The GW Micro blog has been discontinued. For instant updates on GW Micro products and events, follow us on Twitter, and like us on Facebook.
More Thoughts on Braille Driver Signing
by Doug on Wednesday, December 23 2009On September 15, 2009, I created a blog entry entitled "A Thought on Braille Driver Signing". Since then, the industry has been in an uproar regarding the fees and user disruption caused by this new requirement. An online petition was created to oppose this initiative, and has already received over 1,100 signatures. I listened to the September 2009 FScast where Glen Gordon was "interviewed" by the Freedom Scientific employee, Jonathan Mosen. Several of the questions posed to Mr. Gordon seemed to be specific to my blog post although my post was never acknowledged. I found it curious that Mr. Gordon never mentioned the price they charge for the requirement. Mr. Gordon also spoke about how you can't trust the Braille manufacturer to create stable and reliable Braille drivers. Basically, you can only trust Freedom Scientific to validate that a Braille display driver is reliable. But when posed the question of why the JAWS video mirror driver wasn't digitally signed by Microsoft, Mr. Gordon said it wasn't the cost that was the deterrent, it was the time it took to have it signed by Microsoft, and that, again, users can trust Freedom Scientific because they've done all the necessary testing. This sure seems like a double standard to me. Beyond the double standard I find his comment about Microsoft taking too long baffling. As I pointed out, GW Micro does indeed have Microsoft sign the Window-Eyes mirror driver. Would you be surprised if I told you the entire process takes less than 24 hours and the cost is only $250.00? This is not an annual fee and is only imposed when we update our driver (which isn't very often). Microsoft provides utilities to do extensive automated testing of the mirror driver. All you need is a server machine, a 32-bit client and a 64-bit client. The server then performs a set of rigorous testing scenarios on both the 32-bit and 64-bit client machines. Because we run the test simultaneously for both 32-bit and 64-bit, the entire process takes about 12 hours to complete. Assuming the test results were successful, they are then submitted to Microsoft along with our 32-bit and 64-bit mirror drivers. Within 12 hours Microsoft sends us back the signed mirror driver and later bills us for this process. I wonder if Freedom Scientific returns the signed Braille driver within this period of time. Because Microsoft has set up the testing to be automated, it doesn't require any human intervention. As mentioned, the entire process takes less than 24 hours. So to say they don't sign the driver because it takes too long is simply not accurate. If security were a true concern, their mirror drivers would be signed by Microsoft. I was willing to let this go even though the comments expressed on the FScast seemed misleading, but then I read a more recent posting from Freedom Scientific. This was in their November – December 2009 newsletter. While I was happy to see Freedom Scientific reduced and made public the fee they are charging Braille display manufactures to sign their Braille drivers, even at $9,000 for the first year and $5,000 for all other years, the cost is still simply too high. Perhaps Freedom Scientific could offer their services to assist in testing and such but to require it and charge these fees is simply not acceptable in our adaptive technology community. I also don't understand why testing would be so necessary between versions of JAWS. If a standard is well defined and written, and with each new upgrade that standard is maintained, then everything should continue to work and not require constant testing and redevelopment of the driver. For example, with every new release of Word, Microsoft doesn't have to test it with all available printer drivers on the market. Nor does Microsoft require all printer drivers to pay an annual fee for testing. With every new release of Window-Eyes we don't have to test all available SAPI synthesizer drivers on the market, and charge all synthesizer manufacturers an annual fee. What really caught my attention in their newsletter and caused me to create this follow up blog post was the answer to the question, "Given the benefits, low cost, and ease of participation, why am I waiting for a driver from my Braille display manufacturer?" The answer was "Many manufacturers are developing drivers using the new driver development kit. See Participating Braille Display Manufacturers below. A few have chosen not to sign up because they think they can gain a competitive advantage over Freedom Scientific by attacking the program … these few have taken this approach even though it hurts customers that have purchased their Braille displays." Freedom Scientific has taken a major disruption they created, and tried to turn it around on the Braille display manufacturers. Because of all the reasons I have given, this is obviously a transparent attempt to deflect the real issue. This brings back to mind when we (GW Micro) approached Freedom Scientific a few years ago, and asked them for the JAWS Braille SDK so we could create a JAWS Braille driver for the Braille Sense. The answer was they would not allow us to create the Braille driver because it didn't make good business sense for Freedom Scientific. Freedom Scientific saying companies are using this as a competitive advantage, again, is a transparent attempt to divert the real issues. Currently, the majority of Braille display manufacturers are pressured to conform even though they know this is not acceptable. Because there is no telling in the future what new restrictions or fees may be imposed to Braille display manufacturers, many of these companies are working on a standard interface for Braille displays. Just like software voice synthesizers have standardized on SAPI, hopefully we'll soon see a similar interface for Braille displays. It makes good sense. This way, any application (like a screen reader) would only need to talk to one standard instead of having so many unique interfaces to deal with. This is how the industry should be going, and GW Micro will help in promoting and making this happen. Although it may seem like a good source of revenue, GW Micro will never charge for the Window-Eyes Braille SDK. Meaning we encourage Braille display manufactures to create Braille drivers for Window-Eyes, not discourage them. Because of the way Window-Eyes has implemented and documented the Braille interface, it doesn't break with each release of Window-Eyes. We will continue to make sure Window-Eyes supports as many Braille displays as possible and still remain stable and reliable. Even if that means working with companies that do or may directly or indirectly compete with GW Micro.
