Published on [Permalink]
Reading time: 4 minutes

It used to look like this

Forty-three years ago I was being harangued by university tutors to go and use the computer. Not because it would make me a better writer. Because they couldn’t read my handwriting.

The “word processor” they had in mind was called DSR - DEC Standard Runoff. It looked like this:

Input
When you're ready to order,
call us at our toll free number:
.BR
.CENTER
1-800-555-xxxx
.BR
Your order will be processed
within two working days and shipped
Output
When you're ready to order, call us at our toll free number:
1-800-555-xxxx
Your order will be processed within two working days and shipped

(in RSS? You need to visit the web page to see the formatting)

You typed green text on a black screen. The trick was that you preceded certain lines with a dot and a command. .BR forced a line break. .CENTER centred the next line. .JUSTIFY tidied up the margins. Nothing on screen told you what it would look like when printed - the screen was just a stream of characters. What mattered was the output.

And output was an event. You’d submit a batch job to the computer department and come back several hours later to collect your pages from a dot matrix printer the size of a chest freezer. Loud enough to hear from two corridors away. The paper came out in a continuous fan-fold roll and you tore your assignment along the perforations, hopefully in the right place.

DSR was DEC’s own flavour of a much older idea. RUNOFF was originally written by Jerome Saltzer at MIT in 1964 for the Compatible Time-Sharing System - a text formatter where you annotated a document with markup commands and let the program handle the layout at print time. This is the direct ancestor of troff, TeX, HTML and Markdown. The dot-command concept in that demo above is essentially unchanged from what Saltzer wrote six decades ago.

Two things had to happen before any of this connected to ordinary people. The first was the Macintosh in 1984 - the moment a graphical interface stopped being a research curiosity and became something you could buy at a shop. The second was Tim Berners-Lee, who in 1991 built the first prototype of the World Wide Web on a NeXT workstation at CERN. That machine - a sleek black cube - is still there. On it he invented the URL, HTTP and HTML. He also put a note on it that read: “This machine is a server. DO NOT POWER IT DOWN.”

The web existed. But for most people it was still invisible.

Then in 1993, NCSA Mosaic arrived - the first browser to display images inline with text rather than in a separate window. It sounds like a small thing. It was not. Mosaic made the web look like something. It turned a protocol into a place.

I know exactly when I first saw it. My friend George Bray sat me down in front of a screen 33 years ago and showed me Mosaic running. I had no framework for what I was looking at. A page. With text. And a picture. On a computer. Somewhere else.

I am typing this now with an editor that shows me exactly what it will look like. When I publish it, anyone with a browser could read it - roughly five billion people at last count.

Probably won’t. But the point is they could. Instantly. For free. From anywhere.

That gap - from a batch job queued at a university computer department to a post on the open web - spans 43 years and three of the most consequential inventions in the history of communication. It feels almost too large to hold in your head at once.

The tutors haranguing me to “go and use the computer” were not wrong, exactly. They just had no idea what they were pointing me toward. Neither did I.

Anyhow. That is the observation. DSR would have required at least three .BR commands just to format this ending.


Sources:

✍️ Reply by email