How to avoid spam?

[Zac67]

It's starting to get unreadable for humans, too. ;-)

Another good way is to make a picture with the letters in it. Use different fonts and capitalization, so it'll be too much work to harvest. If you save it as as JPEG with high compression (low quality option) it's nearly impossible to OCR.  :lol:

IMHO it shouldn't be neccessary to filter for spam (chance of false positives and negatives) but instead block based on destination by the ISP. Our mail server uses various external SPAMBL, it's own black list (IP and host name pattern) and dial up detection (via host name) with extremely high success. With numerous 'burned' addresses we only get ~1-2% spam, and that only because the server's pattern matching capabilities are not (yet) good enough.
False positives (rare) are denied right away and the sender gets notified in nearly no time. Since we don't bounce, there are no overflowing outbound queues with invalid senders.  :-)

[Piru]

@Zac67

IMHO it shouldn't be neccessary to filter for spam (chance of false positives and negatives) but instead block based on destination by the ISP. Our mail server uses various external SPAMBL, it's own black list (IP and host name pattern) and dial up detection (via host name) with extremely high success.

SpamAssassin has this aswell, btw.

Small example of SA content analysis:

Code: [Select]


Content analysis details:   (27.8 points, 5.0 required)
 
 pts rule name              description
---- ---------------------- --------------------------------------------------
 2.0 TRADING_ALERT_5        BODY: Talks about shares
 0.1 HTML_50_60             BODY: Message is 50% to 60% HTML
 0.0 HTML_MESSAGE           BODY: HTML included in message
 3.5 BAYES_99               BODY: Bayesian spam probability is 99 to 100%
                            [score: 1.0000]
 1.5 RAZOR2_CF_RANGE_E8_51_100 Razor2 gives engine 8 confidence level above 50%
                            [cf: 100]
 0.5 RAZOR2_CHECK           Listed in Razor2 (http://razor.sf.net/)
 0.5 RAZOR2_CF_RANGE_51_100 Razor2 gives confidence level above 50%
                            [cf: 100]
 0.2 DNS_FROM_RFC_ABUSE     RBL: Envelope sender in abuse.rfc-ignorant.org
 1.6 RCVD_IN_BL_SPAMCOP_NET RBL: Received via a relay in bl.spamcop.net
               [Blocked - see <http://www.spamcop.net/bl.shtml?218.65.219.13>]
 1.7 DNS_FROM_RFC_POST      RBL: Envelope sender in postmaster.rfc-ignorant.org
 1.6 URIBL_SBL              Contains an URL listed in the SBL blocklist
                            [URIs: gloriadj.com]
 3.8 URIBL_AB_SURBL         Contains an URL listed in the AB SURBL blocklist
                            [URIs: gloriadj.com]
 4.1 URIBL_JP_SURBL         Contains an URL listed in the JP SURBL blocklist
                            [URIs: gloriadj.com]
 2.1 URIBL_WS_SURBL         Contains an URL listed in the WS SURBL blocklist
                            [URIs: gloriadj.com]
 4.5 URIBL_SC_SURBL         Contains an URL listed in the SC SURBL blocklist
                            [URIs: gloriadj.com]


[Seehund]

Typing "user@domain.tld" as "user domain tld" is good too, as it doesn't contain any characters that the harvesters are looking for, while it's still human-readable.

Typing it backwards (dlt.niamod@resu) works fine as well, but takes a bit more effort for the human reader to get right.

The method used on Slashdot is probably fine too. Insert a random word (but not "REMOVE", "ANTISPAM" or similar) in the address, for example "user@domSNAILain.com", and after that write "remove invertebrate animal to e-mail me".

If you've got your own website, then either a CGI form, and/or your address written on a picture is best IMO. I don't think address harvesters bother with OCR detection on images -- for a spammer, that CPU time is probably better spent on harvesting another thousand "unprotected" addresses from people who don't make it clear that they're uninterested in spam and who aren't as likely to report your abuse.

If you make websites for the general computer-illiterate audience, then they NEED a clickable link (or a form). No obfuscation will work, no matter how obvious you make it. People are more clueless than you can imagine. Then the HTML entity encoding mentioned in this thread is probably better than nothing.

[Dan]

I was luck with userdonttrytospamme@domain_net for about a year,then they got me from somebodys virusinfected adressbook.
It´s impossible to avoid spam forever. Spammers should burn :madashell: