I have been told by my senior management that my IT services, mostly the spamfilter is hindering our business. Specifically they are saying customers and important suppliers cannot email us.

They also said other companies do not have this quarantine thing and do not ask users to go check the spam themselves, and they have no problem with spam or virus, nor do they have problems with their customers.

I want to hear what you guys' opinions are? In my view, there no system that would not generate any false positive. I cannot think of any solution to that either.

Obviously we ultimately do what management wants, however they have hired you because of your expertise and as such aught to be able to rely on your judgement.  If you have explained all the dangers of opening things up and they still insist on doing it, then I suggest you open things a little until they complain about all the spam etc they have to deal with.  Sometimes a child has to fall aleast once before they believe you that it hurts. 

In our case, the web interface works great.  Most of the time the customer calls in because of the rejection email they get back from SF.  Then the sales person goes to the web interface and forces it through.  Most of our sales people have learned to read the reason why it was originally rejected and the customer either fixes thing at thier end (Reverse DNS) or calls me for assistance.  In the end, you are stuck with either dealing with the complaints about false positves or too much spam getting through.  Unfortunately until everyone is playing be the rules there will be issues like this. 

John

John, really appreciate your input.

Is this true that other companies have no such issue?

I was thinking to open doors for some complaints, but SF does not allow me to pick what filters apply to which person - would be nice if it is that way.

Explanation does not work. When I try to explain, the response I get is "I have no time listening to you, just let me have all the emails". Well, obviously if I did let him have all, he would be bombarded with virus and spam, and I would be the one who would have to clean the mess. So I stood my ground, then he fired emails to all senior management members etc.

I ran into this with some of my customers. Effectively what it boils down to is making the webinterface as simple as possible.

We ended up writing our website to do hourly and daily notifications from 6am to 6pm mon-fri

The email that is generated has a link the user clicks and it logs them in automatically. no username or passwords to remember. We then utilize the autowhite list function and eventually it works it self out.

Our code works so well that you can actually create a link right in outlook so they can just click it just like any other mail folder and they are in the spam interface to check for spam. It literally take zero effort.

I may consider selling my code if anyone is interested.

Yeah, I have that too, kspare.

Selling? I have been giving them out for free.

Maybe I'll consider every two hours notification rather than daily.

you might offer to loosen the reins slightly by using the Custom Domain Matrix.  Explain what each type of protection is available and let them choose.  We have a couple of customers that don't want Bayesian filtering.  However, we continue to block SPF, bad files, missing reverse DNS, missing FROM, blacklist, etc.

regarding notification; we allow the customer to choose the frequency.  below is a screenshot of the settings our customers can make per account/login:

Our clients are initially in "filter shock" when they first use our filter.  Our filters are considered "agressive", but after a little tweaking with the white lists, such as finding out who their important vendors are (i.e. those with their own domain names), we will add those domains to our white lists.  Also the auto-white list on force deliveries has really helped in the web interface.  Some of our clients were getting over 200 spams per day per mailbox and once things have been tweaked for them, they are more happy with the email being filtered out than they are upset with having to deal with spam quarantine, etc.

They also said other companies do not have this quarantine thing and do not ask users to go check the spam themselves, and they have no problem with spam or virus, nor do they have problems with their customers.

you DO realise that you are beeing told a blatant lie, or are faced with ignorance of epic proportions right?

No, I don't realize that because they said this was from their IT manager from their old jobs, and they offered me to talk to the IT manager.

Marco wrote: They also said other companies do not have this quarantine thing and do not ask users to go check the spam themselves, and they have no problem with spam or virus, nor do they have problems with their customers.   you DO realise that you are beeing told a blatant lie, or are faced with ignorance of epic proportions right?

As long as there is money to be earned for spammers, spam will keep polluting the internet, it's as simple as that. Every major ISP, and allmost all companies are fighting spam, they get rid of 95% of the garbage for their users. But unfortunately a couple of legit mails get caught in the crossfire every now and then. Tell your users that what they want is impossible, no matter how good an it-manager is, things will never go 100% right.

One major step in the right direction has been made recently, with the SFDB-collective, i see this filter catch more and more spam, no false positives so far.

My boss doesn't want to sell the interface at this time.  Sorry guys.  I would be happy to answer some questions or give you tips as long as it is not directly supplying code.

SFDB Filter - SpamFilter Distributed Blacklist

The SFDB filter uses a very powerful resource to stop spam:
The entire global SpamFilter ISP user community.

Anytime an IP address is added to SpamFilter's local IP blacklist cache, the SFDB filter updates our Distributed Blacklist centralized database. This allows the SFDB filter to have access to a huge repository of spammer's IPs, updated in realtime by all the SpamFilter ISP users in the world. IP addresses from the database are automatically aged and removed from the database within 24 hours if they receive no further reports.

The SFDB filter detects spam by checking IP addresses against the SFDB database. The "network reliability" level tells SpamFilter how many different users must have reported a specific IP in order to classify it as spam.

The SFDB filter will reject IPs if there are *currently* in the database more that a minimum number of reports for a certain IP. This threshold is the "Network Reliability". Enter 0 to disable the SFDB filter.

Rifleman wrote: Roberto,   Please explain how SFDB works exactly please.   Thanks

For you guys who are interested in the website, please see:

http://www.logsat.com/spamfilter/forums/forum_posts.asp?TID=/5338&PN=1 - http://www.logsat.com/spamfilter/forums/forum_posts.asp?TID= /5338&PN=1