Hypertext before the World Wide Web (1988)

Richard Rathe, 1988 (History/Programming)

When I was a resident in 1988 I wrote a bit of software called HYTEXT—a hypertext publishing system for MS-DOS computers. I knew nothing about networks at the time, but I did create a markup language that anticipated HTML in many ways. I recently came across a paper copy of my original documentation and thought it would be a good idea to reproduce a few pages here for posterity. The original program and documentation are on 5.25 inch floppy discs that I have no way to read. But I found my 22 year old documentation file online! So the circle is complete.

Here is my representation of two hypertext pages displayed on a 80 column by 24 line (!) monochrome display…

I called these pages Topics and you could scroll them 12 lines at a time using the Page Up and Page Down keys. You navigated the links using the Arrow Keys and pressing Enter. This is very similar to text-only browsers still in use today (such as Lynx). In addition there were two means of showing additional information without jumping to another topic. Popup Cards contained brief annotations and definitions, very similar to the hover info/tool tips commonly seen on web pages today. Expansion Text allowed the author to hide lines of text that would only appear if the user pressed a special key. Again this technique is frequently used on websites to compress content for easier scanning/browsing. Here is a conceptual diagram of how it worked…

The syntax was rather simple—using the seldom-used backtick (`) and tilde (~) characters (which are conveniently found together on the same key on modern keyboards).

Sample Hypertext

`start`

    HYPERTEXT can be defined as the NON-LINEAR, HIERARCHICAL presentation
of information. In contrast, a book represents the most well known LINEAR
system for information storage and access. One tends to use a ~book~0~book~
by preceding from beginning to end. The most sophisticated internal
references a book usually contains are the table of contents and index. A
concept on one page has no direct way to reference another concept on a
different page.

    This is not true with a ~hypertext~0~hyper~. A page can contain HOT WORDS which
are directly linked to related TOPICS anywhere in the document…

…

`start`

To define a block of text the writer enclosed it with tags such as `start`. This is the exact equivalent of the html…

<p id=start>a block of text</p>

Hypertext links were constructed using tildes ~hypertext~0~hyper~. This is the equivalent of the html link within a document. It was also possible to jump between documents and even run external programs.

<a href=hyper>hypertext</a>

The ~0~ was used to place a value on a link. This was ignored for regular hypertext, but could be used to score the user's path through the text. I called this hypertesting.