4 ms·

I never understood why email servers never adopted something like the 3-way-handshake, which ultimately would increase the bandwidth and processing power requir

by jemka 16y ago

I never understood why email servers never adopted something like the 3-way-handshake, which ultimately would increase the bandwidth and processing power required per email. It would theoretically make bot-nets easier to find and protect against. However, I admittedly don't know much about the industry, so there's probably a simple answer.

kooshball 16y ago
I dont really follow what you're saying. Why do you think a 3 way handshake would make any difference here? Why would it make bot-nets easier to find?
- jemka 16y ago
  Spam is a numbers game. If you make it harder for spam messages to be sent out, spammers 1. send less spam by default 2. increase hardware. (realize 2 can also include increase cost if not using a "free" bot-net) If all SMTP servers were required to receive and send back a token/packet before their original message was delivered, the bandwidth/cpu etc... required to send the message will have increased. Send original to recipient server, receive from recipient server, send to recipient server. Then the mail is delivered. Example: A single spammer can send out 50 million emails / month on current system. Introduce a 3-way & that number drops. If the spammer wants to get back to sending 50 million / month, a significant increase in hardware will be required.
  - silentbicycle 16y ago
    In my experience, spamd (http://www.openbsd.org/cgi-bin/man.cgi?query=spamd http://www.openbsd.org/cgi-bin/man.cgi?query=spamd) is remarkably effective, and its greylisting works by requiring a request/response before the first valid transmission.
kjksf 16y ago
Short explanation: network effects and backwards compatibility. When the SMTP protocol was designed/evolved, spam and other malicious uses did not exist. They didn't even exist for many, many years afterwards. When the spam really became a problem, everyone was already using smtp and switching the whole world to an alternative wasn't (and still isn't) possible. Designing email protocol that would eliminate spam is not hard if you start with such goal in mind. The people who designed smtp were very smart (it is being used decades later), but they were pioneers and unsurprisingly didn't anticipate all of the consequences. When they were designing it, there number of people using the system were in thousands. I doubt anyone in early eighties was anticipating the magnitude of smtp's success and the abuses of the system only became profitable when email started being used by millions.

by jemka 16y ago

kooshball 16y ago

I dont really follow what you're saying. Why do you think a 3 way handshake would make any difference here? Why would it make bot-nets easier to find?

jemka 16y ago
Spam is a numbers game. If you make it harder for spam messages to be sent out, spammers 1. send less spam by default 2. increase hardware. (realize 2 can also include increase cost if not using a "free" bot-net) If all SMTP servers were required to receive and send back a token/packet before their original message was delivered, the bandwidth/cpu etc... required to send the message will have increased. Send original to recipient server, receive from recipient server, send to recipient server. Then the mail is delivered. Example: A single spammer can send out 50 million emails / month on current system. Introduce a 3-way & that number drops. If the spammer wants to get back to sending 50 million / month, a significant increase in hardware will be required.
- silentbicycle 16y ago
  In my experience, spamd (http://www.openbsd.org/cgi-bin/man.cgi?query=spamd http://www.openbsd.org/cgi-bin/man.cgi?query=spamd) is remarkably effective, and its greylisting works by requiring a request/response before the first valid transmission.

kjksf 16y ago

Short explanation: network effects and backwards compatibility. When the SMTP protocol was designed/evolved, spam and other malicious uses did not exist. They didn't even exist for many, many years afterwards. When the spam really became a problem, everyone was already using smtp and switching the whole world to an alternative wasn't (and still isn't) possible. Designing email protocol that would eliminate spam is not hard if you start with such goal in mind. The people who designed smtp were very smart (it is being used decades later), but they were pioneers and unsurprisingly didn't anticipate all of the consequences. When they were designing it, there number of people using the system were in thousands. I doubt anyone in early eighties was anticipating the magnitude of smtp's success and the abuses of the system only became profitable when email started being used by millions.