Terry Milnes said: > Frederick M Avolio wrote: >> At 06:13 AM 11/25/2003 +0100, Mark wrote: >> >>> Is SA hard to install? Not harder than any other program based on >>> Perl. Perl >>> is the underlying engine, which, in and by itself, has nothing to >>> do with >>> SA. It is good to keep that in mind; >> >> >> Perhaps this is *my* problem. November 14 I mentioned having >> problems >> upgrading to 2.60. One of the rules was giving an error. No one >> suggested upgrading to a new version of Perl. I installed both >> from the >> tar file and using CPAN. Neither time did I get a dependency >> error. >> >> Yesterday I again tried to install 2.60 on RedHat Linux 7.3. Some >> of the >> tests failed. I know my way around UNIX systems but am not a Perl >> expert >> (I know it when I see it). I could find no help for what might be >> the >> problem. Nor did anyone respond to my question about it on this >> forum. >> >> Like many things, SA works very well... if installed properly. It >> installs easily... usually. But when one comes up with a problem >> that no >> one else (apparently) has seen, there seems to be nothing to do >> about >> it. I massaged my system back to using 2.55. I have tried on 2 >> different >> 7.3 systems to get 2.60 to work with the same results. >> >> Fred > > No thats incorrect, you are thinking about commercial software here, > if > no one can solve the problem with the closed source software you are > SOL. This is open source software, if you can't find the answer, > dig > dig dig.... > > SpamAssassin is a product that works on damned near all platforms, > it > isn't some silly beta that is dependant on all kinds of obscure > packages, consequently you can make it work on your system. > > Stop and think, look for the differnces between the SA versions, > look at > the failed tests. > > If only one rule was giving you a problem and you can't figure out > why, > try it without the rule.
I disagreed with the majority of Mr Harbaugh's article regarding SA's effectiveness. I've been using 2.60 for almost two weeks with approx 400 spam correctly identified, 800 ham correctly identified, 0 false positives, and 1 false negative; that's 99.75% accurate. I use Bayes, RBL's, checksumming, and "normal" SA pattern checking; it seems to be doing an excellent job for me. I also have no doubt that whatever other new anti-spam technologies become available will make it into SA pretty quickly. However, I think he did have a point regarding the difficulty of installation. In some ways this is not a fair comparison, since SA is designed to be effective and to play nicely with traditional Unix installation procedures. However, any admins who are used to install wizards a la Windows will undoubtedly find SA more difficult to install. In my mind, there is space in the market for a commercially supported version of SA. This could include a "canned" installation script including paid support. So the choice would be: download SA for free but be willing to get your hands dirty with Perl and config files, or buy the prepackaged version and reduce the complexity of the installation. So, anyone know of a company like this towards which to point Mr Harbaugh? -- Kurt Yoder Sport & Health network administrator ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
