Skip to content

Dr Scott Hollier - Digital Access Specialist Posts

W4A2017 & WWW2017 accessibility conference highlights

After three years of being involved with the logistics to bring WWW2017 and W4A2017 to my home city of Perth, Australia, the Web for All (W4A) 2017 conference arrived in April with a host of great papers. Here’s a few of my personal highlights from W4A2017 and the W3C accessibility track at WWW2017.  W4A2017 conferenceWith this year’s conference focusing on accessibility and work, the first day kicked off with an excellent keynote by Alastair McEwin, disability commissioner for the Australian Human Rights Commission.   The presentation titled ‘Working together: technology as the foundation for better employment outcomes for people with disability’ noted that with 4.2 million Australians having some form of permanent disability, more needs to be done to improve employment opportunities.  In particular, Alastair explained that employers are often not prepared for people with disabilities which can lead to misconceptions about staff needs, and explained that the benefits go far beyond work itself as engagement in the workforce also means inclusion in society. Key recommendations discussed that could improve employment consisted of addressing inaccessible information in the workplace, and the potential benefit of the NDIS. Alastair highlighted that currently Australia has quite a low employment rate of people with disabilities when compared to comparable countries which is particularly concerning, so action is needed quickly.  

Jeffrey Bigham presented a paper titled ‘Scopist: Building a Skill Ladder into Crowd Transcription’.  The focus was on Stenography and the production of captions. It was explained that Stenographers use a special keyboard to reach 300 words per minute. While the equipment can be expensive, tools such as OpenSteno can map the commands onto a traditional QWERTY keyboard.  The challenge is to be able to swap between the two in real-time. The paper focused on how the required keyboard could be predicted meaning that there was no need to actually swap as the software would be intuitive enough to pick it up, and hence speeding up the process.

Another presentation I found really interesting was William Grussenmeyer who focused on the topic ‘Evaluating the Accessibility of the Job Search and Interview Process for People who are Blind and Visually Impaired’. This looked at a study of people who are blind or vision impaired and the main issues they had around the start-to-finish processes of finding employment. Interestingly two of the key issues raised were inaccessible PDFs for applying for jobs, and inaccessible on-boarding content that made it difficult to get set up in a new job.  I’ve had similar experiences personally whereby I have been able to do the job itself but completing the initial setup of a job such as completing the government form relating to personal details and tax information has only been available on paper and needing help with that.  This was picked up in the key note on Day 2 by David Masters from Microsoft who discussed ‘Microsoft’s Journey towards Inclusion’. David made the point that Microsoft has largely addressed its on-boarding issues and ensures that the recruitment and employment processes are accessible from start to finish which is encouraging. Rosemary Spark at W4A2017Rosemary Spark continued the discussion in the topic ‘Accessibility to Work from Home for the Disabled: The Need for a Shift in Management Style’ which made the point that a flexible workforce is beneficial to people with disabilities and despite the general perception that people don’t work as hard at home, the reality is that it actually improves productivity and performance.

 Outside of the topic of work, Keith Fitzpatrick from the City of Cockburn provided a great insight into the processes of local government in creating an accessible website. There was also a great paper on the Internet of Things authored by a number of people from W3C WAI.

 A role I shared with A/Prof Justin Brown for W4A this year was as co-chair of the Google Doctoral Consortium, and it was great to see the winner Tahani Alahmadi from Griffith University present her paper titled ‘A multi-method evaluation of university website accessibility: Foregrounding user-centred design, mining source code and using a quantitative metric’.

 There was also an excellent presentation by Kevin Carey who offered some interesting food for thought on the pursuit of access and much excitement in the Paciello Group challenge. This year the Challenge provided an opportunity to test out some products that had been created around research including a method to navigate website using only eye tracking and a clever use of simplifying words in e-mail to make it easier for people with cognitive disabilities to read.

In addition to my role as co-chair for the Google Doctoral Consortium paper, I was also given the great honour to deliver the William Loughborough After-Dinner address, focusing on the topic Technology, education and access: a ‘fair go’. With 100 people enjoying the beautiful warm autumn night overlooking the Indian Ocean, it was a great way to share a meal and a lot of fun was had by all. I’ll post a separate update with links to the video once it becomes available. With W4A wrapping up after three days, the accessibility-related conference was complete but some accessibity presentations remained.  On Wednesday 5 April the main WWW2017 conference had a W3C track which included three disability-related presentations.  The first topic was ‘Digital Service Standard for the Australia Government’ by Andrew Arch from the Digital Transformation Agency, followed by ‘A conversation without barriers’ by Marie Johnson from the National Disability Insurance Scheme. I rounded out the session discussing the future W3C standards of WCAG 2.1 and Silver that are currently in development.  With approximately 50 attendees in the room it was great to see accessibility issues get some mainstream traction.

 With W4A and WWW moving to France in 2018 it will be exciting to see what new and innovative research projects are presented as the great work being done in this space continues.

Screen readers and web browsers – what’s the best pairing for testing?

Testing web content for accessibility can be a difficult task, but fortunately there’s some great guidance from W3C in the form of the Website Accessibility Conformance Evaluation Methodology (WCAG-EM) 1.0. However, many people get stuck at Step 1 – defining the evaluation scope.  While setting a conformance target is generally straightforward such as WCAG 2.0 Level AA, it’s much harder to decide baseline-related issues such as which assistive technologies should be used for testing against the WCAG 2.0 Success Criteria and in which browsers the content should be displayed.  Recently encouraged by a student question in the course I teach, I’ve put together some information on the topic for you to consider.  

Before outlining my thoughts though there’s two important things to keep in mind: firstly, if you are doing a web audit for a client, the answer to ‘what browser and screen readers should I use to form a baseline?’ should generally be answered as ‘whatever the client wants’.  You’re certainly welcome to point them to this post if it helps to explain your point of view about accuracy of testing, but I appreciate there may be very specific reasons why the client wants to test out, say, JAWS with Chrome.  While some pairings in industry may be unusual, there’s often a method to the madness especially in locked-down enterprise-level systems where the Standard Operating Environment (SOE) gets changed for no one. Secondly, keep in mind that this assessment is a point in time and is purely my opinion based on working in the industry and many anecdotal conversations with screen reader users, so this can change quickly.

With that in mind, here are my recommended pairings for screen readers and web browsers.

1.    NVDA and Mozilla Firefox

If you want to test in a traditional desktop environment on a Windows platform, it’s hard to go past NVDA with Firefox.  NVDA is a fantastic screen reader with the developers at NV Access working hard to ensure the screen reader is up-to-date with great support across a number of recent Windows versions.  In addition, updates tend to come out very quickly ensuring that it caters for changes to web standards and best of all, it’s free. 

 The benefits of NVDA are also in many ways the benefits of Firefox. The browser focuses heavily on standards-compliance with a legacy of effective support in this area and it plays very well with NVDA. As Firefox is also updated regularly you’re well placed to use these two together to maximise your accessibility testing with a degree of certainty for the results and at no charge to your organisation.

 2.    JAWS and Microsoft Internet Explorer

While NVDA and Firefox is arguably the most accurate paring for testing, there’s no denying that JAWS remains the king of screen readers and it’s likely that organisations will want to know how things go for JAWS users as a result.  In my opinion JAWS testing should remain with Microsoft Internet Explorer despite the browser being so old that its existence in Windows 10 is barely acknowledged to the point that you probably didn’t realise it’s still there.  The reality is that there’s still a lot of screen reader users that rely on old versions of JAWS due to the cost of upgrading, and its slow upgrade cycle has caused issues for it with other more modern web browsers.  People testing websites often get frustrated with this combination as Internet Explorer is not the most standards-compliant browser, and JAWS certainly has its own quirks so the two combined may show up errors that you don’t believe most users would experience, yet this pairing is still reflective of a large number of desktop users and you may need to consider this if your organisation has JAWS users.

 3.    VoiceOver (iOS) and Safari (iOS)

 With WCAG 2.1 on the way a time is coming when testing on a mobile will be an essential part of WCAG conformance, and with that in mind it’s important to know which pairing is best for mobile devices. On iOS devices such as an iPhone or iPad, it’s a pretty easy choice – use the built-in VoiceOver screen reader with the built-in Safari browser. Both work great together and in my experience there’s no other iOS browser that comes close to VoiceOver support.

 4.    Android TalkBack and Google Chrome

As noted above, with WCAG 2.1 coming it’s likely you will need to test on mobile devices, and the best Android option is to test using the TalkBack screen reader with the Chrome browser which works very well, better than any other web browser I’ve tested.  If you have to choose only one mobile platform for testing though I’d go iOS at this point as while Android dominates the market in the general population, Apple iOS is far more popular among people who are blind or vision impaired – and I’m acknowledging this despite being primarily an Android user myself.  That said, there’s nothing wrong in using Android for testing and if you do, go for TalkBack with Chrome.

5.    ChromeVox and Google Chrome (desktop)

From a  testing perspective, you may not be aware that there’ is a little-known screen reader tucked away for users of Google Chrome called ChromeVox, and this screen reader is also found in Chromebooks. While its not commonly used by people with vision disabilities, it can be a useful pairing for testing purposes  Just be mindful that while it can help pick up issues, the overall experience is not going to be reflective of most screen reader users hence its further down this list. That said, Chrome is an excellent browser and both Chrome and ChromeVox are free. Furthermore, given the popularity of Chrome as a desktop web browser there’s a good chance it’s already on your computer.

6.    Windows 10 Narrator and Edge – in the near future

 At this point, screen reader purists are likely to start questioning my sanity by including Windows Narrator on the list paired with anything. I certainly won’t argue that Narrator in Windows 2000, XP, Vista and 7 was a terrible screen reader.  However, updates in Windows 8 and significant improvements in Windows 10 including the Braille support coming soon highlight its significant improvements.  I’ve used Narrator for testing and while it should certainly be no higher than six on this list, it has the advantage of being built-in so people are likely to have it, and Edge is a reasonably standards-compliant browser – well, significantly more so than Internet Explorer anyway!  At the moment Microsoft still recommends Narrator users in Windows 10 use Internet Explorer, but with updates to both Narrator and Edge in the insider preview of Windows 10 likely to be released later in the year.  This will become an option for testing.  Not the best option, but good enough to be an option if your machine at work is locked down like a fortress and you have no other choice.  

7.    VoiceOver (Mac) and Safari (Mac)

I’ll be the first to admit that VoiceOver and Safari on iOS are fantastic, but VoiceOver on Mac OS and Safari on the desktop is in a desperate need for a massive overhaul.  It’s unfortunate that VoiceOver on a Mac is good enough that you don’t really need another screen reader, but not good enough when its functionality is compared to other screen readers on other platforms. If you specifically have a blind user in your organisation that only uses VoiceOver on a Mac, then your best option is to test the content in Safari. I should stress that my opinion here is not because I particularly dislike VoiceOver – indeed it was the introduction of VoiceOver in Mac OS 10.4 Tiger that is credited for having the first fully-fledged screen reader in a desktop OS and that deserves respect, but its way overdue for an update.

So that’s a bit of an overview of my favoured pairings of screen readers with web browsers from a testing perspective.  Hope it helps.   

TellMe TV audio description service: hands-on

In December 2016 the TellMe TV online subscription service was established to provide audio described video content.  The service has been designed to meet the needs of people who are blind or vision impaired and offers Netflix-style online streaming of content.  As an Australian with very few audio described content options I decided to give it a go.  

The service has been established by Canadian accessibility advocate Kevin Shaw, which stated in the press release that “TellMe TV is an exciting new destination where 100 per cent of the on-demand programming, including a diverse portfolio of movies, television shows and documentaries, [are provided] in fully described video.”

As a vision impaired person located outside North America, I wasn’t sure if I’d be able to access it at all: it’s not unusual for such services to be geoblocked or content hidden from international viewers. After signing up I navigated my way through the content and found a documentary. On the plus side, it worked without getting any geoblocking message.  The downside however is the notable lack of modern content.

As promised, the service does indeed provide 100% audio described titles, meaning that for people with vision-related disabilities the more visual aspects of the videos are described by a narrator.  While the interface was a little cumbersome to use in my high contrast color scheme, the screen reader picked things up well and it was able to play the selected video. I was pleasantly surprised to see the video start playing given I’m based in Australia, and both video and audio played with no problem.

However, the biggest issue with this service at the moment is the content.  It was hard to determine if the lack of modern content was due to the good stuff being hidden from me or if there was just a very limited amount of titles at this time, but most of the movies, documentaries and services I could play were quite old and presumably available as public domain titles.  That said, if the service is able to secure the rights to the big movies and TV shows then the interface works quite well, and if the service continues to be available internationally then it has the potential to revolutionise the way in which people who are blind or vision impaired watch TV. As such, it’s my hope the big media companies will step up to support this initiative especially as audio description is hard to come by through traditional media sources.  For now though the best option here in Australia remains with Netflix and its small but growing audio described titles.

More information on the service can be found at the TellMe TV website.  The service offers a seven day free trial.

Dr Scott Hollier to give W4A2017 William Loughborough address

It is with much excitement that I’m able to share with you some fantastic news; I’ve been given the great honour and privilege to deliver the William Loughborough After-Dinner address at the Web for All (W4A) 2017 conference.

My topic, ‘Technology, education and access: a ‘fair go’ for people with disabilities’ will focus on my personal journey and on the broader benefits that education and technology can provide..  In particular, the address will focus on how key accessibility developments help to  provide a ‘fair go’ in relation to the inclusion of assistive technologies now built-in to mainstream devices, the future implications of accessibility and the wonderful dedication and hard work of accessibity professionals in their support of people with disabilities.   

Details regarding the after-dinner address and additional conference information can be found at the W4A2017 website.

Microsoft adds Braille and mono audio to Windows 10 insider preview

Microsoft has announced accessibility improvements to its latest Windows 10 build including the addition of Braille support to the Narrator screen reader and the inclusion of a mono audio option.

In a blog post by Microsoft’s Dona Sarkar, developers who are a part of the Insider Preview network can update their version of Windows 10 to version 15025 which contains the new accessibility features.  

The blog post stated that “We love getting feedback from our visually-impaired Insiders and implementing features to support your needs. It’s so important that we keep our diverse customers in mind as we co-create with you.  Today, we are excited to announce braille support for Narrator. This experience is currently in beta.”To enable the feature, Microsoft has provided the following instructions:

  • Ensure Narrator is running. Then go to Settings > Ease of Access (WIN + U) and under the Narrator settings, activate the “Download Braille” button. You will be prompted to install braille support.
  • Under Settings > Ease of Access, activate the “Enable braille” button and add a braille display. Note that USB and serial connections for the display are supported.
  • Under Settings > Ease of Access, choose the language and braille table you want to use. NOTE: There are coexistence issues with braille support and third party screen readers. Until the documentation is available, we recommend that braille be enabled for Narrator only on PCs that do not also have a third-party screen reader configured to use a braille display.

The new mono audio feature is another great accessibity feature and assists both hearing and vision impaired users.  People with a hearing impairment in one ear can benefit by ensuring that information pushed only to one audio channel is available in both channels, while vision impaired users that use a screen reader with one earpiece can also receive audio sent to both channels.   This feature can be found in the Ease of Access section once users have updated to the developer build.

Microsoft has not confirmed when the features will be available in the standard Windows 10 builds but based on development cycles its likely to arrive on most consumer devices before the end of the year.