Re: [Accessibility] Call to Arms

[Chris Hofstader]

eric: it's a matter of pragmatics or something but if the current free
solutions are inadaquite or don't work at all...

cdh: Although almost all of my career until this year was spent making 
proprietary software, I have always held that the ideal of free software 
went beyond the theoretical issues regarding software freedom (to which 
I subscribe entirely) but also that free software is the most pragmatic 
solution.

cdh: In 1983, this whole GNU/Linux thing was gaining some ground but was 
still viewed as a hacker toy in many areas. Today, 90% of the Internet 
runs on some sort of GNU/Linux based system.

cdh: Remember the horrible days back in the late 80s and early 90s when 
most WAN systems were proprietary? Remember things like MS NT Server, 
Novell Groupwise, Lotus Notes (pre-IBM acquisition), Novells thing, IBM 
Domino and a few others? Getting any one of these proprietary systems to 
work reliably was a real big effort; getting two or more of them to 
communicate with each other required a tremendously expensive 
consultant. Free software, programs like MOsiac started popping up and 
people started dropping the proprietary  and hardly reliable WAN systems.

cdh: Then came programs like Apache and so much more from the FLOSS 
world. I don't know what the 10% of the Internet that uses proprietary 
servers are using nor do I understand why they would be doing so.

cdh: NOw, imagine if the Internet had to try to progress using all of 
those different proprietary WAN systems and trying to keep them talking 
to each other? We'd still be thinking that the silly hamster dance was 
pretty cool.

cdh: I've used NaturallySpeaking on Windows and MacSpeech (a DNS based 
dictation program for Macintosh) on a number of occasions. I admit their 
recognition quality is pretty good but we can't go in and make them 
better or add API calls we would like or let some of our scholarly 
friends try out some new theory that might be a major step forward. With 
proprietary software, we get what we get.

cdh: Also, a speech recognition/dictation engine goes way beyond the 
community of people with disabilities and we can cast a much wider net 
for volunteer computer scientists, computational linguists and hackers.

cdh: Meanwhile, Nuance is bound by a huge debt burden that came (of 
course) from acquiring everything from ETI to Dragon to KESI to the Talx 
guys to SpeechWorks to L&H to some Israeli speech synthesizer to 
virtually all OCR companies aside from Fine Reader  business  and lots 
more. While they have incredible market share in most of their business 
segments, they are laying off high priced hackers and hiring commission 
based salespeople and are spamming all of we old customers on at least a 
weekly basis so they can catch up with their interest payments. A lot of 
their employees are looking hard for other gigs as they can feel the 
place going bad and I've even heard rumour that ETI might be spun off 
into the good old Eloquence only company that made the de facto standard 
speech synthesizer for screen readers.

cdh: So, business issues entirely unrelated to a program can serious 
hurt its quality, especially if some of their old reliable hackers walk 
away. In a libre planet, the source is their for all to see, learn and 
help improve. Shaky companies can mean disaster for their users.

cdh: So, protecting our liberties will also be protecting our investment 
in our  software.

cdh