Strategies for dealing with spam?
Strategies for dealing with spam?
Author
Discussion

TUS 373

Original Poster:

5,211 posts

311 months

Wednesday 3rd May 2006
quotequote all
For some time now, I am receiving emails with bogus names and random titles all concerned with 'insider dealing' share tips. In the language they use they also appear to be American. Just recently I have started to get more and more of these, often in duplicate.

The filter in Outlook seems to cope well with diverting the vast majority of these straight into the Junk Mail folder.

What I am wanting to know is:
1. Ss there a way of tracking down the ultimate source of this garbage and blocking it?? It is definitely spam, not a service I have signed up to.


2. Would it help if I gave that particular email address a rest for 4-8 weeks and then reinstated it, or does spam go on getting sent forever?

_dobbo_

14,619 posts

278 months

Wednesday 3rd May 2006
quotequote all
Keeping things simple, Anti-SPAM tech falls into three different areas.

1.) Content checking - this is where an email containing (for instance) the word viagra is blocked. It can get a lot more clever with bayesian and weighting and scoring etc, but it's basically all content checking. This can be done server or client side.

2.) Identity checking - this is where you block a sender based on their IP address. For most people this is not something you can maintain, because Spammers change their IP more often than we change our socks. That said, services called "Realtime Black Lists" are available, which are a publically accessible list of known spammers, relays, open proxies, zombie hosts, etc. Utilising RBLs can result in a significant reduction in junk. It will however require that whoever is the admin of you mail server (eg your ISP, or your tech dept at work) will need to implement this - I'm not aware of any client RBL plugins.

3.) Pattern recognition - this is more recent and is designed to get around the fact that spammers are increasingly bypassing content checks by inserting random text and HTML into their messages to "trick" the content recognition. Pattern matching is resilient to this because it doesn't actually look at the content of the email in the same way. I think people like message labs do this. Again, this is probably something your ISP would have to do server side.

There are lots of other things out there to do with address books etc. but these are usually specific to individual clients, and my experience of any filters that work based on email address is that they aren't that effective, due to the sheer quantity of forged email addresses being used. If you don't have control of your server, or if your ISP can't implement these technologies for you, then you'll need to start looking for client side utils to add on to your existing outlook AS filter. I have zero experience of this I'm afraid, as I'm firmly in the "fix it server side" camp.


edit - just realised I spouted all that crap without answering your question!

1.) It's not feasible for you to do this - see 2.) above.

2.) It's doubtful not using the address will help. I manage a bunch of mail servers and addresses that haven't been active for YEARS still get inundated with junk mail attempts. Once you are on the radar, you wont get off short of changing your address altogether, which again will probably only work for a few months.



>> Edited by _dobbo_ on Wednesday 3rd May 10:00