[K12OSN] OT: Big Brother

[Chuck Kollars]

> Is anybody doing filtering that logs blocked pages 
> with user info? 

See
http://contentfilter.futuragts.com/wiki/index.php?title=Two_Configuration_Families

If you have (or set up) an "Explicit-Proxy"
environment, you can have your filter/proxy (ex:
Squid) request "ident" information, to which the
browser (at least IE) will automatically respond. Many
filters (ex: DansGuardian) can then "sniff" the
information and include it in their logs. (Use of
Firefox --which tries mightily to remain
multi-platform and so isn't tightly tied to Windows--
may throw a monkey wrench into the scheme.)

But if you have (or set up) a
"Transparent-Intercepting" environment, I'm not aware
of any way you can generate or capture the
identification information with any tool no matter
what. 

In many cases the same software filtering tool will
behave differently depending on which environment it's
in. (Beware) 

In these cases the filter logs will typically contain
just an IP Address, which you somehow need to
associate to a username. Often you can get the
username information from somewhere else without
requiring the user to reenter it (i.e. retain "single
sign on"). For example if your users connect to Samba,
you can run `smbstatus` over and over every minute or
so to generate the timestamped IP->name correspondence
information. 

thanks!

-Chuck Kollars