There is no doubt at all that e-mail is the most popular and at least one of the most useful Internet tools. E-mail was originally a way for different accounts on the same computer to communicate with each other, sort of a delayed reaction chat function. The reason e-mail became popular so quickly is because of its relative immediacy and because it was really so simple.

On early UNIX machines, the architecture of which greatly influenced the development of the structure of the Internet on a number of different levels, sending e-mail was really as simple as "mail [user name]." This would bring up a terminal based e-mail client and allow the user to type a message. Closing the e-mail editor would send the message, which would arrive pretty much instantly in the other user's account. When they logged on next, the mail system would tell them they had so many messages available and they could open the mail program to read them.

What the original networked e-mail system allowed was for users to send mail to accounts on other machines. So if within a network one machine was called "dog" and another machine was called "cat" a user on the "dog" machine could send e-mail to a user on the "cat" machine by addressing it "user@cat." This will probably be recognizable to users of e-mail as it is the basic structure of an e-mail address. The only difference between this early UNIX-based e-mail and the current system is the way machines are named. In order to route e-mail properly, it is necessary for each machine to have a unique name, and this is why the domain naming system or DNS was incorporated in the structure of the Internet.

What all of the early e-mail systems had in common was that they were a text-based communications system. There were really no "graphics" or file attachments possible without some extraordinarily arcane configuration which would ultimately be redundant with technologies like the file transfer protocol or FTP which was developed precisely to make it possible to move large files from one machine to the next. There was also no such thing as "rich" text, document-based or HTML-based e-mail either. E-mail started as plain old ASCII text.

With faster network speeds, plain old ASCII text was a very fast way to communicate. Since each letter is just one byte of data, even e-mails with two or three pages of information didn't put much of a strain on the network because each e-mail only represented perhaps a few hundred to a couple thousand bytes of data.

But advances in e-mail clients have brought with them all kinds of advanced ways to generate what are essentially documents as opposed to messages, and with that comes the potential to slow e-mail down significantly. For example, an HTML formatted document can contain three times as much formatting data as actual text, which increases the size of the message significantly. Granted, this allows for convenient features like hyperlinks, lists and formatted text, but it requires a browser to display and it requires more network bandwidth to download.

For one message, this may not seem like a big deal, but as we all know, e-mail is never about just one message. When there's a list of 50 or 100 messages to download, turning every e-mail from 2k of data into 8k or 10k of data in order to format lists and hyperlinks slows an e-mail client down because it takes longer to download all of the messages. Then, add 50k of graphics to every third message and pretty soon what used to be a 200 kilobyte download becomes a 3.5 megabyte download, or almost 18 times as much data!

Then there is the question of actually displaying all of this non-text information. Believe it or not, it does take processor cycles, system memory and video memory to display HTML rendered text, documents, attachments and graphics in e-mail messages. For example, an HTML message requires the e-mail client to load a browser engine in order to display the formatted text inside the e-mail client. A message with an included graphics file requires the e-mail client to set aside system memory to load that file and display it inside the e-mail client. By itself, one e-mail message is unlikely to be a problem, but most current computer systems very rarely have just an e-mail client running which is displaying just one e-mail message.

So, how can e-mail be speeded up? One of the first things to do is allow e-mail to return to its origin. E-mail is a text-based communications technology. It is not a video phone, it isn't the web and it really isn't a file transfer mechanism although most people really do need to exchange documents in e-mail if they don't have access to an FTP server. Many e-mail clients allow a user to turn off things like HTML, graphics or document features in the message display. While this won't change the size of the message, it will reduce the load on the e-mail client.

The second strategy is to compose e-mails using plain text. By doing this, you are doing the recipients of your messages a favor by not requiring them to download 5MB of web page in order to read the two sentences of actual e-mail message.