Re: Webex Accessibility Regression


David Goldfield
 

To add to what has already been said I have heard that automated testing only detects around 30% of all possible issues. In addition, while I agree that it would be ideal as well as nice if a company would perform manual Web testing with at least two screen readers it’s more important that their code complies with the WCAG standards, at least with the goal of achieving 2.1 AA compliance. If that goal can be met or at least if you can strive to achieve that goal then your site will work with any of today’s screen readers. Sure, there’s always a chance that a particular screen reader may, for example, temporarily have a bug where certain types of ARIA might not be properly handled. However, in a case like that that’s up to the screen reader developer and not the corporate Web development team to address. To be fair screen readers do address issues like this and that’s as it should be. In other words if your Web site is using proper code with good semantic markup then the developer has done his/her job. At that point if screen reader A works with it but screen reader B does not and if it’s a bug in that particular screen reader the corporate developer shouldn’t then be expected to write a hacky workaround to accommodate screen reader B’s bug. If that were to happen and screen reader B then addresses the bug, which screen reader B’s developers likely will if they become aware of said bug, the hacky workaround may actually wind up making things worse. Therefore, while it would be ideal for a QA team to perform manual testing with NVDA, JAWS, Narrator, Supernova, VoiceOver on Mac, VoiceOver on iOS and Talkback on Android the team might not have that amount of bandwidth. If they only test with just one screen reader I’m both impressed as well as satisfied. I’m speaking as a product manager who works on an accessibility team at a media and technology company and so I think I’m looking at this issue with what I hope is a balanced perspective.

 

 

David Goldfield,

Blindness Assistive Technology Specialist

JAWS Certified, 2019

Subscribe to the Tech-VI announcement list to receive emails regarding news and events in the blindness assistive technology field.

Email: tech-vi+subscribe@groups.io

 

www.DavidGoldfield.org

 

 

 

From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Brian Vogel
Sent: Friday, June 25, 2021 10:52 AM
To: nvda@nvda.groups.io
Subject: Re: [nvda] Webex Accessibility Regression

 

On Fri, Jun 25, 2021 at 03:46 AM, Sean Murphy wrote:

I like to inform you that Cisco does not test with NVDA.  When I work for them a couple years ago in their accessibility group I was the only one who used NVDA for testing.  You need to raise this with the accessibility@... mailer and really push the point.

-
And as much as I hate to be the naysayer on this, the NVDA group, the question is:  What point?

You can ask and advocate for testing with NVDA, and I think that would be a fine thing to advocate for, but no company is obligated to test specifically with NVDA, or specifically with JAWS, or specifically with Narrator.  Most companies are going to pick one, and only one, screen reader to test with, and be lucky if they have individuals truly competent with a screen reader if they're not a large enterprise (Cisco is, so they should have competent screen reader testers).

But I would not be surprised if even screen reader testing is now largely automated like a lot of other testing is, and "success" dictated by whether analysis of the speech logs has what they expected to be in them actually be in them.

The use of human testers (much like humans that do almost any sort of job) has been on the decrease even when I was still working in software development in the 1990s.  And more's the pity, because as Quentin recently pointed out on this very group, what humans actually do when faced with a given novel situation is very often very much at odds with what the developers presumed they would do and should do.  Determining how an actual person interacts with your software, including how they do things you never, ever thought anyone would do, is at the very heart of through testing.  And it happens far too little.

But I would never hold my breath that any company is doing accessibility testing with two screen readers, let alone three (or more).
--

Brian - Windows 10, 64-Bit, Version 21H1, Build 19043  

I do not understand why some seek to separate a person from their actions.  The self is composed of an individual’s thoughts, actions, and expression, which are contained in and actuated by the body.  What you do and say is the clearest indicator of who you are.

      ~ Brian Vogel

 

Join nvda@nvda.groups.io to automatically receive all group messages.