I think Paul has a point. Basically, there is error in all the measurements
we are using -- e.g., we don't get to count "hits" from cache either but we
still count hits. So we try to know what we can as best we can. turns out
some of the most interesting information is the time series -- long term
increases and decreases kind of wash out the error in a measurement.
i find "sessions" and "time in session" useful approximations of meaningful
events. It is quite interesting to look at whether "time in session"
increases over time and, I suggest, meaningful. Stephen is right that these
measures are soft, but how accurate are other commonly used measures? the
inflation rate? unemployment rate?
not that these numbers shouldn't be footnoted, of course ;)
dave
-----------------------------------------
Dave Witzel [EMAIL PROTECTED]
Forum One Communications http://www.ForumOne.com
ph: 703/237-8537 fax:703/995-4937
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Madley, Paul
> Sent: Friday, July 16, 1999 4:26 AM
> To: '[EMAIL PROTECTED]'
> Subject: RE: [analog-help] Sessions?
>
> > No. Because of the nature of the web there is no way to
> tell that a given
> ip
> > or hostname is the same user. See the item in the Analog
> FAQ about this
> > and the page "How the web works" in Analog docs for a complete
> > explanation.
>
> Hmm, 4.iii on
> http://www.statslab.cam.ac.uk/~sret1/analog/docs/webworks.html
> Is me, I _have_ to come up with some way of measuring
> sessions. If I were
> to try and explain to them about requests versus hits, they'd
> go "Uh?". And
> I'd be forced to actually read through the 16mb of log files
> currently on
> this hardrive!
> --------------------------------------------------------------
> ----------
> This is the analog-help mailing list. To unsubscribe from this
> mailing list, send mail to [EMAIL PROTECTED]
> with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
> List archived at
> http://www.mail-archive.com/[email protected]/
> --------------------------------------------------------------
> ----------
>
------------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
List archived at http://www.mail-archive.com/[email protected]/
------------------------------------------------------------------------
