Re: User Estimates

[Ginger Booth]

> Averaging these numbers seems odd -- why average the number of
> unique posters PER WEEK (?) with the number of people on swarm-support?  

I certainly wouldn't qualify as an active user.  (Applying my application
instead of debugging the back end at the moment.)

FWIW, on our projects back at a prior employer, we hooked up a gizmo that
took stats (counters) for our software usage, collected 'em up locally, and
once a month, emailed the local site stats off to us.  (It also emailed us
tracebacks each time our programs crashed.)  We didn't get stats from
anyone not networked, but that was a minor omission, since our heavy duty
users were quite networked.  It's a little tricky deciding what to count
and when....  Anyway, it's such a basic idea, I imagine some implementation
is floating around free somewhere.  Such stats are useful in management
decisions, justifying funding, and the like.  Be careful not to count something
that will easily overflow the counters (like random deviates generated ;).
My boss didn't believe me that our graphics editor was 100,000 lines of code
that did MOVE_LINE and nothing much but MOVE_LINE 24 hours a day.  So
I counted 'em. :)  I believe it overflowed the counter.

Ciao,
            Ginger Booth, Center for Computational Ecology
                Yale Institute for Biospheric Studies