Present at the 'e'-creation

It was sometime late in 1971, Tomlinson recalls, and the message, sent from one of his computers to another adjacent to it, was probably something like "QWERTYIOP."

In the first e-mail sent to other people's computers, Tomlinson explained to his colleagues at BBN Technologies--a Cambridge, Mass., research and development company now a subsidiary of Verizon Communications--how his new system worked. Tomlinson and his co-workers were assigned to find uses for the newly developed ARPAnet, the precursor to today's Internet. Tomlinson's fellow researchers took to e-mail with alacrity, and from that point, the electronic revolution in written communications unfolded.

E-mail found its way into the mainstream more than 20 years later with the advent of the personal computer and then the World Wide Web, which popularized the Internet. Tomlinson, who this summer won a Lifetime Achievement Webby Award from the International Academy of Digital Arts and Sciences, spoke to CNET News.com from his office at BBN Technologies, where he is a principal engineer.

Q: E-mail has become so central to the Internet it may be hard for some people to imagine a computer network without it. What exactly was the ARPAnet doing before e-mail?
A: It was doing a lot of file transfers and Telnet access so you could log in to a remote computer. There were some other data-transfer operations geared toward output for particular kinds of programs.

What made e-mail different from these types of data transfers?
At lowest level, it was pretty much the same thing. You had a target file going to the mailbox of a recipient. But the source, rather than being a file, was something you had composed in an editor. With e-mail there was an address to a person, and it was obviously intended for human consumption. Previous applications had existed within a single machine, and this one existed across a network.

E-mail came from working on the single-machine version of sending messages, the program called SNDMSG, and playing around with the file transfer protocol. And there was an RFC (request for comment) about sending messages (but it was never implemented and output to a printer). When I saw these things it occurred to me that you should first send the message to a person, and you could use the file transfer program to send the characters of the message from the SNDMSG program to the mailbox at the other end.

What were some of e-mail's earliest uses?
Pretty much the same thing you see today. People used it for what they needed to do--if they needed to arrange a meeting, they would try to negotiate a time, or ask a question about how a piece of software worked. It wasn't quite as versatile as it is today; you didn't have things like attachments and various kinds of media that were described by the MIME idea. That came quite a bit later than the early stuff, but it really codified the various kinds of media you could attach to e-mail and made it much richer.

When did it become apparent to you that you'd created something big, that e-mail was really taking off?
It was actually not until about seven years ago that someone asked me the question about what I'd really done back then, and in thinking about it, it occurred to me that I was the first one to send an e-mail. A lot of other things had happened to e-mail in the interim. Apart from being the first e-mail, it hadn't seemed so noteworthy. I think it is important to say that there were large numbers of people who contributed to the e-mail we know today. I got it started, but most of what we've seen today is the result of hundreds of people who have worked on it one way or another.

But when did you first realize that e-mail was becoming a major mainstream phenomenon?
There were indications very early on. Within about four or five months of writing the program, we were already discussing how to write mail into FTP. Larry Roberts, the head of the IPTO at DARPA, said, "If you want to do business with me, you have to send me e-mail so that I can have a copy of it and conduct business on my time frame." He actually wrote a program to read e-mail. Up until then, they typed their mailbox out. At the baud rates we were talking about then you could read things before they scrolled away from you.

What was Doug Engelbart's role in the invention of e-mail?
(Editor's note: Engelbart is a pioneer in numerous technologies, among them the mouse, display editing, windows and groupware.)
The online system he put together had a version of e-mail, though this was a single-computer thing. He also had a role in hypertext. His theme has always been getting people to be able to interact and communicate. And that was part of the online system that he was putting together at SRI. He had a lot of the early ideas there.

Can you describe in some technical detail exactly what you devised to send the first e-mail?
I had two programs. One was called SNDMSG, which was used for composing messages to be put in the mailbox of another user in a time-shared computer. There were versions of SNDMSG from Berkeley and MIT, but I recoded it. Another program was an experimental FTP program. I wrote a version to act as a server, another to act as a client, to specify what should be transferred, then send the data of the file to the other computer. I took those two programs and put them together with some glue software, the sticky stuff. Rather than getting the source from a file, you'd get the source from the buffer of the editor. And instead of simply writing the file at the remote end, you would append the characters to the mailbox file. The new message would follow the earlier message that would already be there.

And then there's a thing that everyone remembers, or associates with e-mail, which is the @ sign, which gave the editor a way to specify the recipient. You had to have a way of separating the user and the computer name. In English the @ sign is obvious, in other languages it isn't. But being the only preposition on the English keyboard, it just made sense.

What have you been up to in the 30 years since sending the first e-mail?
I've been up to a little bit of everything. I worked with packet radio in the late '70s. Network security, personal computing--I designed the hardware for a personal computer before you could buy a PC. I worked on the designs for a large-scale processor, something that was never built. One of the early ones used TCP to send fax data between these huge, expensive fax machines that we had in the 1970s. The idea was that the post office would do your faxing for you. Now that fax machines are a dime a dozen, the idea seems kind of crude and silly, but at the time it seemed like it might be a good idea.

What do you predict for the future of e-mail?
I think the main thing we're going to see is an increased use of higher bandwidth. The text-based stuff is just going to stay around because it fills a very significant need. But when bandwidth becomes super-cheap, putting your picture or other high-bandwidth media along with your message might just work. Also, we might find e-mail merged with other forms of access. We might see it merging toward the instant-messaging stuff--we may find that there are programs that integrate the two.

What's the best thing to have come out of e-mail?
The best thing is the ability for people to communicate with each other more conveniently--and more globally, too. One would not tend to pick up the telephone and talk to someone you had never heard of before if you heard they had information you wanted.

What's the worst thing?
It's the ability for people to communicate with each other more conveniently (laughs). It's a double-edged sword, obviously, because the convenience means it can be used for things like spam, which is an imposition for everyone who receives it.

What about on the security front?
We'll see more and more secure ways of doing e-mail. There are already ways, but they're not as widespread as they could be. There's PGP, but it's hard to get it started. It's a bit of a chicken-and-egg scenario--you have to know that the recipient is participating. And for 99 percent of things carried by e-mail, you don't really need high security. While the cost of capturing someone's e-mail is relatively low, the benefit is also pretty low. Who cares if you're going to have pasta for dinner?

What about e-mail-borne viruses?
If you take the biological metaphor, viruses are opportunistic. They will use routes of infection that are the easiest for them to use. Cold viruses use hand-to-hand transmission because that's a good way to get around. Viruses will use the best medium for getting from place to place, and e-mail happens to be the most prevalent for communications from computer to computer. So it's an ideal way for viruses to spread. If it weren't e-mail it would be something else.

How do you use e-mail? How has it changed your life?
I use it all the time. But I don't think it's really changed my life, because I've been using it for roughly half of it. It makes a lot of things more convenient.

What do you think has been lost with the decline of the letter?
People are influenced by the immediacy of e-mail. It's not so much that they can't do the niceties and etiquette of the letter, but that people tend to not do it. Someone once asked me, "What's your advice for writers of e-mail?" I went through the usual stuff, but I also have advice for the readers of e-mail. You have to fill in those things e-mail senders tend to leave out.

E-mail often makes the news, and it's often things like Bill Gates having his words read back to him in court, or a new virus. How much personal satisfaction or mortification do you experience when these sorts of events transpire?
Most of what I see looks good to me. Certainly my own experience with e-mail has always been positive. I receive very little negative e-mail and a lot of positive e-mail from people who have had their lives changed for the better because of it. I don't respond very much to what I see in the news about e-mail. If it weren't e-mail, it would be something else--recordings of telephone conversations, or letters in a file someplace.

With the 30th anniversary, people are comparing your invention to the printing press, the telegraph, the railroads--revolutionary advances in communications and industry. What do you think of those comparisons?
They're kind of intriguing, but I don't know how true they are. Time will tell. E-mail has been noticed to be popular only within the past 10 years or so because of the computer explosion. That's not enough time to see how significant it really is.