The Craft
of Plain Text

Plain text

Rich text

Developed by Microsoft in the mid-1980s, RTF (Rich Text Format) is a proprietary document file format that supports text formatting, such as bold and italics, as well as graphics. Unlike plaintext, RTF retains style and formatting information, making it suitable for exchanging formatted documents between applications and platforms. While not as ubiquitous today, it has played a significant role in the evolution of document interchange formats.

The typewriter marked the beginning of mechanical text editing, offering a more efficient way to write than handwriting. Originally invented in the early 1870s, it remained a staple in offices and homes well into the 20th century.

Punch cards were one of the earliest methods used for programming and data storage, especially in early computing. A sequence of punched holes represented data or commands. They were widely used from the early 1900s until the 1970s.

Before the advent of video displays, teletype systems were used as a means of entering and displaying text data in real-time. These machines were often connected to early mainframe and minicomputers and marked the transition to electronic text editing.

ED (short for "editor") was one of the first command-line text editors, created in the early 1970s. Designed by Ken Thompson for Unix systems, ED set the foundation for many text editors that followed. While primitive by today's standards, it introduced key editing concepts still used.

With the advent of video display terminals in the late 1970s and early 1980s, text editors evolved to show text directly on a screen rather than print them on paper. Editors like "vi" (on Unix) and "WordStar" (on DOS) emerged, offering users a visual interface to interact with their documents. They marked a significant shift from command-line based editors, providing real-time feedback and enhancing user experience.

Developed in the mid-1970s by Richard Stallman, Emacs is more than just a text editor; it's an extensible and customizable editing environment. With its unique system of key bindings, and its own built-in programming language (Emacs Lisp), users can customize and extend their editing experience. It is renowned for its powerful features, enabling tasks from editing text to reading emails or even playing games, all within the same interface.

WYSIWYG editors revolutionized the world of text editing in the late 1980s and early 1990s. Instead of users having to remember markup or commands, these editors allowed for direct manipulation of the text and its appearance. With WYSIWYG, the display on the editor closely matched the final output.

In computers, everything is stored as binary data – sequences of 0s and 1s. Text is no exception. Each character is mapped to a unique number using a character encoding, like ASCII. For example, in ASCII encoding, the letter 'A' is represented as the number 65, which in binary is '01000001'. When we type a letter, the computer processes its binary representation, but displays the character we recognize, allowing us to interact using familiar symbols.

ASCII, standing for American Standard Code for Information Interchange, was developed in the early 1960s by the American National Standards Institute (ANSI). Designed as a standardized character encoding for electronic communication, ASCII represents text in computers, telecommunications equipment, and other devices that use text. Originally, it was created to address the compatibility issues between different types of data processing equipment.

The ASCII table consists of 128 characters, encompassing control characters (like carriage return and line feed) and printable characters (letters, numbers, and symbols). Each character is represented by a unique 7-bit integer. This table became the backbone for text encoding in early computing and forms the foundation of many encoding systems used today.

Control characters, often non-printable and invisible, are special characters used in text to represent non-graphic instructions. Originating from the early days of telegraphy and teletype, they served to control the behavior of devices or instruct them to perform specific operations. In character encodings like ASCII, control characters occupy the first 32 codes (from 0 to 31).

Text encodings are essential because they define how characters are represented as numbers in computers. As the digital world expanded, the need arose for a system that could accommodate more than just English letters and symbols. Different regions and languages necessitated their own encoding schemes, leading to a plethora of text encodings tailored to various linguistic and technical needs.

Codepages are sets of characters for a particular encoding. In the era before Unicode, different regions used different codepages to represent their languages. For instance, Western European languages might use the ISO 8859-1 codepage, while Cyrillic-based languages might use ISO 8859-5. Switching between codepages was often necessary to display multiple languages correctly.

KOI-8R (Kod Obmena Informatsiey-8 Rossiya) is an 8-bit character encoding, designed to cover Russian, which uses the Cyrillic script. Introduced in the 1980s, KOI-8R became one of the main encodings used in Unix-based systems in Russia. An interesting feature of KOI-8R is its resilience to transmission errors: if a single bit is lost, a Cyrillic letter turns into another Cyrillic letter.

• Insertion: Adding characters or words at a specific position.
• Deletion: Removing characters, words, or lines from the text.
• Search: Locating specific words, phrases, or patterns within the document.
• Replace: Swapping out specific characters or words with others.

• Navigation: Moving the cursor or focus to different parts of the document.
• Selection: Highlighting specific portions of text for other operations.
• Undo/Redo: Reversing or re-applying recent changes.
• Copy/Cut/Paste: Duplicating or moving text within the document or between documents.

The gap buffer is a dynamic array with a "gap" or empty space. Cursor movements and text edits occur at the gap's location. When characters are inserted, the gap size decreases, and when they're deleted, the gap size increases. The gap moves with the cursor, and while this requires moving the characters around, it ensures that insertion at the cursor location is always fast. Gap buffers are used in popular editors like Emacs.
Read more

The piece table is a data structure that maintains original text and its modifications in separate buffers. Instead of editing the original content, modifications are stored separately, and a table keeps track of "pieces" of the content to construct the current state of the text. This approach makes operations like undo and redo very efficient. Piece tables were used in early word processors like Word for DOS.
Read more

Ropes and piece trees are a binary-tree-based data structures where each leaf node holds a string fragment. Trees efficiently handle operations like insertion, deletion, and concatenation by manipulating tree nodes rather than characters. Ropes have been employed in editors like Xi, while piece tree is utilized by VS Code.
Piece tree
Rope

Text rendering is the process of displaying text in a visually readable form on a digital screen. It involves a series of steps, including layout calculations, glyph selection, and pixel rendering, which collectively bring text from a font file to your screen.

• Typefaces vs. Fonts: Typeface is the design; Font is the implementation.
• Font Styles: Includes regular, bold, italic, etc.

Bitmap fonts, sometimes known as raster fonts, are fonts in which each character is represented as an arrangement of pixels in a grid. Each glyph is essentially a bitmap image, predefined for a specific size and resolution.

• Fixed Size: Each bitmap font is designed for a specific size. Scaling can result in jagged or blurry edges.
• Pixel Perfect: At their intended size, bitmap fonts are crisp and clear, especially useful for screens with low resolution or specific rendering needs.
• Limitations: Lack of flexibility in scaling and transformations compared to vector fonts.
• Use Cases: Early computer displays, retro video game graphics, and specialized screen applications.

Vector fonts, also known as outline fonts, are fonts where each character is defined using geometric shapes like curves and lines. Unlike bitmap fonts, which represent characters using a grid of pixels, vector fonts use mathematical equations to describe the contours and paths that make up each glyph. The most common vector font formats include TrueType (TTF) and PostScript Type 1.

• Scalability: Vector fonts can be resized without any loss of quality, ensuring sharp and clear characters at any size.
• Flexibility: Can be easily transformed (rotated, skewed) while maintaining clarity.
• Print Quality: Provides smoother edges, especially for large print sizes.
• File Size: Generally, smaller file sizes than bitmap fonts for the same character sets.

Rendering a line of text isn't as straightforward as merely displaying glyphs side by side. The process takes into account various factors such as kerning, baseline alignment, and text directionality. Properly rendered lines ensure text is readable, aesthetically pleasing, and maintains the intended meaning.

The viewport is the visible area of a text document or any other digital content. As documents exceed the size of the viewport, scrolling becomes necessary to navigate through the content. Effective text scrolling is crucial for user experience, allowing seamless exploration while keeping the content readable and well-rendered.

Anti-aliasing techniques smooth out jagged edges in rendered text by subtly coloring the pixels around the edges. This makes the text easier to read and more visually pleasing.