I’m learning procmail. Every time a piece of spam gets through my filter, I write a new recipe to filter it, and reply to the spam, figuring they’ll send me more mail within a few days, and I can find out if the new recipe works. I got a message today from (random string of letters)@msn.com, with a reply-to of (believable username with random string of letters appended)@yahoo.com. I replied to the reply-to, and it bounced. I replied to the from: address, and it bounced. There’s no “click on this link” or “email (some other address)” anywhere in the mail - just “fill out this form and send it back to us” (no snail mail address, either). It looks like an online version of vandalism to me - i.e. it serves no purpose, wastes time, and is annoying. I can’t see, though, why anyone would do this. It’s neither highly visible nor lucrative. Can anyone suggest what the purpose of this mail might be?
Hm. Just realized how stupid that subject is. If a mod wants to be superkeen and get rid of the “what’s the point” part, it would make me smile. If not, life will go on, I’m sure.
[note: Smile, mna! Rather than get rid of it, I made it seem like there’s a question here (which there is), as opposed to a rant. -manhattan]
[Edited by manhattan on 03-16-2001 at 06:32 PM]
My WAG: This is a way to see which of the email addresses in the spammer’s database are actually valid, before sending out the real spam. (The invalid addresses would bounce.) I have no idea why you would want to test out the addresses first, because it seems it would require just as much in the way of resources to send a test message as it would to just send the spam. But that’s the only thing I can come up with.
I like the tagging idea. One of my e-mail addresses is so obscure that it could only be randomly found by an address tagging program. Yet I get plenty of spam.
Another silly possibility. Could it be that the spammer got shut down before you could reply?
Ok, this would make sense, if the spammer were in the business of selling email addresses to other spammers. In which case they wouldn’t have a “legit” message to send out, but would want to have the highest ratio of good to bad addresses. Thanks, MsWhatsit, that’s an entirely plausible idea.
SofaKing: I considered that, but it seems unlikely. There were less than 2 minutes between receipt of the message and reply.
The problem with spam is that it costs virtually nothing to send so there is no incentive to attempt to direct it towards people who are more likely to respond. It just isn’t worth it to cull your list and a ridiculously low response rate validates the effort.
From a practical point of view what I do is keep a couple of hotmail accounts. I never give my primary emails unless I want someone to have it. I have one hotmail for things that might marginally have some interest and one for virtually everything else.
I have had the same email address for over 10 years and I get almost no spam at all.
Common reasons-
[list=1]
[li]They are trolling for valid email addresses to sell, as MsWhatsit and typo mna suggested. They don’t actually have any spam for you, but they’re going to sell your address to spammers, with a guarantee of a certain percentage of valid addresses in the list.[/li]
[li]The mail included html or java code that put cookies on your system, or automatically caused your machine to send something back to their server. If you’re email client is set to interpret html or java, just the act of reading the mail can send information to them. No actual response from you is required.[/li]
[li]Hi Opal![/li]
Arjuna34
A good reason not to use an email client that can interpret HTML or Java, I might add.
Yes, one of many good reasons to turn off HTML, Java, VBScript, etc. in your mail client. If you do leave them on, at least disable the preview function, so you don’t accidently run the code without even opening the mail.
Argh, I used “you’re” instead of “your” in my last post- one of my pet peeves, even! 
Arjuna34
There was no naughty code in there. I use pine (which is why I’m learning procmail - pretty much every graphical email client I’ve seen has a built-in filtering feature, which I’d use, if I used that sort of client), which, of course, is text-based, so HTML, improperly attached files, and all that sort of stuff shows up as a bunch of garbage. This email was all text.
Speaking of email address harvesting, Sneakemail is a great way to avoid it. Every time you need to fill out a form with your email address, you can go there and generate a new one (e.g., 47kij1klm8@sneakemail.com).
Then when you receive mail at that address, Sneakemail forwards it back to your real address, after changing the headers so that 1) you know which address it was sent to (using a label you supply when you generate the address), and 2) your reply will go through Sneakemail, so the sender doesn’t see your real address.
If you start receiving spam at an address, you know who gave it away. You can also set up detailed per-address rules to reject certain senders, or only allow certain senders - or you can close the address entirely.
How do you know if your e-mial client can interpret HTML or Java?
If you can see pictures or animations or whatnot in the body of your email, or there are links you can click on, you’re HTML-enabled. If HTML code shows up in the form of a bunch of “tags” in brackets, your client isn’t interpreting it.