Man, that article is indeed rather damning - in affect, though unlikely effect.
The article, flagrantly irresponsible, says more about the magazine, and the author, than any commentary about the reality of SA's effectiveness as an anti-spam tool. The fact that older SA versions don't hold up well against modern spam is probably because spammers have gotten better as a function of SA's effectiveness, as opposed to anything SA lacks. And SA developers continue to respond with effective solutions. Given that spammer ingenuity at getting past filters evolves with the filters' effectiveness, measuring a filtering tool based on an outdated version has to be the height of incompentence (or deceit) - especially when the author doesn't even go so far as to mention the program version he used, not the program itself, was outdated. The author responds to Robert by hiding behind his editor, though his article refers to SA as an older solution, first generation, as if SA were a long time forgotten, crashed, and burned test pilot. He gives the version number, but does not say what the current version is - that 2.44 is an older version. Somehow, I don't think the copy editor gambit excuse quite obscures his (or the magazine's) obvious anti-open source bias - e.g., the article headline: "[Various and sundry commercial products] overwhelm open source in accuracy, flexibility, and ease." I suppose the editor gambit is code for, 'My magazine is influenced by it's advertisers,' and is therfore not a reliable source for information technology reviews - though I don't know, as I don't generally read his magazine. Shops, professionals will opt for commercial products over open source for a variety of reasons I suppose. Defaming an open source product to grandize a commercial one is not neccessary, and really bad form. If not for SpamAssassin, it's highly unlikely spam filtering technology would be a fraction of where it is today. Bryan Robert Menschel wrote: > > Hello Logan, > > Sunday, November 23, 2003, 2:51:18 PM, you wrote: > > LH> The point of using the old version of SpamAssassin was to show how > LH> much the technology has changed in the last few years. That was > LH> stated in my original article but edited out of the final version. (I > LH> love copy editors.) > > Then it would have been good to have tested BOTH versions of > SpamAssassin, and to have compared them just as you compared each of the > commercial products against the ancient and aged version 2.44. > > The result of your article was simply to denegrate one of the best > anti-spam packages available. > > You have experience with copy editors. Could you have submitted an > article to them that wouldn't have been so anti-SpamAssassin? > > Will there be a correction printed in the next edition? > > Bob Menschel > > LH> Thanks, > > LH> Logan G. Harbaugh > LH> 530 222-1164 > LH> 693 Reddington Drive > LH> Redding, CA 96003 > LH> www.lharba.com > > > -----Original Message----- > > From: Robert Menschel [mailto:[EMAIL PROTECTED] > > Sent: Sunday, November 23, 2003 2:46 PM > > To: Logan G. Harbaugh > > Subject: http://www.infoworld.com/article/03/11/14/45FEspam_1.html?s=tc > > > Dear Mr. Harbaugh, > > > In your article, you state, > >> The five products I tested: ... and SpamAssassin 2.44, an open source > >> spam filter included with Red Hat Linux 9. > > and > >> In contrast to the commercial products, SpamAssassin represents an > >> older, first-generation anti-spam solution, and its age showed in my > >> tests. It filtered only 62 percent of spam, ... > > > Why would you intentionally test an ancient version (2.44) of a product, and > > then blame its age on the product? > > > Version 2.5x was available in April or May, and version 2.60 was released > > last month. Version 2.5x made great strides against spam, implementing not > > only new rules-based filtering capabilities, but also a Bayes database > > methodology. 2.6x has continued the improvement. > > > Run version 2.60 with network and Bayes checks activated, and SpamAssassin > > will easily catch 95% of all spam. Spend just a little time tweaking the > > scores and adding a few rules, and you can reach 99%. > > > My system consistently runs at 99.8% or higher. (Last week I processed over > > 5000 spam messages, of which only 5 slipped past SpamAssassin's filtering.) > > > Your report penalized SpamAssassin, not because of anything SpamAssassin > > does or does not do, but because you yourself used an ancient version of the > > product supplied by RedHat. You penalized SpamAssassin because RedHat > > provides an old version. IMO that is a serious disservice to your readers. > > > Do be more careful in the future. > > > Robert Menschel > > SpamMaster > > www.contractorswarehouse.com, www.xeper.org > > ------------------------------------------------------- > 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/ -- Nothing in the world has more potential for beauty than woman. Nothing has more potential to destroy it, than the world. - (Anonymous) http://www.wecs.com/content.htm This signature file is generated by Pick-a-Tag ! Written by Jeroen van Vaarsel http://www.google.com/search?hl=en&ie=ISO-8859-1&q=pick-a-tag ------------------------------------------------------- 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
