Hi Mark,

You've stepped into quite a sticky wicket. I can't offer specific advice, but maybe some helpful leads and my experience.

From my understanding PDF/UA is not a file format but a set of policy implementations, hence the difficulty 'validating' any specific instance. My experience with massive PDF authoring/508 remediation efforts suggest that no PDF authoring tool itself will get you 100% of what you need -- there is often some manual intervention needed. The easiest path to authoring a WCAG 2.0 (A, AA, or AAA) compliant PDF usually depends on the capacity of the source data/software, like using tooling available in MS Word to properly structure a document before PDF export, or a WCAG 2.0 compliant HTML document converted to PDF. The PDF validator link you offered seems to rely on a viewing software's capacity to render PDF/UA features rather than the technical aspects of a PDF file itself. We've been playing with VeraPDF's validators and I don't see PDF/UA listed as a target format. Acrobat XI Pro and Common Look are some other PDF-specific tools.

Unless someone is insisting on PDFA/UA as a deliverable, you might be better served focusing on the WCAG 2.0 piece especially focusing on the POUR concepts. The wiki site you mentioned in another listserve post states "The proposed rule identifies PDF/UA as equivalent to WCAG 2.0 for "appropriate content"." -- enough holes there to drive several trucks through. There is more validating technology and checklists for WCAG 2.0 compliance (which BTW is not the same as Section 508 accessibility) such as WAVE and Deque Axe. DHHS's new accessibility standard is WCAG 2.0 AA, although most public-facing websites mostly still refer to the old 508 checklists, so beware of the online content you consume!

FWIW, we are transitioning from an in-house tool to Deque Axe for HTML accessibility checking and issue management. Axe also has some developer integrations so you can run tests as you develop rather than remediate afterwards.

We're getting away from retrofitting/policing file formats to focusing on authoring/UX, so you are taking the path to righteousness.

https://www.w3.org/WAI/intro/wcag
https://www.deque.com/


John


John P. Rees
Archivist and Digital Resources Manager
History of Medicine Division
National Library of Medicine
301-827-4510



-----Original Message-----
From: Custer, Mark [mailto:[log in to unmask]]
Sent: Monday, January 22, 2018 11:17 AM
To: [log in to unmask]
Subject: Re: Accessible PDF finding aids? Print finding aids at all?

This is a great discussion so far; thanks to everyone who has weighed in already!

To be clear: Personally, I'm a strong advocate for printable finding aids, but I don't think that archival repositories should be the ones printing them out (though I guess many do, like mine, during the editing / proofreading process anyway!) and/or filing them on site.

What's surprised me is that it seems like fewer and fewer finding-aid delivery services are offering PDFs or a similar option for download. ***I also can't find any examples of PDF/UA finding aids in the wild***. It is a relatively new standard, but it is also subject to Section 508, for example: https://www.section508.gov/blog/check-pdf

As for the technical aspects, Brian, I am relying on Apache FOP to create PDF/UA finding aids with XSL-FO (I thought about pursing a CSS option, but I don't know if any are viable right now to create PDF/UA documents). Apache FOP, which is open source, has been offering accessibility options since version 1.1, https://xmlgraphics.apache.org/fop/1.1/accessibility.html, it looks like. I'm now using version 2.2, though, since it includes a number of fixes, and it also ensures that the document title of the finding aid is immediately recognized by Adobe Acrobat rather than the file name.

Anyhow, more in a few weeks once I have a few other concrete details worked out. But I'm still interested to hear more about this discussion in the meantime, especially if anyone else out there is working on creating PDF/UA finding aids. If so, let's talk ASAP. I've got a lot to learn still about this topic, so I'd love to share notes.



-----Original Message-----
From: Encoded Archival Standards List [mailto:[log in to unmask]] On Behalf Of Brian Sheppard
Sent: Monday, 22 January, 2018 10:44 AM
To: [log in to unmask]
Subject: Re: Accessible PDF finding aids? Print finding aids at all?

I did a lot of work on building PDFs via XSL-FO. They have the advantage of course, of being portable, but they can also provide good context — with breadcrumb running headers — and good navigation via the headers or the TOC. (Only useful if viewed on an electronic device of course, but it's complete and relatively compact.)

Unfortunately, Apache FO didn't seem up to the task and the proprietary FO engines we experimented with were very good but beyond our budget.

> On January 22, at 9:19 AM, Bowers, Kate A. <[log in to unmask]> wrote:
>
> Printed finding aids are vital. Not just because of the look of horror on researchers' faces if you were to take them away, but because archivists, too, need to understand whole collections wholistically sometimes. Especially for large collections with large finding aids, print is essential. PDF online or print eliminates confusing and distracting navigation, searching, filtering, and branding stuff that sits on a website. Then there's the problem of eyesight and projected light as opposed to reflected light....hurrah for print!
>
> Kate Bowers
> Collections Services Archivist for Metadata, Systems, and Standards
> Harvard University Archives [log in to unmask]
> voice: (617) 998-5238
> fax: (617) 495-8011
> web: http://nrs.harvard.edu/urn-3:hul.eresource:archives
> Twitter: @k8_bowers
>
>
>
>
>
> From: Encoded Archival Standards List <[log in to unmask]> on behalf
> of Custer, Mark <[log in to unmask]>
> Sent: Monday, January 22, 2018 9:45 AM
> To: [log in to unmask]
> Subject: [EAD] Accessible PDF finding aids? Print finding aids at all?
>
> All,
>
> I've just embarked on a project to update how our PDF finding aids are created to ensure that they are compatible with the PDF/UA standard (en.wikipedia.org/wiki/PDF/UA), by converting ArchivesSpace-produced EAD3 finding aids to PDF (yay, EAD3!). So far so good, but getting things set up was a bit more trouble than I expected, though that's probably because I knew essentially nothing about the PDF/UA standard before I started. Additionally, I have not been able to find any evidence of PDF/UA finding aids available online. Given that, I thought that I'd ask the EAS Section the following questions:
>
> • Can anyone point me to examples of PDF/UA finding aids?
> • Is anyone using any PDF alternatives to fill this gap? I've seen quite a few examples of print-friendly HTML finding aids, for example, but I'm not sure if any of those a) meet WCAG 2.0 guidelines and b) help researchers (since there are lots of ways to use an offline PDF, which I'm not sure that static HTML page would fulfill). At the same time, those print-friendly HTML files could easily be converted to other formats.
> • Last, and most importantly: are print finding aids a requirement at this time or not? I seem to remember some really compelling data from user studies (probably about 10 years old now) that would suggest that they are crucial for researchers.... And yet, I see fewer places offering PDF finding aids. There are a whole host of reasons for why this might be, but right now I'm just really curious what the list thinks about user needs in relation to print finding aids in 2018.
>
> All my best,
>
> Mark

—————————————————————————————
Brian Sheppard
UW Digital Collections Center
[log in to unmask]