[Athen] How do WCAG & PDF/UA authorize removing running headers and footers from PDFs?

[Karen McCall]

Thanks Bevi! I'd left work when Adina's post came in and was going to reply this morning.

If anyone needs more clarification, let me know on or off list.

Cheers, Karen

From: athen-list <athen-list-bounces at mailman12.u.washington.edu> On Behalf Of chagnon at pubcom.com
Sent: Thursday, June 29, 2023 7:47 PM
To: 'Access Technology Higher Education Network' <athen-list at u.washington.edu>; EDUCAUSE-ITACCESS at ConnectedCommunity.org
Subject: Re: [Athen] How do WCAG & PDF/UA authorize removing running headers and footers from PDFs?

Clarifying about headers/footers in PDFs...apologies in advance because this is long and detailed.

WCAG doesn't address headers, footers, and page numbers correctly. And WCAG isn't the right standard to reference for accessible PDFs.

Instead, you'll want to reference the PDF/UA-1, the current ISO standard (14289) for PDFs. You can purchase the standard directly from either the ISO in Geneva or the PDF Association, the designated and authorized manager of the ISO standard.

*         ISO's website is https://www.iso.org/standard/64599.html

*         PDF Association's is https://pdfa.org/resource/iso-14289-pdfua/#pdf-ua-1  Purchase button is in the upper right corner of the page.

There's also a freebie reference guide I recommend to our students and clients. Written by the working group that produces the PDF/UA standard itself, it provides very solid guidelines on how to use PDF/UA tags for accessible PDFs. The "Syntax Guide" is free from the PDF Association at https://pdfa.org/resource/tagged-pdf-best-practice-guide-syntax/  Warning: it is written for programmers so read between the lines of code to see the actual tag structures.

In it, you can reference Section 3.5 Artifacts, which discusses both headers/footers and repeating page numbers. In a nutshell, they all must be artifacted so that assistive technologies skip them to avoid interfering with the content stream, but they're still visible on the page for sighted users. That's what Karen McCall was saying in her previous post. Screen reader users don't want to hear "Chapter 3: The History of the 100 Years War page 123" in the middle of a sentence that begins at the bottom of a page and reflows to the top of the next page. And they don't need to hear "Chapter 3: The History of the 100 Years War page whatever" because they already know they're in Chapter 3 and they can announce the specific page number they're on any time they need to know it.

The best way to artifact these elements is in the source document, not the exported PDF.

*         In Word, everything placed inside Word's header and footer sections is automatically artifacted when the PDF is exported, so everything should be correct in the final PDF. Learn how to make an accessible Word document that exports to an accessible PDF.

*         In PowerPoint, certain items on the Slide Masters are automatically artifacted when the PDF is exported. Learn how to make an accessible PowerPoint slide deck that exports to an accessible PDF.

*         And in Adobe InDesign, all items placed on the Master/Parent pages is automatically artifacted. Additionally, individual objects (text frames and graphics) in the body of the page can be manually designated as artifacts, too. Learn how to make an InDesign layout that exports to an accessible PDF.

When the tools in these authoring programs are used correctly, they will produce a compliant PDF that doesn't need any remediation afterwards...well, for these items at least! <grin>

However, if you're stuck with a PDF and can't get your fingers on the source file, then you can manually correct this in Acrobat:

1.       Select the parent tag, such as <P> that holds the header/footer.

2.       Expand the tag to expose the yellow content container inside the tag.

3.       Right-click on the yellow content container and select Artifact. The yellow content container will now disappear from the tag tree.

4.       It leaves behind a now empty <P> tag which can safely be deleted. But be careful: only truly empty tags can be deleted so make sure the <P> doesn't have any content containers or the grey expansion arrow to its left.

FYI, both myself and Karen McCall are on the ISO committees for PDF and PDF/UA. There are over 50 specifications for different types of PDF files and their usages, and we're both on a smattering of them, as well.  Both of us were on the first WCAG teams that created the original accessibility standards.

And personally, I've been a traditional programmer, PDF programmer, web developer, and digital media developer. A website's code is exceptionally simple compared to at PDF's code. The code structure of the two are extremely different, and there are differences between what a webpage presents as content from what a Word/PDF document presents. Like headers and footers and page numbers and text content that reflows from page to page, column to column.

Know all of the standards and understand which one to use for different types of content. We have recent blogs on this at:

*         Accessibility Standards: https://www.pubcom.com/blog/standards/wcag-pdf/index.shtml

*         US Laws for Accessibility: https://www.pubcom.com/blog/us-laws/all/index.shtml

*         PDF/UA tag set: https://www.pubcom.com/blog/2020_05-02_tags/pdf-ua-tags.shtml

Hope this helps everyone make accessible PDFs as quickly and painlessly as possible.

-- Bevi Chagnon

- - -
Bevi Chagnon  | Designer, Accessibility Technician |  Chagnon at PubCom.com<mailto:Chagnon at PubCom.com>
US Delegate to the ISO Committees for PDF and PDF/UA Accessibility
- - -
PubCom: Technologists for Accessible Design + Publishing
consulting * training * development * design * sec. 508 services
Upcoming classes at www.PubCom.com/classes<http://www.pubcom.com/classes>
- - -
Latest blog-tutorial<https://mailchi.mp/e694edcdfadd/class-discount-3266574> - The 4 Reading Orders in PDFs<https://www.pubcom.com/blog/2020_08-18_ReadingOrder/reading-orders.shtml>

From: athen-list <athen-list-bounces at mailman12.u.washington.edu<mailto:athen-list-bounces at mailman12.u.washington.edu>> On Behalf Of Adina Mulliken
Sent: Thursday, June 29, 2023 2:23 PM
To: athen-list at u.washington.edu<mailto:athen-list at u.washington.edu>; EDUCAUSE-ITACCESS at ConnectedCommunity.org<mailto:EDUCAUSE-ITACCESS at ConnectedCommunity.org>
Subject: [Athen] How do WCAG & PDF/UA authorize removing running headers and footers from PDFs?

Hi all,
Apologies for the cross posting.

I'm wondering if anyone could advise me what success criteria or techniques within WCAG you think could most clearly be interpreted to authorize "removing" repetitive headers and footers from PDFs?  I'm aiming to find the most specific criterion or "technique" or "failure" that I can.   Maybe 2.4.1 Bypass blocks?  Or Info and Relationships?  If nothing else, I suppose this could fall under POUR principles.  Or anything within PDF/UA or any other standards you know of?  Maybe this sentence that I found in PDF/UA: "Artifacts shall be marked as such and shall not be tagged in the structure tree."?

I would also appreciate opinions from any of you all in the accessibility community saying whether you think it's important to remove repetitive headers and footers that interrupt the flow of text, as that could be helpful for me.

I want to see if I can justify erring on the accessibility side of a possible conflict between copyright/licensing versus accessibility.  (This is a situation where exceptions to copyright for people with disabilities wouldn't apply.)

Please respond to me directly if you prefer.

Thank you!
Adina

Adina Mulliken
Associate Professor, Librarian Silberman Social Work and Urban Public Health Library
Hunter College, City University of New York
2180 3rd Ave. New York, NY
Phone 212-396-7665
Pronouns she/her
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman12.u.washington.edu/pipermail/athen-list/attachments/20230630/fce29a42/attachment.html>