>Google’s spam filter does a TERRIFIC job of catching spam. But it put’s spam in a “Junk” folder. So, I ~still~ have to review, then delete, spam. Got to check for the rare instance when good mail is misidentified as spam.

gmail auto-deletes everything in spam folder when it gets older than 30 days.

But still have to look through it, it is good as catching spam but quite a few false positives as well. I used gmail to consolidate all of my email and overall quite happy with it.

Plus, you have to join before you can even attempt posting. THAT
cuts down on a lot of spam, as spammers don’t usually want to take
the time to fill out on-line forms to join.

My personal e-mail is on two accounts: Gmail, and my ISP.

Google’s spam filter does a TERRIFIC job of catching spam. But it
put’s spam in a “Junk” folder. So, I ~still~ have to review, then
delete, spam. Got to check for the rare instance when good mail is
misidentified as spam.

Use Gmail for my ‘public’ e-mail address. This really minimizes the
amount of spam received on my ISP account address. I reserve that
for friends.

But -in all cases- it’s (as you say) critical never to do anything
with spam except delete it. NEVER open it. Respond to it. And
certainly never click on anything in it !

And, I never open attachments or click on links that I’m not sure
were actually sent by a “friend”. Even if from a friend, I won’t
click on it if I wasn’t expecting it. Sometimes this means e-mailing
back first, asking if they actually sent the link or attachment.

Only time I ever downloaded an e-mail virus was a few years ago in
an attachment sent by a dear friend. The virus on her PC sent itself
out to everybody on her address list !

The DataRat

.

Well, it totally practical to me. :) And using only list is just an example.

I don’t like Akismet much and consider it overhyped because of bundling with WordPress. I hadn’t used it much as blogger but as reader I lost count of times my totally normal comments where treated as spam.

Black list is essentially a set of powerful custom rules. Marking message as spam using external service like Akismet tells service that something about that message as spammy. Then service may or may not deduce signs correctly and in either case will try yo apply result (you have no idea about btw) to following comments.

By maintaining black list guessing part is eliminated and create very specific and precise rules. That are very not probable to affect any legitimate comments.

If (when?) spam gets trickier to manage I will probably add bayesian filtering to the mix, it’s rather effective and unlike black list is self-learning.