| Term | Explanation |
|---|---|
| Tag | Describes a content item, such as a paragraph or table cell. |
| Tag tree | The document’s hierarchy of those descriptions. |
| Reading order | The sequence in which content should be read. |
| Alternative text | Text conveying an image’s relevant information or function. |
| Artifact | Content separated from the meaningful structure, such as certain decoration. |
| OCR | Optical character recognition: converting characters in a scanned image into machine-readable text. |
| Screen reader | Assistive technology presenting content and structure through speech or braille. |
| Validation | Checking against a defined set of rules. The result depends on the rules and tool. |
WCAG is W3C’s Web Content Accessibility Guidelines. A success criterion is a testable requirement, such as not conveying information through colour alone. A, AA and AAA are conformance levels; AA includes A. A PDF technique describes one way to satisfy criteria, rather than a separate standard.
PDF/UA is an ISO standard series for accessible use of PDF technology. PDF/UA-1 and PDF/UA-2 identify different parts and technical foundations, not scores in which two means twice as accessible. PDF/A is a separate standard series for long-term preservation.
EN 301 549 is a European accessibility standard for information and communication technology. Its document requirements form part of a broader specification. Legislation may reference one edition, while a contract may add requirements. A commission therefore needs to state what will be assessed, rather than merely collect acronyms.
Metadata describes the document, including its title and certain identifiers. The document language is another technical property and may be supplemented for individual passages. A filename and a document title are different things.
An annotation is a PDF object used for features such as links or form fields. Its interactive area and the text’s structure need to work together. Blue, underlined text does not become a working link merely through its appearance.
Unicode identifies characters. A PDF needs to convey which letter or digit a drawn character represents. ActualText supplies replacement text when the ordinary text representation does not work. Alt supplies an alternative description, such as a description of an image. Their purposes differ; they should not routinely receive identical strings.
Scope and Headers describe relationships between table cells. Scope can identify a row or column header. Headers can associate a cell with explicitly identified header cells. A namespace identifies the vocabulary to which a structure type belongs. A role mapping relates a custom tag name to a standard type.
A tagged PDF has structure, but that structure may be wrong. A searchable PDF contains text that can be found, but may lack headings and reading order. A PDF/UA identifier in metadata is a declaration, not an independent test result.
When someone calls a file compliant, ask which requirements, which assessment method and which file version they mean. This makes the claim possible to examine.
- Find the object or page identified by the report.
- Translate the rule name into a problem for the reader.
- Check that the repair solves the problem, rather than merely removing the message.
Sources
- Tagged PDF Best Practice Guide: SyntaxPDF Association
- PDF7: OCR for scanned PDFsW3C
- ISO 14289-1:2014 (PDF/UA-1)ISO
Examples are our own unless stated otherwise. These guides distinguish legal requirements, standards and practical advice. Passing an automated check is not a complete assessment of document accessibility.