In May 1982 I joined Tandem Computers and became part of the second-level support team in Frankfurt am Main in Germany. At the time, Tandem had been running a proprietary global network, called EXPAND, for several years. I had seen it in operation three years earlier, but never got to use it.

All that changed when I joined Tandem. We did our business on the net.

In those days, nobody knew what the term “computer network” meant. A computer was one of those big things in hermetically sealed rooms with lots of white-coated people tending to them. The only connection with a network was the power grid that supplied electricity for the computers.

All our interfaces to the computer were text-only. Most of the text was line-oriented, but the first full-screen editors were just coming out. The transmission speed from the computer meant that it took a whole second to rewrite the screen at 19,200 bps, so many people continued to use line-oriented editors. This was pretty typical of the way most people used computers in those days, and we were no exception. For us, the difference was the network. It changed our lives and the way we did business.

In those days, the network comprised maybe 100 nodes in four continents: America, Asia, Australia and Europe. The connection between corporate headquarters in Cupertino CA and Europe was a ring, so any single link could fail and connectivity would remain. On one occasion the direct link between Frankfurt and London went down, and all connections went via Cupertino, about 15 times the distances. It was slow, but it worked.

A day in the office, 1982

Probably the best way to get a feeling for what things were like in those days is to consider the day's routine. When we came into to office in the morning, we'd log on to our machines—maybe, for those who were security-conscious enough to log off the night before—and start our email program. We were 9 hours ahead of Cupertino, so they basically started work when we stopped. By the morning we could expect a lot of mail from the USA.

But it wasn't all company mail. Private mail exchange was allowed and almost encouraged. Messages went out to all employees. Thanks to the employee stock purchase plan, we were all stockholders, and one of the daily messages was the stock price.

The email program was primitive. It was line-oriented, it required an external editor, and it didn't have a proper reply facility: all messages were new. It didn't really do filtering, and it was much slower to use than modern Unix email systems.

Email wasn't the only thing. Tandem was a pioneer in distributed databases, and even normal files could be distributed across network nodes. The entire network looked like one big file system. That's similar to NFS (Network File System) today, but it was automatic: file names included the name of the network node on which they were located, along with the disk on which they were located. This allowed anybody with the appropriate access credentials to automatically access a file anywhere in the network.

One use was for “reply files”. People might send out a question on email, and include a line at the end:

Useful replies will be kept in \SOSII.$WASTE.OF.TIME
      

\SOSII was our network node, and when we got an enormous disk (580 MB!), we didn't know what to do with all the space, so we called it $WASTE. There really was a file called \SOSII.$WASTE.OF.TIME, and anybody who was interested could read it.

But there was more than just file access. We had multiple databases. One example was the personnel database mentioned above. From this database others were created, like the TELE database with people's phone numbers, email addresses and some other details. This was available across the network.

Back to our work day: after handling email, the main part of our work was problem resolution. For that we had another database, part of the “Product Reporting System” or PRS, which maintained problem reports. The database was accessible world-wide, and as we processed problems, they could be reassigned to support personnel anywhere in the world.

Normally our work flow would take a TPR (“Tandem Product Report”) from the person who issued it at some field node, through to us, on the support headquarters in Cupertino and finally to Software or Hardware Development, also in Cupertino. But there were exceptions, for example with strange and unusual problems which required expertise that only certain people had, and on occasions we sent TPRs on to Reston VA, High Wycombe, UK or even Australia.

This was not a high-speed network by modern standards. The main backbones only ran at 56 kb/, still nearly 3 times the speed of our terminals. It was practical, but sometimes slow, to run interactive shells on systems in other continents, but we'd expect email to be delivered in less than an hour. Sometimes it was much faster, and on one occasion in 1983 I exchanged mail with somebody in Cupertino with delivery times of only a minute or two.

What didn't we have? Graphics of any kind. And of course there was nothing like the World-Wide Web. But if you don't know about it, you don't miss it. We loved our network.

It's interesting to think of what else was going on in the world at the time. ARPANET had been operational for nearly 7 years, but it was about the only one in “production”, and it had few non-US nodes. BITNET had only just started, The Internet was just about to start operations. 4.1cBSD, the first operating system that enabled the Internet, was released in late 1982, and production networks didn't come until even later. In many ways that started with NSFNET several years later. We didn't feel like pioneers, but looking back, we were to some extent.

There's much more to say, but my 1000 words are up. Here is a longer version.