a11y-discov-vocab
a11y-discov-vocab copied to clipboard
w3c
taggedPDF definition
Currently defined as:
<The structures in a PDF have been tagged to improve the navigation of the content.>
This is a bit misleading. An untagged PDF presents no content to a screen reader. It isn't just inadequate navigation/structure.
Proposed change:
<The contents of a PDF have been tagged to permit screen reader access.>
It could be more general, such as "AT access" or "to ensure accessibility". Other AT that might need it would be read-aloud tools, for example.
An untagged PDF presents no content to a screen reader.
It should still give you access, no? There's just no telling how the content will be read out or the structures interpreted. It all depends on how the text is laid out and how the PDF viewer tries to read that into a structure on its own.
The focus on navigation is a bit misleading, though, as this gets down to the logical reading order potentially being a mess when the tagging has to be inferred.
An untagged PDF has literally no readable text for AT.
Ok, you can run the tagging service inside Acrobat to do your own conversion, and then, yes, it's a question of reading order quality, navigation quality, image and heading tags possibly wrong, etc. But if you are using Preview or some other PDF reading tool, you won't have that option and you won't have any content.
There is metadata available if it conforms to PDF UA, but one will be hard pressed to find many with this metadata. However, it should be mentioned. Sorry if this is already there.
Hello All,
I want to understand and to confirm this "meta itemprop="accessibilityFeature" content="taggedPDF"" properties for metadata in PDF.
My PDF (which is hosted online for user to download) is auto-tagged using indesign software, but is not completely following the semantic tagging structure and there are some issues in the reading order as well.
Should i still use this metadata field "taggedPDF" in my accessibility file for this pdf document? Will this comply with EAA requirement?
kindly advise urgently on this topic and question.
IMO autotagging approaches need to be confirmed by a person. We have found that many of the tags need to be adjusted. I would say that the metadata should not be marked that it conforms to accessibility standards.
We should probably add mention that a file needs to conform to PDF/UA to the definition.
Yes, referencing the PDF/UA standard would be the most objective way to say "it must be accessible to use this metadata term".
As per daisy, definition of taggedpdf -
"The taggedPDF value is used to indicate that a PDF includes accessibility tags to improve its readability.
Accessibility tags are a way of labeling the components of a PDF to try and retain some of the semantics that are otherwise only visually present."
Therefore, if my pdf is tagged with partial semantic tags utilized/included in tag tree, does it meet this metadata value?
On Mon, 16 Dec, 2024, 8:44 pm Matt Garrish, @.***> wrote:
Reopened #40 https://github.com/w3c/a11y-discov-vocab/issues/40.
— Reply to this email directly, view it on GitHub https://github.com/w3c/a11y-discov-vocab/issues/40#event-15672029350, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDJU2DKVSNIFFQNKPXCX5YL2F3U5TAVCNFSM6AAAAABTVHODPOVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJVGY3TEMBSHEZTKMA . You are receiving this because you commented.Message ID: @.***>
The accessibility metadata is for making accessibility claims, and a PDF with incomplete or incorrect tagging is unlikely to meet any accessibility standard. If you can justify that your PDF fully meets EAA requirements, including for tagging, with its known problems, then go ahead, but it's not a risk I'd be willing to take in a jurisdiction with legislated accessibility requirements.
Yes you are absolutely correct. What if i use this "taggedpdf" value included in <meta.. field, and also mentioned its deficiencies under "accessibilitySummary" field, example like "This hosted online pdf is auto-tagged pdf using composition platform, without semantic structuring and fully accessible".
The end user will upfront be aware of what he/she is going to use or get from this product. This information will be retrieved from the "accessibility summary" that included the details of the product.
Does that make sense?
On Mon, 16 Dec, 2024, 10:40 pm Matt Garrish, @.***> wrote:
The accessibility metadata is for making accessibility claims, and a PDF with incomplete or incorrect tagging is unlikely to meet any accessibility standard. If you can justify that your PDF fully meets EAA requirements, including for tagging, with its known problems, then go ahead, but it's not a risk I'd be willing to take in a jurisdiction with legislated accessibility requirements.
— Reply to this email directly, view it on GitHub https://github.com/w3c/a11y-discov-vocab/issues/40#issuecomment-2546184178, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDJU2DKVCA6M2GXANU3WQEL2F4CRDAVCNFSM6AAAAABTVHODPOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBWGE4DIMJXHA . You are receiving this because you commented.Message ID: @.***>
You write: without semantic structuring and fully accessible".
Some of the semantic structing if for headings. Without heading navigation and other features, I don't think you can say fully accessible.
Thanks for the tip! Yes, headings are appropriately tagged as h1, h2 etc. but other features like list etc. are auto-tagged as paragraph.
In that case, should we consider that approach?
On Mon, 16 Dec, 2024, 11:20 pm George, @.***> wrote:
You write: without semantic structuring and fully accessible".
Some of the semantic structing if for headings. Without heading navigation and other features, I don't think you can say fully accessible.
— Reply to this email directly, view it on GitHub https://github.com/w3c/a11y-discov-vocab/issues/40#issuecomment-2546271704, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDJU2DNK5SLTZV4XB4J3RKD2F4HFVAVCNFSM6AAAAABTVHODPOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBWGI3TCNZQGQ . You are receiving this because you commented.Message ID: @.***>
In the accessibility summary, you are free to add what you want. Letting people know that there are headings tagged properly is good.
There are a ton of other features and I don't know what you can say. What about MathML? Tables properly tagged, etc?
Thats bring me to thought about another question. As per EAA guidelines and requirements, any digital product should be comply for full accessibility as per wcag 2.1 standards or later.
If online PDF of any article or issue that are not made accessible or just partial accessible, should it be removed from online edition? Or, should we notified its deficiencies online in the accessibility statement but still it will be available online for European market? Any guidance would be appreciated.
On Mon, 16 Dec, 2024, 11:53 pm George, @.***> wrote:
In the accessibility summary, you are free to add what you want. Letting people know that there are headings tagged properly is good.
There are a ton of other features and I don't know what you can say. What about MathML? Tables properly tagged, etc?
— Reply to this email directly, view it on GitHub https://github.com/w3c/a11y-discov-vocab/issues/40#issuecomment-2546338016, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDJU2DMQMH4Q2UHKJXHSE732F4LDDAVCNFSM6AAAAABTVHODPOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBWGMZTQMBRGY . You are receiving this because you commented.Message ID: @.***>
Thanks for kind sharing your thoughts on this topic.
On Tue, 17 Dec, 2024, 4:29 am Matt Garrish, @.***> wrote:
Sorry, that's a question for the lawyers.
— Reply to this email directly, view it on GitHub https://github.com/w3c/a11y-discov-vocab/issues/40#issuecomment-2547055725, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDJU2DKDG4Z3YB5KI7NGZ3L2F5LN5AVCNFSM6AAAAABTVHODPOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBXGA2TKNZSGU . You are receiving this because you commented.Message ID: @.***>