Diversity in Silicon Valley Should Include Disabilities

Everyone's talking about efforts across Silicon Valley to spread diversity in the workforce. Inclusion is a great thing, but one has to wonder why disability is hardly touted as a characteristic worth recruiting.

Recent Headlines

Silicon Valley takes on diversity issue

Silicon Valley has a diversity problem, a contentious issue that has come into sharper focus in recent months as tech firms sheepishly released updates on their hiring of minorities. The companies have pledged to do better. Many point to the talent pipeline as one of the main culprits. They’d hire if they could, but not enough black and Hispanic students are pursuing computer science degrees, they say. Read more.

Silicon Valley, Here's That List of Boss Ladies You Needed

This week Silicon Valley veteran Sukhinder Singh Cassidy unveiled The Boardlist, a database to help startup CEOs find qualified women to appoint to their private boards. The list includes more than 600 female leaders compiled by more than 50 high-profile tech industry insiders and venture capitalists (both men and women) in the San Francisco Bay Area. ... The Boardlist aims to provide an easier system for the discovery and matching of CEOs and pre-vetted female board candidates, letting users of the database (which is in beta for now), dig into categories like functional expertise, industry, company stage, and board experience. Read more.

Hey, Silicon Valley, give your nanny's child an internship

Hey, Silicon Valley CEO or director of engineering. You say you care about diversity in tech. So here's a novel idea: Give an internship to the daughter of your housecleaner. Or it might be a scholarship for the nephew of your nanny. Or a summer job to the grandchild of the office janitor, security guard or bus driver. Read more.

Obama’s top tech adviser takes fight for Silicon Valley diversity to Washington

The problem has taken new urgency with another round of embarrassing workforce data released in recent days by tech's leading firms. Despite splashy promises to pour more resources and attention into recruiting women and minorities, the work forces at Google, Yahoo, Facebook and Apple haven't changed. Women make up less than 20 percent of employees, Hispanics hold about 5 percent of jobs, and blacks are particularly under-represented at below 2 percent of those work forces. Read more.

Possible Explanations

There must be a logical reason for the lack of disability representation across the tech sector. Silicon Valley is home to leading companies producing some of the most innovative products designed by some of the world's brightest innovators.

It's not lack of awareness. Google support's the NFB's Scholarship Program and must see the potential a blind software developer could offer. The Google Impact Challenge seeks to "propel even more technological innovation that will create meaningful and positive change for people with disabilities." Clearly the company appreciates the challenges persons with disabilities face and wants to invest funds toward solutions.

It's not lack of confidence. Apple's position on accessibility proclaims: We’ve done everything possible to make anything possible, and it's a fine tagline befitting a company that continues to enjoy dominance over accessible products out of the box. One hopes Apple has ensured equal productivity across its own apps not because it was forced to do so, as some still speculate, but rather, because it believes persons with disabilities are equal contributors.

Perhaps it's a lack of applicants? Consider the following from the University of Minnesota:

When considering students with disabilities, their participation in postsecondary education is lower than their representation in the U.S. population, and significantly lower in comparison to their peers without disabilities. For students with disabilities who are enrolled in postsecondary education, only 11% of those students in undergraduate programs are pursuing STEM degrees (Burrelli, 2007). This number drops to 7% in graduate programs in STEM, with only 1% earning a doctorate degree in STEM (Commission on Professionals in Science and Technology, 2007) and only 4.8% entering the science and engineering workforce (Burrelli & Falkenheim, 2011). This low number indicates a need to encourage and support students with disabilities to enter and complete postsecondary education in STEM so they can compete in this growing job market. Read more.

Possible Solutions

If it is a question of insufficient candidates, tech leaders should consider battling the problem at its root. Students with disabilities are still inhibited by the same technology they wish to revolutionize. Modern technology has done great things to advance education. Unfortunately, modern technology also spawns more exercises in advocacy than exercises in learning. Tech leaders could be more proactive about ensuring the core accessibility of products widely adopted by the education sector. For instance, it's well past time for Amazon to make its Kindles accessible.

Silicon Valley should consider dedicated recruitment campaigns to identify qualified persons with disabilities. The NFB has begun incorporating job fairs into its national conventions and Washington Seminars. It would be great to see the number of mainstream employers outnumber employers in the blindness field as part of these small but promising gatherings. Strategic partnerships with cross-disability groups could inspire a collective effort to identify, groom, and ultimately employ competitive candidates. Of course internship opportunities provide one way for employers to cultivate ideal staff, not to mention a perfect means to observe accessibility and productivity comingling in a real world scenario.

Taking the recruitment idea one step further, it's not unusual for larger tech companies to partner with universities to undertake specific projects or research studies. In September 2014 Apple announced partnerships with Duke and Stanford on medical trials for HealthKit. One month later Google announced a partnership with Oxford to make advances in artificial intelligence. Perhaps it's time to identify a state school for the blind or a rehabilitation training center that could yield both potential candidates as well as create an incubator for technologies benefiting persons with disabilities.

Are there other solutions? Absolutely. These are only a start to what could be a fruitful dialogue.

Final Thoughts

The 25th anniversary of the ADA saw a chorus of voices vowing to make technology more accessible for persons with disabilities. That's a noteworthy step and commendable of the coalition participants. Yet, it is not enough to train existing engineers in accessibility best practices. Silicon Valley ought to move from a passive hiring of employees with disabilities to a dedicated effort to identify qualified candidates. Persons with disabilities want to be given the opportunity to contribute, not just consume.

Diversity is a hot button issue. The technology sector is being asked to extend itself beyond a traditional workforce to one that encompasses people who have historically been overlooked. Persons with disabilities have been overlooked. It's time to reverse the trend, acknowledge disabilities are a part of a diverse working environment and make them as much of a priority in building the next generation of Silicon Valley leaders as any other factor.

Did you find this post interesting? Follow me on Twitter @ScribblingJoe or sign up for updates straight to your inbox. No spam!

Enter your email address:



So I am blind, work in Silicon valley; have done so since 1980 in product support, tech writing and as a software engineer. But I've spent these last fifteen years at a community college, in Cupertino, supporting the technical needs of disabled students and puttering with access technology.

And the way I see it, the biggest barrier to hiring qualified blind people is technology access, even more so than in previous decades.

When DOS was king, there were a few programs that didn't play nice with the screen reader. But lots of stuff was accessible enough you could use it. After all, the screen was a text-based thing, and advanced screen readers like ASAP could intelligently render the information.

After Windows screen reading matured, most software again became accessible. There was always a file menu to the left of an edit menu. There was a mouse cursor you could move to locate, label and click on toolbar icons. There were panes and status bars, and all the usual dialog box controls, but they were predictable, familiar with well-established keystrokes for accessing them. You could download some shareware and have a pretty good chance of it working with your screen reader. You could walk in to the job and its database, customer contact application and email software probably would work with a screen reader. Those were the days when I was downsized many times due to all the corporate upheavals in Silicon Valley and I always landed on my feet. Sometimes I used console Linux and more often Windows, but I could write a few JAWS scripts to patch up any minor access issues I encountered.

About the time I got this current job, where I am sheltered like many blind people from the realities of the real world, I noticed a paradigm shift that seriously affects access. It's the elephant in the room that GW Micro, Freedom Scientific, Dolphin and even Sarotek don't talk about. It's the problem with the disappearance of the standard interface.

It used to be that if a Windows package didn't work with your screen reader, you were advised to find a comparable package that used standard controls.

Since when have you downloaded, or in some other way confronted a program that sticks with standard controls; those beloved tree views, drop-down lists and check boxes of the past? Now most software is created with libraries that adhere to no Windows standard at all. There is no longer routine keystroke access to most elements. Most screens are a morasse of graphics and unlabeled buttons. The menus are gone, sub-windows float everywhere, and you can't depend on any software interface behaving like any other. Take for example, those programs that let your computer burn CDS, bundled with most every drive. Take malware scanners. Take everything from accounting software to media players. It's a long hard struggle to even figure out if they can be made accessible, let alone trying to use them.

And most in-house software on which companies rely is even worse. The developer probably knows nothing about making an application accessible. It might have graphical controls and edit fields that no screen reader can see. Look at Quicken. The Windows 98 version worked fine without scripting. Today, there is this mass of publicity about how Freedom Scientific and intuit have temed up to improve the accessibility of Quickbooks. Nobody is going to team up with your average employer to fix his inaccessible software.

Yet screen reader manufacturers are busy adding useless features; more than a speech and sounds manager, I need a tool that will analyze an application and help me figure out whether and how to use it. More than extra voices, I need extra tools for adding keystrokes quickly to clickable elements.

And in to this newly standards-free chaos comes web 2.0. Google and Yahoo patch up their sites with a bit of keystroke access. But have you tried administering an online group with a screen reader lately? How about joining, or leaving a facebook group? What do you do when a web application's help instructs you to drag an element and drop it somewhere or click on the blue arrow?

More and more employers are using web-based applications, but screen readers still have no perfect way to deal with pages that refresh certain parts. When you arrow through a virtual buffer and jump between elements, you still have completely lost touch with the visual layout: those geographical relationships elements have to each other. You don't know where the menyu on the left and the buttons on the right begin. You don't know what's highlighted or shown in red. If something pops up while you interact with the page, the screen reader often puts it at the end of the virtual buffer even though visually it's front and center on the page. And half the time, the screen reader fails to notice that the page has changed!

You toggle in and out of forms mode, or browse mode, and if you tab too far forward or back, you land in the browser interface, interacting with that instead of the important interface for your web-based app. Even a simple web application, like a calorie counter (see myfitnesspal.com) isn't as easy for a screen reader user to work with as most software which ran under Windows 98. And what's worse, since the demise of XP the mouse cursor is gone. Something with the internals of Windows 7 and above made it impossible for screen readers to freely move the mouse everywhere. Screen review is a thing of the past; you are stuck working exclusively with the screen reader's rendering of the document object model.

Last week, JAWS and Outlook which were humming along together, not entirely glitch-free, but at least predictably, suddenly stopped working. JAWS would no longer read my email. Mail would open but it would appear empty.

After much research, I learned an add-on, supplied by Microsoft, on which JAWS depended had failed. I had to remove and re-install it and everything worked again. JAWS, I knew, but hadn't really thought about before doesn't read the screen any longer. It depends on Microsoft's exposure of the Office object model to work. It depends on components of office, which through scripts, it's able to acquire information about Microsoft Office, and its current state. So for example, if you ask JAWS to go to the next sentence, it asks Microsoft Office where the next sentence is and then tells Microsoft Office to take the focus there and then it reads the current sentence. Now that's great if you are using Office. But as screen readers focus more and more on making particular applications behave as if they were self-voicing, the rest of the big wide world of Windows software is forgotten.

And all that other software is what your future employer will expect you to be able to access. If you are talking to customers on the phone, you'll need the client database and remote control software for taking over their computer, if you do tech support. If you are in sales there is order management and contact tracking software too, and it might not be something generic and mostly accessible like Excel or Outlook.

And heaven help you if you need to organize meetings, manage another person's calendar, alter a sighted person's workbook, or create a newsletter. It's no longer easy working with the supposedly accessible Microsoft office. The keystrokes microsoft has cobbled together are from Office 2003 and below; you can tab around the ribbon and get completely lost. Ever tried to orient a document for printing in landscape? The steps are accessible but there is no way to easily confirm that a document is already oriented in landscape because it's a funky ribbon thing. And another gripe I have with the ribbon is the non-standard way you select things; sometimes it's space, sometimes enter. How do you know when to press what? Mouse users click, but keyboard users need standardization. Even the elements on the ribbon of a supposedly accessible suite behave differently compared to each other.

Last week I was trying to follow an article in PCWorld which showed you how to set up "Quick Steps" in Outlook. It was click on this, pull down that, type this in to an edit field. and I could not translate those instructions in to keystrokes. There were edit fields but unlabeled so I didn't know which was which. I couldn't right click on some elements because I couldn't get my mouse cursor there and the applications key wouldn't work even when I was focused on the proper element. I've gone through some basic training, but any advanced features are sill out of my reach because I can't figure out how to get to them with the keyboard.

And forget the free Office alternatives like OpenOffice and LibreOffice. Nobody's working to improve their accessibility. The big problem is that the underlying tools used to build today's applications don't have accessible components. The controls they supply for the user to interact with themselves were not designed to be accessible.

Truthfully, I'm scared to venture out in to the real world again, where I'd be confronted with questions about how I could perform the job. In the Windows 98 days, I confidently dragged a laptop to the interview and quickly squashed any employer's fears. Today I'm pretty sure I won't be able to access his software.

Add new comment