I have to completely agree with Robert here. I don't really have time to teach the users about SA. While I'm sure this isn't for an ISP, but more along the lines of a local sysadmin. It is better for them to tell me about problems, and let me take care of it, less trouble.
Chris Santerre System Admin and SA Custom Rules Emporium keeper http://www.merchantsoverseas.com/wwwroot/gorilla/sa_rules.htm "A little nonsense now and then, is relished by the wisest men." - Willy Wonka > -----Original Message----- > From: Robert Menschel [mailto:[EMAIL PROTECTED] > Sent: Saturday, September 20, 2003 4:27 PM > To: Don Saklad > Cc: [EMAIL PROTECTED] > Subject: Re: [SAtalk] How to deal with the learning curve for > spamassassin > > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hello Don, > > Saturday, September 20, 2003, 8:35:45 AM, you wrote: > > DS> a. How do you folks deal with the learning curve for spamassassin > DS> ?... It's too very steep for a novice! > > DS> b. If spamassassin is already on your providers' system, > how does the > DS> novice user configure things not knowing how to write dotfiles?... > > I'm not your average end-user, but perhaps I can answer some of your > question. > > I'm a domain master on a shared domain server. My host added > SA to their > system last Spring. The first I knew of it was when a new > option appeared > on my domain management control panel: SpamAssassin On/Off. > > I turned it on to see what it would do. > > I started getting spams flagged as ****SPAM****. Cool. > > I read the message info added, and headed to http://www.spamassassin.org where I learned enough to be able to modify a few settings and change some default scores. My knowledge of SA grew from there. Now MY users don't know anything about SA other than it's there. Because of the way my systems are set up, they have no direct control over SA (each domain has one user_prefs which applies to all POP addresses within the domain). They let me know when there's a problem, and I work on it. I have provided learn-spam and learn-notspam folders to my webmail users, so they can copy/move emails to those folders when they want to teach SA how to better identify these. I then feed the contents of those folders to Bayes via sa-learn regularly. That's about it. SA catches 90-95% of all spam out of the box, with almost no false positives. That's amazing, and more than satisfactory. With the tweaking I've done we're around the 99% mark, with one or no FP in any given month. My people have no complaints. Bob Menschel ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
