Standard
PDF/UA-2: what changes, and when it will matter to you
The successor to PDF/UA-1, built on PDF 2.0. What is genuinely new, what is a restatement, and why you should not rush.
ISO — ISO 14289-2:2024 · technical standard: ISO 32000-2 (PDF 2.0) with accessibility requirements
What actually changed
PDF/UA-2 (ISO 14289-2) is built on PDF 2.0 rather than PDF 1.7, and it tidies several things that were awkward in UA-1.
Namespaced structure types. UA-1 has one flat vocabulary, so anything outside it has to be role-mapped to the nearest standard type — which loses information. UA-2 allows namespaces, so MathML can be used for equations as MathML rather than as a Formula with an /Alt string. For technical and scientific documents this is the substantive improvement.
Clearer table model. Header association rules are stated more precisely, which should reduce the disagreements between checkers on complex tables.
Associated files. A first-class mechanism for attaching alternative representations — a source file, a MathML fragment, a data table behind a chart.
Tidier artifact and pagination handling, with the ambiguities of UA-1’s clause 7.8 resolved.
Why you should not rush
Three reasons, all practical.
No instrument names it. ADA Title II names WCAG 2.1 AA. Section 508 names WCAG 2.0 AA. EN 301 549 names WCAG. None of them names PDF/UA-2, or PDF/UA-1 for that matter — PDF/UA is how you satisfy those requirements in the file format, not what they ask for.
The ecosystem has not moved. Checkers, screen readers and the procurement questionnaires that drive most of this work are all built around UA-1. A UA-2 file today is more likely to confuse a reviewer than to impress one.
PDF 2.0 adoption is partial. Producing UA-2 means producing PDF 2.0, and not every tool in a typical pipeline handles it cleanly yet.
How Taggart is positioned for it
Taggart’s checks are defined in a single rule file — the same file that drives the check engine, the report and this site — rather than being hardcoded against clause numbers. A UA-2 profile is therefore an addition of data plus the checks that are genuinely new, not a rewrite of the engine.
That was a deliberate design choice made against exactly this eventuality, and it is why the roadmap can promise a UA-2 profile without promising a rebuild.
Frequently asked questions
Should I be targeting PDF/UA-2 now?
Almost certainly not. Every legal instrument in force names WCAG, and the tooling ecosystem — checkers, screen readers, procurement processes — is built around PDF/UA-1. Meet UA-1 and WCAG 2.1 AA today. Watch UA-2.
Does a PDF/UA-1 file automatically conform to UA-2?
No. UA-2 is built on PDF 2.0 and its requirements are not a strict superset. A UA-1 file is a good starting point, but conformance is a separate claim against a separate standard.
What is the biggest practical change?
Namespaces for structure types, which finally makes it possible to mix vocabularies — MathML for equations, for instance — without abusing the role map. For documents with mathematics that is a genuine improvement rather than a restatement.
Will Taggart support it?
The engine models its checks against a rule file rather than hardcoded clauses, precisely so a UA-2 profile can be added as data rather than as a rewrite. It will land when the checker ecosystem and the procurement language have moved, not before.
Related failure conditions
- 01-005 Content is neither marked as Artifact nor tagged as real content Every text-showing, image-painting and path-painting operator on a page must sit either inside a marked-content sequence that maps to the structure tree, or inside an artifact sequence. Content that is in neither is orphaned.
- 09-001 Tags are not in logical reading order The order of elements in the structure tree does not match the order a human would read the page. This is the single most common substantive defect in tagged PDFs and the one Acrobat's own tooling handles worst.
- 13-004 Figure tag alternative or replacement text missing A Figure element has neither /Alt nor /ActualText. Every Figure must have one or the other; if the image is decorative it should be an artifact instead of a Figure.
Check your own file. Taggart's validator is free and unlimited — no page cap, no watermark, no account.
Primary source: https://www.iso.org/standard/82278.html. This page is a working summary, not legal advice.
Last updated .