Long term accessibility (was: Re: Development cost proposal needed.)

[Tony Baechler]

At 07:40 AM 1/24/02 -0500, you wrote:
>we call it.  My view is that if we can get a robust Linux solution that
>would allow blind folks to get real access to an inexpensive operating
>system which we could use to do what we need and also to access the
>Microsoft apps through emulators, then we really could go a long way to
>creating a level playing field with a host of options rather than
>constantly trying to catch up to main apps and forsaking the others.  Not


Hello.  I think I am coming in late on this thread, but your comments 
interested me.  I would like nothing better than to have a workable Linux 
solution which allows me to get everything done which I use Windows for 
now.  However, I just do not see this as realistic, regardless of available 
resources.  As much as I would like to criticize Mr Mosen's comments, he is 
mostly right.  There are two things here which are going to be stumbling 
blocks.  I write from my own experience with using LInux.  One is that 
emulators are generally not accessible because of the GUI and different OS 
standards.  What works one way in Linux does not work at all in 
Windows.  PINE vs. PC-PINE is a good example.  The standard "pine" is a 
good command line program which does everything with the keyboard.  The 
Windows equivalent uses dialogues and the GUI for everything and does not 
even work with pop3.

Another serious difference is between Lynx and IE.  Lynx does not handle 
cookies correctly.  Lynx does not do Java or javascript.  Lynx knows 
nothing about Activex or being able to display images.  IE is much slower 
and has far more security problems.  IE crashes at the least opportune 
times.  IE requires some form of MSAA implementation to be useful at 
all.  With that said, I personally use Lynx for 95% of my surfing and only 
use IE when I have to.  However, Lynx for win32 is vastly different than 
Lynx for Linux.  Besides the win32 version being slower and having 
stability problems, (it randomly crashes sometimes) it does not handle 
cookies at all (it just throws them away on exit) and has no idea about 
handling multiple users.

The other serious problem has nothing to do with the blind, but equally 
affects sighted.  That is that Linux is still relatively new and is written 
by amateurs.  I think the kernel does a very good job and the programmers 
are very, very good considering what there is to work with, but the fact of 
the matter is that new kernel versions are coming out constantly.  It has 
not got to a point where it is stable enough that it can go for months 
without bug fixes, new drivers, general code tweaks, etc.  You could argue 
that "I have been using 2.2.19 for months without any problem," or any 
other kernel version, but those are not the "latest" or "newest" kernels in 
most cases because they are updated too often.  When I was actively running 
a Linux box, I was compiling a new kernel once a week or more.  That was 
last summer.  In a very short time, I went from kernel 2.2.19, 2.4.5, and 
ended up at 2.4.12.  I never got 2.4.13 or higher to work.  How, then, do 
you expect to break the MS monopoly when suddenly your kernel (assuming you 
use modules) is yanked out from under you because a new version came 
out?  Sure you could do it like Speakup does now, but how often has it 
needed updates (I know of at least once, if not more) because it could not 
patch the latest kernel correctly?  On the other hand, it amazes me how 
very stable Linux, Speakup, and the software is in general and I wish MS 
would learn from Linux.  Obviously Microsoft has serious problems which 
they are in no hurry to address, but they have the funds to blow people off 
without a thought.  The gnu project was established way back in the late 
1970's or early 1980's.  Their goal was to have a totally free OS.  Years 
later they are just now seeing this reality.  Therefore, it is unreasonable 
to expect overnight progress for the blind community.  Things take time.

In conclusion, I think that the days of Windows being dominant are coming 
to an end, but very slowly.  If a majority of the blind appealed to screen 
reader manufacturers such as GW Micro and Freedom Scientific, they would 
move their focus to Linux.  That has not happened either, and I would be 
surprised if it does for awhile.  As Kirk W wrote, Linux is not for the 
masses.  Personally, I would like to keep it that way.  Emulators are still 
very much in their infancy.  It will take years before that 
changes.  Rather than trying to focus all our resources on an immediate 
solution, let us wit and take one step at a time.  Too many other things 
need to happen before it can be a level playing field.  Linux needs to grow 
up a little more, along with Speakup, the ODS and Windows emulators, and 
blind computer users.  There are still the majority of the blind who prefer 
Windows and are not going to change yet.  They could change if Linux was as 
accessible as Windows, but that is unreasonable for now.  Sorry for my 
rambling.  I really do like and prefer Linux (otherwise why would I be on 
the list?) but we must deal with the limitations facing us.  Please feel 
free to correct my impressions if they are wrong, but no flames please.  Thanks.