SGML - Standard Generalized Markup Language

SGML (Standard Generalized Markup Language) is a standard for defining generalized markup languages for documents. It's a meta-language that provides a formal framework for defining document structure and is the parent language of both HTML and XML.

Overview

SGML was developed by Charles Goldfarb, Edward Mosher, and Raymond Lorie at IBM in the 1960s and became an ISO standard (ISO 8879) in 1986. It provides a rigorous framework for defining document types and establishing rules for document structure, making it ideal for large-scale document management systems.

File Characteristics

• File Extension: .sgml, .sgm, .dtd
• MIME Type: text/sgml
• Character Encoding: Various (ASCII, UTF-8, ISO-8859-1)
• Structure: Tag-based markup with DTD definitions
• Standard: ISO 8879:1986

SGML Components

Document Type Definition (DTD)

The DTD defines the structure and syntax rules for SGML documents:

<!DOCTYPE book [
<!ELEMENT book (title, author, chapter+)>
<!ELEMENT title (#PCDATA)>
<!ELEMENT author (#PCDATA)>
<!ELEMENT chapter (title, section+)>
<!ELEMENT section (title, paragraph+)>
<!ELEMENT paragraph (#PCDATA | emphasis)*>
<!ELEMENT emphasis (#PCDATA)>

<!ATTLIST book
    id ID #REQUIRED
    edition CDATA #IMPLIED>
<!ATTLIST chapter
    number CDATA #REQUIRED>
]>

Document Instance

<!DOCTYPE book SYSTEM "book.dtd">
<book id="book001" edition="2nd">
    <title>Introduction to SGML</title>
    <author>John Smith</author>
    <chapter number="1">
        <title>Getting Started</title>
        <section>
            <title>Basic Concepts</title>
            <paragraph>
                SGML is a <emphasis>powerful</emphasis> markup language.
            </paragraph>
        </section>
    </chapter>
</book>

DTD Syntax Elements

Element Declarations

<!-- Empty element -->
<!ELEMENT image EMPTY>

<!-- Text content only -->
<!ELEMENT title (#PCDATA)>

<!-- Mixed content -->
<!ELEMENT paragraph (#PCDATA | emphasis | strong)*>

<!-- Element content only -->
<!ELEMENT book (title, author, chapter+)>

<!-- Any content -->
<!ELEMENT notes ANY>

Attribute Declarations

<!ATTLIST element-name
    attribute-name type default-value
    id ID #REQUIRED
    class CDATA #IMPLIED
    status (draft|final|review) "draft"
    version CDATA #FIXED "1.0">

Entity Declarations

<!-- Parameter entities -->
<!ENTITY % text "title | paragraph | list">
<!ELEMENT section (%text;)*>

<!-- General entities -->
<!ENTITY company "Acme Corporation">
<!ENTITY copyright "©2024 &company;">

<!-- External entities -->
<!ENTITY chapter1 SYSTEM "chapter1.sgm">

Content Models

Occurrence Indicators

• element - Exactly one occurrence
• element? - Zero or one occurrence (optional)
• element* - Zero or more occurrences
• element+ - One or more occurrences

Group Connectors

<!-- Sequence (all elements in order) -->
<!ELEMENT book (title, author, chapter+)>

<!-- Choice (one of the alternatives) -->
<!ELEMENT media (image | video | audio)>

<!-- Mixed groups -->
<!ELEMENT article ((title, author), (section | appendix)+)>

Attribute Types

String Types

<!ATTLIST element
    name CDATA #REQUIRED          <!-- Character data -->
    id ID #REQUIRED               <!-- Unique identifier -->
    ref IDREF #IMPLIED            <!-- Reference to ID -->
    refs IDREFS #IMPLIED          <!-- List of ID references -->
    nmtoken NMTOKEN #IMPLIED      <!-- Name token -->
    nmtokens NMTOKENS #IMPLIED    <!-- List of name tokens -->
>

Enumerated Types

<!ATTLIST document
    status (draft|review|final) "draft"
    format (html|pdf|print) #REQUIRED
    language (en|fr|de|es) #IMPLIED
>

Default Values

<!ATTLIST element
    required-attr CDATA #REQUIRED    <!-- Must be specified -->
    implied-attr CDATA #IMPLIED      <!-- Optional -->
    fixed-attr CDATA #FIXED "value"  <!-- Cannot be changed -->
    default-attr CDATA "default"     <!-- Default value -->
>

Processing Instructions

SGML Declarations

<!SGML "ISO 8879:1986"
    CHARSET
        BASESET "ISO 646-1983//CHARSET
                 International Reference Version (IRV)//ESC 2/5 4/0"
        DESCSET 0 128 0
    CAPACITY
        SGMLREF
    SCOPE
        DOCUMENT
    SYNTAX
        SHUNCHAR CONTROLS 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
                         19 20 21 22 23 24 25 26 27 28 29 30 31 127
>

Marked Sections

<!-- Conditional sections -->
<![INCLUDE[
    <paragraph>This content is included</paragraph>
]]>

<![IGNORE[
    <paragraph>This content is ignored</paragraph>
]]>

<!-- Conditional with parameter entity -->
<!ENTITY % debug "INCLUDE">
<![%debug;[
    <debug-info>Debug information here</debug-info>
]]>

Tools and Applications

SGML Parsers

• SP (James Clark): Widely used SGML parser
• OpenSP: Open source version of SP
• SGML Parser: Various commercial parsers
• nsgmls: Command-line SGML parser

Authoring Tools

• ArborText Epic: Professional SGML editor
• FrameMaker+SGML: Adobe's structured document tool
• XMLSpy: Supports SGML editing
• Emacs with PSGML: Text editor with SGML support

Conversion Tools

# Parse and validate SGML document
nsgmls -s document.sgml

# Convert SGML to other formats
sgml2html document.sgml
sgml2latex document.sgml
sgml2xml document.sgml

Document Management

Large-Scale Applications

<!-- Technical documentation DTD -->
<!DOCTYPE manual [
<!ELEMENT manual (title, toc?, chapter+, appendix*, index?)>
<!ELEMENT chapter (title, section+)>
<!ELEMENT section (title, subsection*)>
<!ELEMENT subsection (title, (paragraph | figure | table)+)>

<!ATTLIST manual
    part-number CDATA #REQUIRED
    revision CDATA #IMPLIED
    security-level (public|restricted|confidential) "public">
]>

Publishing Workflows

1. Content Creation: Authors create SGML documents
2. Validation: Documents validated against DTD
3. Processing: Documents processed for different outputs
4. Publishing: Generated formats (HTML, PDF, print)

Relationship to Other Standards

HTML Ancestry

<!-- SGML DTD fragment that influenced HTML -->
<!ELEMENT p - O (%text;)*>
<!ELEMENT em - - (%text;)*>
<!ELEMENT strong - - (%text;)*>

<!-- HTML equivalent -->
<p>This is a <em>emphasized</em> and <strong>strong</strong> text.</p>

XML Simplification

SGML features not carried over to XML:

• Optional closing tags
• Attribute value omission
• Complex minimization rules
• SGML declarations

Best Practices

DTD Design

1. Modular Structure: Use parameter entities for reusability
2. Clear Naming: Use descriptive element and attribute names
3. Flexible Content Models: Allow for document evolution
4. Proper Documentation: Document DTD structure and usage

Document Creation

1. Validation: Always validate documents against DTD
2. Consistent Structure: Follow established patterns
3. Semantic Markup: Use elements for meaning, not presentation
4. Version Control: Track changes to both DTD and documents

Processing Considerations

1. Parser Selection: Choose appropriate SGML parser
2. Error Handling: Implement robust error handling
3. Performance: Consider document size and complexity
4. Compatibility: Test with target processing systems

Legacy and Modern Usage

Historical Importance

• Foundation for HTML and XML
• Established principles of structured markup
• Influenced document management standards
• Set precedent for markup language design

Current Applications

• Legacy document systems
• Technical documentation
• Publishing industry archives
• Government document standards

Migration Strategies

<!-- SGML to XML conversion considerations -->
<!-- SGML: Optional closing tags -->
<p>Paragraph one
<p>Paragraph two

<!-- XML: Required closing tags -->
<p>Paragraph one</p>
<p>Paragraph two</p>

Advantages and Limitations

Advantages

• Rigorous document validation
• Flexible markup rules
• Powerful DTD capabilities
• Platform independence

Limitations

• Complexity for simple documents
• Limited tool support
• Steep learning curve
• Processing overhead

SGML remains an important milestone in document markup history, providing the foundation for modern markup languages while demonstrating the power and complexity of comprehensive document structure definition.