I know what you are thinking now: 'come on, I use e-mail every days, what the hell do you want to tell me again?'. Well, we send or receive e-mail messages every days but maybe we don't know how e-mail works. 1969: ARPANET, the first experimental network was born. 1971: two years later, Ray Tomlinson of BBN invents email program to send messages across a distributed network. There were scarcely 23 hosts at that time...The e-mail system works like the traditional mail system (friendly named 'snail' mail) but it is much quicker. An e-mail message takes seconds to reach its destination. However you won't receive the answer within seconds! In fact you have to consider some things: the recipient's time zone (while you are sending the recipient is perhaps sleeping), when the recipient checks its mail (today? Tomorrow? Every weeks?), when the recipient answers (today? Tomorrow?) and so on. When you send an e-mail message, you have to specify the recipient. So you write the message (usually you use a 'client' to do this job, but you could simply write the message 'on fly' by means of telnet. So an e-mail client is not required) and then you send it. What's happen? You send your message to your server and it deliver the message to the recipient's server. To understand this mechanism we must have a look to e-mail addresses. An e-mail address is composed by two parts: the user's name and a domain name. These two parts are separated by the '@' symbol (you have to read it so: 'at'). Ok, let's suppose you send an e-mail to 'user@host.domain'. You send the message to *your* SMTP (Simple Mail Transfer Protocol) server. It takes 'host.domain' and asks to another server to translate it into the corresponding IP address (it asks to the DNS server, that's Domain Name Server which searches for the corresponding IP address). If the DNS server knows that IP address, then the SMTP server sends the message to the recipient's server. On the contrary the DNS server supplies the 'name' of another DNS server which should know that address (and so on, until the right IP address is known). An IP address identifies univocally an host in the net. Anything you do on the net, you need an IP address. You don't notice that, but a DNS server is almost always involved. Well, unless you already know the IP address of course (when you are surfing on the web for example, if you type the IP address instead of 'http://www.something.com' in your browser, you can avoid to call the DNS server. But I can't remember numbers such as: '277.122.0.22', and you?). When the SMTP server knows the recipient's IP address, it sends the message to the recipient's SMTP server. When the message reaches the recipient's SMTP server, that server looks at the first part of the e-mail address (the user's name) and it sends the message to an user's reserved area. This is true in a LAN environment, but in a dial-up environment, things are different. In fact SMTP can't be used for the 'final' transfer, since the destination machine is typically off-line, and only connected for short periods. So the SMTP server transfers the e-mail to a POP server (Post Office Protocol), where it is queued. When the recipient dials in with an Internet e-mail client, he uses the POP protocol to retrieve the e-mail. So we could say: you send e-mail messages by means of the SMTP protocol and receive them by means of the POP protocol.