Editor's note: Basically, this discussion is only here for historical record. Our SpamAssassin, greylisting and blocklisting policies mean GreenNet users now have very little problem with spam. For information correct as at 2009, see:
I am just receiving too many Spams, Porno sites, people wanting account details while offering to share millions of dollars and people trying to sell, all of these unsolicited and obviously able to come through the GreenNet portal. The rubbish flooding into my mailbox is just too much for me to cope with. What does GreenNet do against spam?
GreenNet provides a unique anti-spam service geared to NGOs, the voluntary sector, charities and activists.
Comments
Re: I get so much SPAM
An interesting article about spam, where it comes from and how it is done has been published on the Guardian's web site.
[url]http://www.guardian.co.uk/uk_news/story/0,3604,976332,00.html[/url]
Re: I get so much SPAM
Can I suggest a site that I've just started using? It's SPAMcop, and you can find it at:
http://www.spamcop.net/
For $30 (about £20) per year, it cleans up your e-mail without bouncing anything and without your having to download it to your computer. I just started using it and today is a great example of why I think it's so good.
When I logged in, 10 messages were downloaded to my computer.
Of those, just 3 were spam. I reported each of these using the spamcop systesm.
When I logged into the spamcop webmail system, 112 messages were being "held" for me. These had been identified as spam and so not put in my inbox. I checked all of them (looked at the name/subject) before deleting. Not a single one was real mail.
The great thing about this is that you don't end up spending ages downloading spam onto your computer! You also don't have to worry about configuring message filters etc.. All you have to do is use spamcop to pick up your messages from your greennet account, and then get your computer to pick up from spamcop.
One tiny disadvantage is that, if you're on a network and are constantly checking for e-mail, you might find a slight delay (10-15minutes) in getting mail compared to what you're used to.
Anyway, I'm a convert.
Re: I get so much SPAM
A reminder that there are some instructions for filtering spam out using the (spamassasin) headers for Eudora and (now) Pegasus Mail at
[url]http://www.gn.apc.org/support/filter/[/url]
Cedric
Re: I get so much SPAM
I expect SpamAssasin will catch up with them eventually, but the current trick by the Spammers is to B.REAK UP W.ORDS so a robot reader can no longer recognise them.
I see Mr Mobuto's son is apparently alive and well and living in Nigeria :)
Are all messages via GreenNet run through SpamAssasin? Some seem not to contain the X-Spam-Status header - presumably they have not been assessed?
A recent message contained the header:
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
followed by four more bits of information about the message, which originated from yahoo.com. Would this be one of ours or one of theirs?
(The message was unsolicited and asked for an exchange of links but the sites they provided links to could not be found.)
Re: I get so much SPAM
SpamAssassin does process everything, but if the total spam score is very small, it doesn't add the X-Spam-* headers. (update, 2005: it should almost always now include the X-Spam-Status header)
X-AntiAbuse and similar are added by some providers of webmail, news servers, etc, to give instructions on how to report abuse of their services. They aren't added by GreenNet.
Re: I get so much SPAM
An additional protection that may be complementary to GN's SpamAssassin is something like MailWasher. This and similar programs preview your mailboxes, do some tests of a similar kind to, but much more primitive than, SpamAssassin, and list suspected spam for deletion. It also checks against SpamCop's extensive 'dirtylist'. [url]http://www.mailwasher.net[/url]
It can therefore be a time-saver, particularly if you don't want to set up a filter on your email program itself, either because you have a slow connection or Outlook Express.
MailWasher also offers to send fake bounce messages to the return-path addresses. This is of limited usefulness since most addresses are fakes, but can be helpful where you suspect a 'spambot' (automated address verification).
(Update, 2005: Mailwasher is now basically redundant on GreenNet accounts. It is recommended that you instead look at the [url=http://www.gn.apc.org/support/filter/index.html]'Spam filter' section of our support pages[/url] if you find some spam is still getting through the first-line barriers.)
Post Edited (10-24-03 11:25)
Re: I get so much SPAM
We've updated SpamAssassin and made it distinctly more sensitive today, so GreenNet members will see an improvement in detection accuracy.
Re: I get so much SPAM
We've now implemented a greylisting system which cuts out >95% of spam, in addition to the possibility of filtering on commercial content mentioned above.
You may notice some delays in incoming mail, usually around 30 minutes - this time is under the control of the connecting relay, not GreenNet. If you find problems with email forwarded from another address to your GreenNet account, please ask the ISP handling that address to check that their servers obey the two internet standards RFC 821 & 822.
If you want to someone to get something through to you urgently, ask them to send it twice about 10 minutes apart. Similarly, if you are e.g. waiting for a mailing list subscription confirmation, it may be worth sending twice. The second message will come through immediately, but the first may be delayed.
On balance, greylisting seems to be working very well.
Re: I get so much SPAM
As I asked in another thread but got no response. I want to know if Greennet has a system in place for training SpanAssassin. I use it for my personal account as well and know tat it needs constant training to keep the Baynesian filter up to date and catching the most SPAM while also training it to recognize ham and miscategorized messages.
How can we make sure greennet's spamassassin implementation is training properly?
Cheers,
Peter
Bayesian filter
Sorry for overlooking your post for so long. Basically, we don’t rely on the Bayesian filter. URL lists and the rules we have written – specifically with the needs of civil society groups and activists in mind – are more important. However, you can see how the Bayesian filter is doing by checking the X-Spam-Status header added by SpamAssassin. Non-spam (vege-ham) should get BAYES_00 or BAYES_20, while the “spammiest” messages get BAYES_99.
If you want to contribute to the training of the communal filter, which is most useful for unrecognised or conflicted (BAYES_50) messages, you can forward spam and nonspam messages to fairly obvious email addresses that can give you. These need to be sent from a GreenNet account and as an attachment to be accepted.