E-mail was one of the first services available on the internet. At that time messages were text only and it has been an amazing possibility to send them across the globe in just a few seconds. It was much more convenient than telegrams, usual letters are not even worth mentioning. At the same time it became clear that simple text messages were not enough. And so message format has been improved several times allowing the user to decorate the text using different fonts, size, etc, to paste pictures into the message body, attach files, make use of the message encryption and digital signature.

It all became possible thanks to the development of the e-mail clients. Primitive text editors with the mere ability to send and receive messages evolved into powerful programmes with lots of handy features. Now it's possible to create colourful HTML-messages, use templates to speed up the work process, automatically sort incoming messages into different folders and make use of the feature-rich address books. But what happens when we press the "Send" or "Receive" button in the e-mail client? And here the mail server steps into play.

An e-mail client knows nothing of how to deliver a message to the recipient. Its main goal is to correctly compose a message, put in the right sender's and recipient's e-mail addresses, and then hand over that message to the mail server. Incoming messages an e-mail client also receives from a mail server. That's on one hand, on the other one is the mail server, which does not care at all how a message is laid out, but it knows how to deliver it to the respective recipient. In a sense it resembles a usual post-office and people coming in. A post-office stands for the mail server and people represent e-mail clients.

Naturally you would ask why do we need a mail server at all? Can't the e-mail clients send a message directly to the recipient? After all, internet, unlike usual mail, doesn't have any frontiers and transmitting data to another part of the world is as easy as though it were the room next door. So, just imagine you want to send a message and the recipient has already gone to bed or simply shut down the computer. Or someone sent a message, but at that time you were on vacation. It's hard to imagine that the sender will keep on waiting to send you the message until you are back. It's much more convenient to transmit a message to the mail server and it will make sure the message is delivered. Moreover, the message will not be delivered directly to the recipient, but to the respective mail server that will put the e-mail into the user's inbox folder, where it will reside until the user checks his/hers incoming mail.

But how can one know to which server should the message be transmitted to so the recipient finally gets it? The answer to that question lies on the recipient's e-mail address. It consists of two parts split by the "@" sign. What you have to the right of that sign is called domain name and what you have to the left of that sign is the user name within that domain. In order to deliver a message to its recipient it is necessary to determine which server is responsible for that domain's mail. That is why the so called DNS (Domain Name System) is used, it keeps records of each mail domain and the list of servers attached  to them.