The Need for Multiple Screen Readers

This morning I read a post on Blind Access Journal (link above) that Darrell made yesterday criticizing Freedom Scientific– a topic on which I will mostly reserve comment. One concept in Darrell’s post, though, that I find both interesting and annoying is that, to perform many full time jobs, a blind person needs to have multiple screen readers installed. On my primary work computer, I have JAWS 8.0.2107, Window-Eyes 6.1, System Access 2.3, NVDA, Thunder and, of course, Narrator which is there by default.

As I’ve said on numerous occasions, I mostly use JAWS. One reason for this choice comes down to the fact that nothing comes even close to the support one gets in VisualStudio with the combination of JAWS and the scripts on which Jamal Mazrui has led the development and many blind hackers from the blind programming list and elsewhere have contributed. Perhaps the others can catch up if they add a scripting facility but, for now, JAWS remains the only game in town.

I also spend a lot of time in Microsoft Word. I write this blog in Word as well as do my scholarly writing for publication, write various reports for professional tasks and do my creative writing using the popular word processor. For reasons I do not understand, JAWS performance in MS Word requires that I spend a lot of time waiting – it has grown too slow for my taste. Meanwhile, Window-Eyes and System Access both perform with a level of efficiency that I find quite usable. So, when I want to use Word, I quit JAWS and launch one of its competitors. Unfortunately, neither Window-Eyes nor SA does a very good job with the more advanced Word features that I must use when working on a collaborative project. Thus, if I need to merge my work with that of another on the same project, I need to quit SA or Window-Eyes and launch JAWS, deal with the sluggishness while merging the edits submitted by my colleagues and then jump back into another screen reader to do large scale writing. Microsoft Word is probably one of the single most important programs used by screen reader users – it boggles my mind that I cannot use any single screen reader to accomplish everything I need to do in Word.

Those of us who can afford to keep multiple screen readers installed can feel lucky. Switching from one AT product to another might annoy but it sure beats having only one solution. This, quite unfortunately, faces most blind computer users.

One of the biggest problems with the screen reader market is that the customers, those who make the purchasing decisions and write the checks, rarely also use the screen readers themselves. Thus, many purchasing decisions happen without a fully informed consumer and, in the worst cases, a blind person receives a bit of AT with which he cannot actually perform his job. This problem can only partially be blamed on AT vendors as, with a increasing frequency, accessibility decisions are informed by a mainstream company’s claims that their product works with a specific AT product.

So, if screen readers can, at best, provide a 90% solution, how can a blind person perform 100% of a job? In this case, JAWS, because of its incredibly powerful customization facilities becomes the only true workplace solution. If a blind employee needs to use a particular program, it is far more likely that JAWS can be customized to work with it than any of its competitors. Of course, only a small number of companies have the financial wherewithal to hire a consultant to write JAWS scripts for a small number of employees who happen to need a screen reader.

There are quite a lot of jobs that can be performed with a single screen reader. Unfortunately, a screen reader user can find themselves passed over for promotion because the next job in an organization’s hierarchy requires applications that do not work with the screen reader the purchasing people chose for the employee to use. I find it sad that Microsoft Project, a program used in many companies that is essential to getting a management job cannot be used with any screen reader. No major accounting software, to my knowledge, works with a screen reader. Visio and other programs used to draw diagrams are not accessible. Few, if any, UML editors have been made accessible. The list of applications and application types that have no screen reader accessibility that are essential to performing some of the highest paying jobs can not be used by blind people which, in effect, holds our community back.

None of this can be blamed on Freedom Scientific or JAWS as, by quite some distance, more professional applications work with it than any of the other players. I will remind the reader, though, that JAWS users have enjoyed much greater access for a long time. Given technology available today, I could not do my various jobs without JAWS, I admit that I do my work more efficiently because I have a number of screen readers installed and I find a lot of value in the JAWS competitors but, if I had to give up all but one, I would keep JAWS as its power out distances its flaws by a large margin.

The economics of blindness, however, with sighted people making buying decisions, mainstream companies doing the minimum to avoid discrimination complaints and the fact that all AT companies remain fairly small and need to be cautious about investing in engineering that might not pay off in growing sales, will continue to cause difficulty. Open source solutions, with large corporate sponsors like Sun and IBM, seem to be gaining traction on the GNU/Linux platforms, perhaps NVDA will catch fire on the Windows platform?

Afterward

I know that I promised not to write about AT because I cannot be even moderately objective. Thus, the reader should be reminded that the above was written by someone who proudly worked for FS for six years and still feels strongly about the good work he did while there and who maintains a high degree of confidence in the team of programmers at FS who work on JAWS.

Update: I received a phone call about the original version of this post that pointed out some factual errors. I have removed the offending passage but the rest of the post is as it was originally published.

-- End.