jwcolby
jwcolby at colbyconsulting.com
Fri Aug 10 21:04:05 CDT 2007
I find it amusing that anyone takes the time to generate a 10 page "why CR is bad", the basis of many of which is "it should punish the spammer", as if ANYONE has a way to do that. My personally opinion is that hunting down and colbyizing a handful of them VERY publicly would be the only effective deterrent but that won't be happening either. Additionally, so far I have never had email delivered to me because someone spoofed my email address as the sender of spam and a CR system was bouncing it back to me. Given that my name is on the spam lists (I get spam) THAT argument isn't keeping me up at night. I did not design the wonky system that allows spammers to spoof senders and I can not do anything about others doing so. In the meantime, I have at most 100 people in my address books, all of which have already been picked up (automatically) by the system. The rest are trickling in and I am manually dealing with them. Within a single week all of my regular emails will be handled. I used a Bayesian filter with outlook and tried to do so again but it wouldn't install. When it worked, it worked fairly well (98% rate) but had false positives and false negatives, few but still there. Having 2% hiding in the 100 is almost worse than 50%. You have to look at each one to find the 2 in 100 that you need to recover. THAT is as much of a PITA as just hitting the delete key 50 times a day. There are a million systems out there for handling spam, none of them perfect. I have tried about 500,000 of them so far, I know none of them are perfect. Of course if any of you fine folks wants to volunteer to set up and maintain an email server / AntiSpam system on my server machine, or install your favorite variation of Linux and your favorite variation of anti spam, please take my invitation to do so. I do have a beater box (not even so beater) and I will give you remote access to the box in order to do your thing. Of course YOU will be responsible for all maintenance for the rest of your life. I have real work to do unfortunately. In the meantime, I will be trying this one for awhile. I have had to respond to a handful of such "response required" from a handful of people I have emailed, and I did so, no biggie. I can see that some think it is a poor idea but such is life. John W. Colby Colby Consulting www.ColbyConsulting.com -----Original Message----- From: accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of Steve Schapel Sent: Friday, August 10, 2007 7:50 PM To: Access Developers discussion and problem solving Subject: Re: [AccessD] OT Friday: Comodo AntiSpam Someone may be interested in this critique: http://linuxmafia.com/faq/Mail/challenge-response.html Regards Steve Gustav Brock wrote: > Hi JC et al > > JC has sent this out, so now you cannot get to him directly. Aside the > good intentions behind doing so, this will cause a lot of trouble, and > I don't say too much if I - as a general warning - mention that the > consensus between system people is, that as tempting these > challenge-response system my seem, they represent a bad idea because > the negative impact outweighs the positive. > > If you or your client run a mail server, I can strongly recommend > SpamBunker: > > http://www.aimconsulting.ch/Spambunker/sbk-strategy.asp > > which is not free but cheap, and is superior to the common mail > filtering systems. Thus, it requires no "learning", no maintenance, it > works from the minute it is installed, and it handles massive amount > of spam and/or mail even on modest hardware. We operate it here where > we went from about 4000 spam mails per day to a handful per week after > we had been under a week long attack with constantly 20 connections > and more than 16000 connections per day. > > As a result we now promote this to our clients with serious spam > troubles. > -- AccessD mailing list AccessD at databaseadvisors.com http://databaseadvisors.com/mailman/listinfo/accessd Website: http://www.databaseadvisors.com