perl CLF web analysis tool, testers wanted

[Jamie McCarthy]

Hi. I'm not happy with any of the free web log analysis tools,
and I don't want to pay for something so basic, so I'm writing
my own.

More strictly speaking, I'm writing a perl module that (I hope)
will allow anyone with a modicum of perl programming experience
to do some sophisticated log analysis.

In addition to the normal features you'd expect, like hit count
and so on, the code:

* recognizes hits by search engines, counts them separately,
and tracks each engine's spidering
* can recognize hits from the site administrator(s) and count
them separately (customizable)
* tracks referers
* recognizes referrals from search engines and tracks keywords
* tracks sessions, where a session is defined as one or more
hits from a particular IP number with no pause of longer
than n minutes between hits
* writes "pre-processed" analysis files for any time segment,
e.g. weekly or daily
* tracks the most common paths taken through your site
* is not terribly fast, but fast enough (1000 lines/sec on a
typical 200 Mhz PC)

The main feature will be that it's (1) free, (2) extensible,
and (3) provides enough tools that an average perl coder can
make it do what s/he wants.

What I'd like is feedback on (1) what features would be most
desirable, (2) how those features should be incorporated into
the module, and (3) how it can all be presented for maximum
friendliness to a typical webmaster who may not know a lot of
perl or may not have a lot of time to spend making it work.

Drop me a line if you're interested.

--
Jamie McCarthy
ja...@mccarthy.org
http://jamie.mccarthy.org/