As mentioned previously, markup languages both inherit characteristics from earlier print-based markup practices, while extending and augmenting those practices for electronic reading, analysis, and distribution. Markup can be expressed, as we saw previously, via handwritten characters or symbols. For computer-based processing, the markup or annotation vocabulary must be itself expressed in machine readable form in a standardised manner.

But why, you might ask, do we even need markup? Why not simply use plain text? Plain electronic texts, such as txt-files downloaded from online archives such as Project Gutenberg, do usually not contain any markup. For a human reader this tends not to be a problem as long as she knows the language, writing system, and genre conventions.

Take a look at the image to the right containing a piece of writing in which all the characters have been replaced with Xs. Most readers of Indo-European languages will recognize the genre type, without needing to read the content.

We learn how to read texts, how to distinguish individual words, and how to interpret words in the context of their linguistic structure. Even if texts are written without spaces between the words or without punctuation in so-called scriptura continua, a human reader, used to this writing system, can to identify words and sentences and hence discern their meaning.

Similarly, Cross-Hatch writing, a common convention in the Eighteenth Century to save money on post, in which the author of the letter would write horizontally across the page, and then when she reached the bottom of the page, would turn it vertically and write across the previous writing. Ancient manuscripts written in scriptura continua and 18th-century cross-hatch letters prove that humans can adapt to writing conventions that lack white space and/or punctuation. For computers, however, plain text is more difficult to process, particularly when features of individuals words (such as parts of speech), grammatical constructs, or structures of meaning within a text need to be taken into account

Markup is an important enhancement that expands the potential of electronic texts. Markup languages tend to be human readable (i.e they do not need to be compiled like programming languages) as well as descriptive. For example, this fragment of HTML indicates that there are two headings, and the first is more major:

<h1>Markup Languages and Text Modelling</h1>
<h2>Markup Languages vs Plain Text</h2>

Early forms of electronic markup languages were used in the 1960s. In a humanities context, the software COCOA used a form of markup language that was very influential. COCOA was used for the analysis of texts and to generate indexes of words and concordances. These texts had to be prepared and enriched with markup for annotation and meta information (for further information, see Hockey 27-30). The following fragment shows a short example from a dramatic text. Author of a work, the Title of the work, the Act, and Scene, etc. could be encoded and processed with COCOA:

<W SHAKESPEARE> <T HAMLET> <A 1> <S 1><C HAMLET>

Compare this with the following example in LaTeX, a language used to describe documents for typesetting. LaTex has been developed since the 1970s and is one of the most frequently used languages for typesetting in the print industry. For instance, the above HTML example could be written LaTeX in the following way:

\begin{document}
\chapter{Markup Languages and Text Modelling}
\section{Markup Languages vs Plain Text}
\end{document}

Markup languages are used to enrich plain text with information which is readable by both humans and machines. The previously mentioned HTML and XML, the language we will learn in this course, both derive from SGML (Standard Generalized Markup Language). SGML became an international standard in 1986 and is based on an earlier language called GML (Generalized Markup Language). SGML is not an encoding scheme per say, 'but a syntax or a framework within which encoding tags can be defined (Hockey 33). SGML, and hence XML as well, provide a set of conventions in which communities of practice can describe tags, and hence the concepts that are important to capture in their view of texts and textuality. All the languages derived from SGML follow the same convention, specifically angle brackets to distinguish the annotation, known as tags, from the primary text. Recall the above HTML example:

<h1>Markup Languages and Text Modelling</h1>
<h2>Markup Languages vs Plain Text</h2>

There is always an opening tag (e.g. <h1>) and a closing tag (e.g. </h1>) indicating the start and the end of a heading. The only difference between an opening and a closing tag is the slash / which appears after the opening angle bracket in the closing tag. It's not exactly what Freud had in mind when he wrote of the "narcissism of small differences", but this small difference does matter a lot. We'll see why in the rest of this course.

Further Reading
SGML Users' Group. 'A Brief History of the Development of SGML'. 1990. Online.
Hockey, Susan M. Electronic Texts in the Humanities: Principles and Practice. Oxford ; New York: Oxford University Press, 2000.