.
> My impression is that newsgroups nowadays are only read by people who
> posted to them in years gone by, and that the armies of lurkers there
> used to be have all been demobbed.
Yes, but where does that impression come from? By definition,
we know nothing about the lurkers.
> I would be mildly astonished to find that more than a hundred people
> have read uk.s.c in the last month.
Two data points which may well be irrelevant but are perhaps of
interest:
(a) In my PP, I referred to a legal discussion about libel. In that
discussion, there was a reference to a specific court case, in which
the readership of the Huffington Post came into question; and, IIRC,
it was agreed that there were approximately 6000 readers. That needs
slightly more activity than lurking here. But it should be possible
to estimate the number of contributors to the HP [should anyone here
admit to reading it], which would give a data point towards a lower
bound on the lurker:poster ratio.
(b) In my capacity as Admissions Tutor, before I retired [2008], I
spoke regularly to large audiences of actual or potential applicants,
parents, grandparents and general hangers-on. Around once per year,
I would be asked, usually by a middle-aged gentleman, whether I was
the Andy Walker who contributed to this group [followed usually by
some comment on a then-current debate]. That would suggest that in
its heyday, roughly [say] 1 in 3000 of the population, call it 20000
people, read this group sufficiently to recognise my name. But that
is a serious underestimate, for the same "lurker:poster" reason --
most of the audience would not care to queue to talk to me just to
exchange pleasantries about cricket, and even less to stick their
hands up while the talk was in full flow. OTOH, there may have been
a degree of bias in the audience. So 20000 is at least a decent guess
at that peak audience. I accept that numbers have declined since.
But we still get ~30 posts/day [which would have made us one of the
top few groups in "days of yore", and would have taken around 5 minutes
of CPU time to process on a VAX 750 under B News, and a similar time
to expire each night -- multiply that by the number of active groups
to see that there was a real problem with newsgroups in those days,
with people complaining that they were taking over the departmental
computer]. I doubt whether we ever got more than 100 or so. Someone
with sufficient interest could no doubt count up numbers of posters
and articles around a decade ago from one of the archive sites.
Summary: From peak, I think we have lost perhaps 70% of the
articles, and, at a pure guess, there may well have been a similar
decline in the number of posters, whether regular or occasional. A
similar decline again in the number of readers would take us down to
perhaps 6000 readers. A *further* 90% decline would take us down to
about 600 readers. I'm guessing that that is a reasonable number.
You are claiming to be mildly astonished if even 17% of those are
still with us. Go figure!
--
Andy Walker,
Nottingham.