E-mail List Archives
Re: Accessibility assessment of pdfs.
From: Duff Johnson
Date: Oct 25, 2012 9:49AM
- Next message: Sarah Horton: "Re: Fwd: Web Accessibility position at Stanford"
- Previous message: John Martyn: "Spotify Blind www.SpotifyBlind.com"
- Next message in Thread: None
- Previous message in Thread: Steve Green: "Re: Accessibility assessment of pdfs."
- View all messages in this Thread
Steve,
In general, I agree with you, but want to add a few clarifications.
On Oct 24, 2012, at 6:11 PM, Steve Green wrote:
> As with all automated testing, the CommonLook tool can only perform some of
> the necessary tests with certainty. For instance it may be able to tell if a
> document does not contain any headings at all, but if there are heading tags
> it can't tell if they have been applied properly. It can check that images
> have alternate text but it cannot tell if that text is appropriate.
This is totally true - but I'd just add (since people seem to want to talk about our product)
1) CommonLook Clarity allows for the use of Regular Expressions to *help* identify "bad" alt. text. Examples include: text that's obviously a file-name (because it ends in ".jpg" or similar), or alt. text that's less than two words.
2) The product recognizes and implements the concept of "User Verification Required" in its workflows. For alternative text, to take that example, CommonLook Clarity defaults to assuming that "User verification is required" in order to assess the quality of any alternative text.
> Where you have a lot of legacy PDFs, a pragmatic approach might be to test
> them all with an automated tool and test a small sample manually to get a
> general assessment of the quality of tagging (if indeed they are tagged).
> You will then know how good they could be made and if they are fixable at
> all - sometimes they are not, depending on how they were authored.
CommonLook Clarity was designed to facilitate precisely these sorts of workflows, not by assuming that automated checks can resolve the question of accessibility, but simply by making it possible to drive prioritization with data.
Example:
If a PDF includes data tables and if the PDF is tagged AND
If the tables all include column and/or row headers AND
If all the images have alt. text that passes automated checks AND
There are no untagged elements in the PDF AND
There's no use of color, no scripts, etc, I could go on)
THEN it's highly likely that the PDF is not only tagged, but already quality-controlled, at least at some level. This sort of information really helps with prioritization, and also with measuring results.
Ok end of Marketing Speech! We now return to your regally scheduled program!
I hope to meet (some of) you in San Francisco next Monday morning!
Best regards,
Duff Johnson
President, NetCentric US (Makers of CommonLook)
ISO 32000 Intl. Project Co-Leader, US Chair
ISO 14289 US Chair
PDF Association Vice-Chair
Office: +1 617 401 8140
Mobile: +1 617 283 4226
<EMAIL REMOVED>
www.net-centric.com
This e-mail message is confidential, may be privileged and is intended for the exclusive use of the addressee(s). Any other person is strictly prohibited from disclosing, distributing or reproducing it. If the addressee(s) cannot be reached or is unknown to you, please inform the sender by return e-mail immediately and delete this e-mail message and destroy all copies.
- Next message: Sarah Horton: "Re: Fwd: Web Accessibility position at Stanford"
- Previous message: John Martyn: "Spotify Blind www.SpotifyBlind.com"
- Next message in Thread: None
- Previous message in Thread: Steve Green: "Re: Accessibility assessment of pdfs."
- View all messages in this Thread