Corporate pioneers in computer text processing and typesetting tended to work alone. This is pretty normal in the early stages of developing a new technolgy: everyone wants to be first to market with the new systems.

The result of this was that each organization developed its own specific way to "mark up" text for their computer system, like MIT's runoff system and IBM's page-1. Unlike the human editors and typesetters, there was no standard markup language for computerized text processing and printing.

So text input into one brand of computer and marked up for printing couldn't be transferred to any other brand of computer. This was inconvenient and very expensive - espcially as newer technology came out, incompatible with the older markup language.

The expense and inconvenience of specific computer markup languages drove people to consider creating a generalized markup language, that could be used on many systems. The first person to campaign for a generalized markup language was William Tunnicliffe, chairman of the Graphic Communications Association, in 1967, according to Charles Goldfarb, in his talk "The Roots of SGML -- A Personal Recollection" (http://www.sgmlsource.com/history/roots.htm).

Goldfarb himself worked for IBM with Ed Mosher and Ray Lorie to create in 1969 the Generalized Markup Language (GML). The purpose of GML was to allow markup to serve two purposes:

• Mark the function of different parts of a text for information retrieval; and
• Allow formating of the different parts of text for printing.

Here's how Goldfarb himself put it, in "Design Considerations for Integrated Text Processing Systems", IBM Cambridge Scientific Center Technical Report G320-2094, May 1973 (but written in 1971):

This analysis of the markup process suggests that it should be possible to design a generalized markup language so that markup would be useful for more than one application or computer system. Such a language would restrict markup within the document to identification of the document's structure and other attributes. This could be done, for example, with mnemonic "tags". The designation of a component as being of a particular type would mean only that it will be processed identically to other components of that type. The actual processing commands, however, would not be included in the text, since these could vary from one application to another, and from one processing system to another. (Quoted in Goldfarb's The Roots of SGML -- A Personal Recollection)

The idea of a generalized markup language caught on rapidly, but there were problems with GML:

• It was owned by IBM, so it couldn't be developed by any other company or group;
• It was difficult to extend to meet new needs.

Because of these problems, people from the information and text processing community got together to create something new. These people included Goldfarb, Tunnicliffe, and Brian Reid - a Carnegie Mellon University researcher who had developed a generalized markup language called Scribe. What they needed was a language that was...

• An open standard, rather than proprietary;
• Flexible enough to extend to lots of new needs.

These researchers formed the nucleus of a group of hundreds of people from around the world who worked on this effort for eight years. The result, in 1981, was the Standard Generalized Markup Language, SGML: ISO 8879. (Like all ISO standards, this is available for sale, but not for free viewing.)

SGML is known as a meta-language because it is not used directly in text markup, but is used to create generalized markup languages.

The original 1981 version of SGML was revised and a new version issued in 1986. Beyond that, the standard has stayed stable and continues to be widely used. SGML is the basis of:

• HTML
• XML
• XHTML
• and scores of other markup languages.

We turn now to some of the best-known descendants of SGML: those used on the World Wide Web.

The World Wide Web has been based on two "languages":

• HyperText Markup Language (HTML) for marking up text to be displayed on the Web; and
• HyperText Transfer Protocol (HTTP) for transferring information between Web clients (primarily browsers) and Web servers.

We'll be looking in more detail at HTML here.

Tim Berners-Lee put the concept of hypertext together with the capabilities of the Internet in 1989, primarily for the benefit of the scientists with whom he worked at the European Particle Physics Lab, "Conseil Européen pour la Récherche Nucleaire" (CERN) near Geneva, Switzerland.

The great success of the Web was in part the result of the purpose for which Berners-Lee created it:

This resulted in a markup language based on SGML that was...

• Simple:
  Berners-Lee used a very small subset of SGML, which meant that preoccupied scientists didn't have a complex system to learn, and browsers could be relatively small and fast.
• Flexible:
  As the Web became increasingly popular, people were able to do many things with it besides mark up scientific reports.

It seems now that there is no facet of information exchange that isn't handled on the World Wide Web. Millions of people create Web pages for thousands of purposes. The Web has become an art form as well as a source of information.

The Web also became an arena in which giant corporations wrestled with each other for market dominance. One of the tricks these giants used to gain dominant positions for their products was adding new features to the markup language their browsers could display. This led to helter-skelter growth of HTML, frustration for Web developers, and confusion for Web users.

By 1994, the situation was difficult, and by 1996 it was intolerable. Tim Berners-Lee left CERN in 1994 to found the World Wide Web Consortium (W3C) , his idea being to get the major Web stakeholders together and convince them to develop standards that would allow the Web to grow in a less confusing way. His effort has largely succeeded, owing to his vision, combined with the efforts of hundreds of members and staff at W3C. The results was increasing acceptance of standard versions of HTML, up through version 4.01.

We'll take a look at some of the fruits of this effort next.