This sounds like an interesting idea

[Tom]

Deal with spam

I'm interested to know how people would do this. For instance, say a friend sends you a legitimate message, for the first time ever, and he has an old email client. Would he have to upgrade it, or would you use discretion in choosing to receive the mail? Also, wouldn't it be a server-side upgrade at your ISP, rather than a client upgrade?

 

[Shovel]

Very interesting read. It's a unique concept, and certainly worthy of more thought - it would be very easy to laugh it off.

Two key points that aren't reported:
1) How do you prevent a spammer cracking this code so as not to do the processing? Maybe it would be based around the email address of the recipient?
2) I don't understand how this works - surely if a spammer just doesn't update their email system then they wont have the means to process the puzzle, and therefore will output as much as before. It seems to suggest that only people using the new technology would benefit, and at the extreme they have to block everyone who doesn't complete the puzzle: This then means, surely, that anyone stuck on old technology gets indentified as a spammer in the same way? If this system is depending on people updating at the client end, then it's dead in the water: Microsoft know all too well how hard it is to get people patched against system critical virii, let alone spam.

 

[TheJkWhoSaysNi]

It won't work. Spammers will just have their own client to send mail, what about php/perl scripts that send mail? I dont think server admins would want to spend 20 seconds of processing time sending one email.

 

[fatbusinessman]

1) How do you prevent a spammer cracking this code so as not to do the processing? Maybe it would be based around the email address of the recipient?

Presumably it would be something along the lines of "factorise this huge number, take the larger of the factors and put it in a special 'passkey' field in your e-mail". Therefore you need to expend a certain amount of resources to determine the passkey. Presumably the problem would also have to be time-dependent in some way, so that spammers couldn't simply buy lists of addresses and their corresponding passkeys.

 

[Shovel]

If you factored in the send time, so that the key was unique to each email, you'd be covered against buying passkeys?

 

[xane]

The next method for killing spam will be the "confirmation" mail, basically, you receive an e-mail, then a request is sent back to the sender before you confirm the mail and open it.

As most spam is from unreachable addresses it will not be confirmed and opened.

Spam from valid addresses have the sender destroyed, utterly.

 

[Tom]

But the main issue will remain if Xane's method is used - bandwidth. We're told that 50% of emails are spam, and that can't go unnoticed in terms of performance, and costs.

 

[fatbusinessman]

If you factored in the send time, so that the key was unique to each email, you'd be covered against buying passkeys?

You might be better off factoring in the day it's sent instead, as that would minimise inconvenience to genuine e-mailers.

 

[Jonty]

The next method for killing spam will be the "confirmation" mail, basically, you receive an e-mail, then a request is sent back to the sender before you confirm the mail and open it.

As most spam is from unreachable addresses it will not be confirmed and opened.

Spam from valid addresses have the sender destroyed, utterly.

Hello

Not much help, but if anyone is interested in xane's idea, it's in operation at KnowSpam (amongst other places).

Kind Regards

Edit ~ For what it's worth, I've recently written a brief commentary on unsolicited email for my Information Technology Law module at university. Although you wouldn't know it, unsolicited email is actually prohibited in many circumstances by various dispirate legislation, from EU Directives to the US's soon to be enacted CAN-SPAM (Controlling the Assault of Non-Solicited Pornography and Marketing) Act. However, there are loopholes throughout all this legislation, and its disparate nature and lack of cooperation with the IT industry probably means it'll have little impact on the average user

 

[xane]

I'm betting the confirmation e-mail system will be embedded into Hotmail, Outlook and Outlook Express, unilterally by Microsoft, at some point next year, then all the others will follow suit, as they always do.

I disagree with the "bandwidth" argument, most bandwidth is wasted downloading the spam from your post office into your inbox or local environment, the confirmation mail will eliminate this and the confirmation message will be very small, considering the sheer weight of spam nowdays it will make it worthwhile.

A more advanced system would be used by the mail provider to verify whole swathes of mail using a whitelist, and prevent them from coming into your post office in the first place.

It might even make the spammers reconsider

Remember the key issue of spam is that it is (a) anonymous, and (b) cannot be replied to. These are the main weapons of spammers that allow them to ply their trade so actively, once you remove that weapon they either perish or diminish.

 

[Ch3tan]

The only spam I now recieve is from freeloader.com in associationw tih any cpmany under the sun. It's very annoying and I can't seem to find a way to stop it. unsbscribing from it did work once, but it just started again after a month.

Oh well, it will all disappear with the loss of my BW mail account.