Friday, September 3, 2010

Why Can't Microsoft Build a Screen Reader into Windows?

As Windows screen reader users will know, there is no screen reader included in Windows. Instead, users requiring a screen reader must obtain and install a third party product. Yes, there is Microsoft Narrator, but even Microsoft know that this is hardly worthy of the name "screen reader". :)

A few years ago, Apple revolutionised the accessibility industry by building a fully fledged screen reader, VoiceOver, right into Mac OS X. Ever since, many have asked why Microsoft can't do the same for Windows. Many are angry with Microsoft for this continued lack of built-in accessibility, some using it as support for the "why Apple is better than Microsoft" argument.

Here's some food for thought. I'm not sure Microsoft could do this even if they wanted to; their hands are probably tied in a legal sense. If they did, they could very likely be sued by assistive technology vendors for anti-competitive conduct, just as they have been sued several times concerning their bundling of Internet Explorer with Windows. Once again, Apple don't have to be concerned with this because there wasn't an existing screen reader on Mac OS X and they don't have the dominant position in the market.

I have no evidence for this argument. Perhaps I'm wrong, but history suggests that it is highly likely that I'm not.

Even as one of the lead developers of NVDA, I'm first and foremost a blind user who wants the best possible access, both for myself and other blind users. As such, I would very much welcome a screen reader built into Windows. Competition is good. A built-in screen reader doesn't mean that other screen readers can't exist. If the built-in solution were good enough, then there would be no need for NVDA to exist. If it weren't, NVDA would drive accessibility to improve through innovation and competition.


  1. You could be correct about the legal argument, however this hasn't stopped Microsoft from doing just about everything else from including a browser (they still do) to adding anti-virus more recently. So I think they could add a screen reader if they wanted too. Its more likely they dont' see a need since there are third-party applications available, and let's face it, a screen reader by Microsoft for Microsoft isn't going to be much of a money maker.
    I have long suspected Apple did not originally create VoiceOver entirely because they were feeling generous. When OutSpoken for Mac was in existence, Apple did not make a screen reader. When OutSpoken died, I think this forced Apple's hand if they wanted to maintain some of their government and education-based sales. Now they have learned that being the leader in included accessibility is also a marketting advantage so it looks like everyone (most anyway--se below) are happy.
    I wonder if you have seen one of GW Micro's latest blog postings in which they summarize their thoughts on the future of screen readers? I agree with a lot of GW Micro's thoughts and positions and applaud them. However they did have a statement about included screen readers which troubles me. Quoting: "GW Micro believes that having a free screen reader as part of the operating system does a disservice to Blind computer users." Their main concern/complaint is that they feel it reduces competition. I wonder though if instead in the case of Windows which has a much larger market share, and established screen readers, if more competition would force them to make even better products? I am sure this is also a goal of NVDA. I wonder if they feel the same way about NVDA for that matter, that free/low cost is bad? I find it troubling that we should be expected to accept that to use our PC we should pay out an additional $1,000-1,500 USD on top of the hardware and other software. I realize that the manufactuers need to earn a living as well, but we need some balnace.
    Thanks for continuing to work on NVDA and give the other vendors some competition. I hope between all we end up with the best most accessible and productive solution that there can be.

  2. Travis: Thanks for your thoughts.

    I agree that this argument hasn't stopped Microsoft in the past. However, from their perspective, it isn't worth it for them to fight and potentially lose a legal battle over something which will be used by a relatively small percentage of their users. In contrast, losing money over inclusion of IE is probably "worth it" for them, as it is used by almost all of their users. The same is probably true for anti-virus.

    I saw the post from GW Micro to which you are referring. That particular quote troubles me also. I would argue that on the contrary, an in-built screen reader cannot possibly be a disservice. As I noted in my post, other screen readers will then drive accessibility to improve through innovation and competition.

  3. I know it's only a small part of using a computer, but would a built-in screenreader in the browser be an improvement? Certainly not for a blind user, but for other reading impaired users? I keep thinking about the implications of the Audio API in Firefox 4 where it is possible to generate sounds with JavaScript, and probably the other way round as well: voice input, though I haven't seen a demo for this yet. How would you start building a text-to-speech JavaScript library?

  4. There are ways in which MS could offer an improved screen reader while assuring consumers have options. Imagine purchasing a new computer and starting up the machine when a page appears asking if speech or magnification is required. If not, the speech is turned off. If so, the user is presented with an additional page allowing them to select from a list of providers including an improved Microsoft product. This list would provide links to NVDA, Thunder, JFW GW... The user will have the option to choose to install which ever screen reader they choose.

    P.S Thanks for your work on NVDA - it has gotten me out of jams on many occasions!

  5. I have several concerns.

    * Above all I would question the motive.

    Why would a multibillion dollar company that owns something like 95% of the market' suddenly feel compelled to improve integrated accessibility features? Out of the goodness of its heart? There's definitely next to no prophet gains in adding a fully functional screen reader. Either way, potential losses through lawsuits and potential earnings are just marginal. In plain English: "Microsoft does not give a flying fuck". Sad as it may seem.

    * So Apple can flash the fact that they have built in accessibility features. In turn Microsoft can then give the oh so well known and well rehearsed spiel on how they've been working so hard all these years to provide the best platform for 3rd party accessibility solutions. …And you know where the argument goes from there and where it ends just to start all over again.

    * I dig NVDA man. Its system unobtrusive and is a hack of a lot more intuitive and easier to use then VO and in some cases JAWS. I'd like to see it grow and develop. Am I reading too much between the lines in assuming that your blog post basically says "Why doesn't Microsoft incorporate NVDA into Windows". Just kidding there. ;)

    *Apple accessibility features. I'm not gonna rant. I'd like to but I'm not here to hijack your blog.

    1. Universal Access Zoom. Hasn't been updated in over 8 years. Things get even better. Now some of its functionality is hosed by compatibility issues with VO.

    2. Vo. A mac user can have any screen reader as long as its VO. Looking at VO v.3 I can easily say that Apple is establishing a trend of trading functionality for more features. Crashed on me a dozen times today.

    At this point I'd rather actually see some viable 3rd party alternatives to VO. Monopoly isn't good. Especially when the product is free or integrated.

  6. As much as it would be cool to see MS make a decent attempt at a screen reader, thenthey would need to try and make it work with any app, not just the closed-source MS versions. In other words, Firefox, Thunderbird, and so on. There's still a market for screenreaders even if MS develops one, because it won't be mature and feature-rich and stable enough to be a replacement any time soon. I could be wrong, but that's how things work most of the time. They implement something that works fundimentally works, then slowly roll it out over time.

    Honestly, if MS were to come out with a built-in screenreader, I would probabyl still use NVDA if it were supported in the new OS, just because it's what I've been using for the past four years and I really don't want to learn something new.

    GW says having MS build a screenreader into windows is a disservice? No, they just want to stay in business. At the end of the day, a lot of the AT industry is about making a buck, as are most businesses.

    Honestly, if MS can make a new a11y solution that can be called when installing Windows, now that would be something to celebrate!

  7. As a former softie I have to admit I'm very disappointed in Microsoft failing to put this in place. In my defense in the brief time I was there they never bothered to take my feedback *grin*.

    I have an online screen reader accessible text rpg and its really frustrating listening to users talk about the sad state of assistive technology and voice synthesis on readers. NVDA is a huge improvement on what was there before but there's such a long way for us to go in this area.

    By the way if you dont mind my asking what does a dev rig for the seeing impaired look like. What technology stack do you use. Do you make language choices around screen reader accessibility?