Hi Susan, all,
Thanks for kicking-off this important event.
Managing Spam is such an overwhelming task it’s no wonder we find it difficult, especially when every trick we employ seems to be countered and bested by Spammers in this never ending battle for control of our Inbox.
One of the most common social engineering tricks employed by Spammers is to counter the effectiveness of the [
Can Spam Act ] by making people afraid to click an ‘Opt-Out’ link for fear of receiving even more Spam. In the US it is now a legal requirement for all Spam to carry a working ‘Opt-Out’ feature, so the latest scam is to make people afraid to click these links – thereby making the Act effectively worthless even though most Opt-Out links today are genuine (this wasn’t true prior to the Act). Marketing has always been a mind game and these new techniques catch a lot of people unawares.
Part of my role is to administer our national mail gateways and even just from a cost perspective the statistics are truly frightening. On average we process 135,000 mail messages per month of which 65,000 are Spam (we receive more than 2,000 Spam messages daily). Our costing for Email infrastructure and management are $0.15 per message received, so Spam costs us on average $9,000 per month. Add to this the cost of computer virus management (about $4,000 per month) and it’s easy to see why many businesses now lock Email systems to the extent of blocking all uninvited Email – including that sent in good faith by charities and many NPO’s. It is simply too expensive to receive and try to segment ‘good’ Spam from ‘bad’ Spam even though this tremendously disadvantages charities and NPO’s trying to send legitimate messages to potential donors or beneficiaries.
Of the management tools available – we use a variety however the most effective I have found is the [
SURBL ] method of identifying Spam by URL’s (or parts thereof) contained within a message body. This works on the premise that all Spam has one thing in common – the recipient must be able to contact *someone* in order to complete a financial transaction. Sending IP’s can be spoofed; mail headers forged; key words obfuscated to bypass content word and phrase scanners. However at the end of the day there must be some legitimate reference in a Spam message so recipients can contact the individual or company marketing the product. Block these references and you block 99% of Spam with very few false positives (we average less than one false positive out of the 2,000 Spam messages caught daily).
In practice we use SURBL in the same manner as any other RBL block-list. Our mail gateway scans message bodies for references matching the SURBL lists and rejects or quarantines any mail with a positive match. We also maintain a local URL-RBL list to add any false-negatives not caught by the public lists. One very good implementation of SURBL support suitable for gateways that supports command-line tools is [
ASTPS ] – a free add-on for many mail gateways. An obvious benefit of this method is that it does not block legitimate NPO mail through false positives generated by a content/context scan (unless the NPO is actually engaged in Spamming!).
Rgds, Don