Main advantages of SBL over Speakup

[Tony Baechler]

I have a couple comments.  First, last summer we looked extensively at 
newegg.com.  Other than some low-end motherboards, the majority did not 
have serial ports or headers.  They were obviously designed for gamers.  
They had great onboard audio and tons of USB ports, but no serial.  We 
found a grand total of about two.  This was last summer, so I can't help 
but think the situation has only become worse.  Even then, it was 
limited to one or two brands.  We did find one, but it took some 
hunting.  I think it's for servers.

I have two answers why I personally refuse to use software speech on my 
desktop.  First, it's a matter of preference.  I don't like ESpeak.  It 
is free and that is good, but I find it hard to understand.  In the case 
of the Ubuntu live CD, it chops off words, making it almost useless.  
Secondly, not all sound cards are supported or are supported properly.  
Many desktops come with Nvidia chipsets.  HP is definitely one.  Yes, 
there are Nvidia drivers, but they are not free.  Either you have to 
install probably buggy non-free drivers which might or might not work or 
you have no sound, period.  That also applies to the Ubuntu install.  
The sighted person who uses the HP machine complained that the installer 
screws up the font on the monitor until after the installation is 
complete and the non-free drivers are set up.  In my case, Linux 
supports my sound card but at a very low volume.  I've tried everything 
I can think of in amixer but even at full volume, it's only a whisper.  
Windows doesn't have that problem.

On 2/11/2010 9:00 AM, John G. Heim wrote:
> Maybe its accurate to say that *most* modern motherboards don't have 
> serial ports. I question even that. But you didn't address my 
> questions about how you determined your priorities. We both agree that 
> most servers have serial ports. That seems unlikely to change any time 
> soon. So I don't understand why its so important to you to have 
> hardware speech on your desktop. I understand it for servers. In fact, 
> that was part of my original point --  having speakup in the kernel is 
> very important if you need to find out what a server won't boot or why 
> it can't get a network connection. But on your desktop, why don't you 
> just use software speech? In fact, why don't you just get one PC for 
> yourself with a serial port and use software speech on all the other 
> desktops that you do support on. That's what I do even though all of 
> our PCs have serial ports.