Accessible Document Generation via D/LDL Structure Embedding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies face challenges in accurately determining the structure of digital documents, particularly in formats like PDF, where structure is separate from presentation, leading to errors in accessibility features for diverse disabilities and devices, due to complex elements like tables, lists, and missing Unicode mappings.
Innovation Solution
A system using Data/Layout Description Language (D/LDL) with placeholders generates an accessible layout file, populated by a preprocessing module, converted into an object model by a Model Builder, and formatted by Position and Standards formatters to create standards-compliant documents with exact positioning and accessibility attributes, enabling conversion into various formats like PDF and HTML.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated processes are used to determine document structure after generation, then processing speed is improved, but accuracy deteriorates due to error-prone structure determination
Solution Approach 1:
The patent applies preliminary action by embedding structural information directly into the document generation process itself, rather than attempting to detect structure after the document is created. The system uses a structured template approach where the document structure is predetermined and enforced during generation, ensuring both speed and accuracy simultaneously.
2Ease of operation
If pattern recognition is applied to guess document structure from layout, then ease of operation is improved, but accuracy deteriorates for complex structures like tables and lists
Solution Approach 1:
The patent introduces an intermediary structured template layer between the raw document content and the accessibility requirements. This template serves as a mediator that provides predefined structural frameworks (for tables, lists, headings, etc.) that automatically guide the document generation process, ensuring both ease of operation and high accuracy for complex structures.
3Ease of manufacture
If documents are created without structure to simplify creation, then ease of manufacture is improved, but accessibility deteriorates due to missing structural information
Solution Approach 1:
The patent makes the structured template universal by designing it to handle multiple document types and complexity levels through a single unified framework. The template system can accommodate simple documents as well as complex ones with tables, lists, and nested structures, providing both ease of creation and reliable accessibility compliance across all document types.
4Adaptability or versatility
If PDF format is used where structure is separate from presentation, then adaptability is improved for different layouts, but difficulty of detecting structure increases
Solution Approach 1:
The patent resolves this contradiction by performing preliminary structuring action during the document generation phase itself. Rather than attempting to detect structure after the flexible PDF layout is created, the system embeds structural metadata and organization information into the document creation process, maintaining layout flexibility while eliminating the difficulty of subsequent structure detection.
Data Source
AI summary
An architecture for generating accessible documents uses a Data/Layout Description Language (D/LDL) and data sources. The D/LDL and the data source are fed to a pre-processing module which generates an intermediate document containing all the data and positioning hints and structural information required to create the accessible file. The intermediate file is converted into an object model using a Model Builder and is passed to a Model Formatter to calculate the exact positioning of elements. The object model is then passed to a Standard formatter to ensure compliance with required standards and is then passed to a model writer which translates the object to an accessible file in the required format. The design allows for skipping the preprocessing module if the full file is available and allows to plug different model writers for different formats.