[9fans] Are there any blind users of Plan 9?

[9fans] Are there any blind users of Plan 9?

[Chris Brannon]

Hi there,
I'm blind, and I use Unix from the text console.  I'm interested in
trying out Plan 9.  It appears to be a very clean system.  Are there any
blind people in the Plan 9 community?  If so, I am very interested in
hearing from them.

-- Chris

Re: [9fans] Are there any blind users of Plan 9?

[Rudolf Sykora]

[-- Attachment #1: Type: text/plain, Size: 716 bytes --]

Hello Chris,

I don't know anybody blind using the system.
I actually don't know personally anybody using it. :)
But, although the system seems clean to me, I am afraid its use is rather
mouse-centric, which, for you, may be a huge downside. Plan 9, in my
opinion, is not suited for just-keyboard use, which you may possibly need...
But don't take this as a discouragement.

Ruda

2008/10/20 Chris Brannon <cmbrannon@cox.net>

> Hi there,
> I'm blind, and I use Unix from the text console.  I'm interested in
> trying out Plan 9.  It appears to be a very clean system.  Are there any
> blind people in the Plan 9 community?  If so, I am very interested in
> hearing from them.
>
> -- Chris
>
>

[-- Attachment #2: Type: text/html, Size: 1088 bytes --]

Re: [9fans] Are there any blind users of Plan 9?

[Gorka Guardiola]

On Mon, Oct 20, 2008 at 11:27 AM, Chris Brannon <cmbrannon@cox.net> wrote:
> Hi there,
> I'm blind, and I use Unix from the text console.  I'm interested in
> trying out Plan 9.  It appears to be a very clean system.  Are there any
> blind people in the Plan 9 community?  If so, I am very interested in
> hearing from them.
>

It is a very mouse oriented system which is probably bad.
By mouse oriented, I mean a lot of the user interface can only be interacted
with a mouse.

On the other side, acme is text oriented and rio serves
files with the (text) contents of the window.
Both can probably be tweaked a little so that
an external program (which has yet to be written/ported)
reads aloud stuff or prints things out through a Braille line, but to
my knowledge this hasn't been done.

--
- curiosity sKilled the cat

Re: [9fans] Are there any blind users of Plan 9?

[Jeff R. Allen]

I was just playing with Inferno this weekend, and what's nice about it
is that when you are using it hosted, it is simply a process named
"emu" running in a window and talking to you in text. Your screen
reader would be able to understand it's output easily.

Inferno lets you play with the 9p protocol that shares resources
across a network, and that's really the cleanest and most attractive
parts of Plan 9. Also, Inferno gives you access to Limbo. Limbo's
support for concurrency and cross-platform execution is quite
attractive and should be compatible with a screen reader. The
"mouse-intensive" part people are talking about is the rio window
manager and the acme editor.

On the other hand, programming Limbo with only cat and sed won't be
too easy, so you'll need suggestions from people on the list for less
mousy editors. Seems unlikely that Plan 9 people have ported Emacs...
the massive cognitive dissonance would have likely created a
singularity which annihilated the programmer in question. :)

  -jeff

PS: Heh, just found this: http://lsub.org/magic/man2html/1/emacs

Re: [9fans] Are there any blind users of Plan 9?

[Rudolf Sykora]

[-- Attachment #1: Type: text/plain, Size: 207 bytes --]

> On the other hand, programming Limbo with only cat and sed won't be
> too easy, so you'll need suggestions from people on the list for less
> mousy editors.
>  -jeff


There is a port of vim...
R.

[-- Attachment #2: Type: text/html, Size: 475 bytes --]

Re: [9fans] Are there any blind users of Plan 9?

[Pietro Gagliardi]

[-- Attachment #1.1: Type: text/plain, Size: 785 bytes --]

I'm blind in only one eye and have low vision in the other, so I run
Plan 9 in a virtual machine with an enlarged screen using Mac OS X's
Universal Access.

The concept of a Text-to-Speech program for Plan 9 has been floating
in my head for some time. How can it be made to use some of Plan 9's
features (/dev/*ctl, /srv, text-based commands, etc.)? I was thinking
either something like
	echo say (voice) (ipa-pronounciation) > /dev/speech
	echo sayword (voice) (word) > /dev/speech
then use such a device to build a screen reader.

PS -
On Oct 20, 2008, at 6:15 AM, Jeff R. Allen wrote:
> PS: Heh, just found this: http://lsub.org/magic/man2html/1/emacs

That's in the PDF. In fact, it's on the GNU Humor page now, too. I
wonder how vi is related anymore, though.


[-- Attachment #1.2: Type: text/html, Size: 1839 bytes --]

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 201 bytes --]

Re: [9fans] Are there any blind users of Plan 9?

[john]

> I'm blind in only one eye and have low vision in the other, so I run
> Plan 9 in a virtual machine with an enlarged screen using Mac OS X's
> Universal Access.
>
> The concept of a Text-to-Speech program for Plan 9 has been floating
> in my head for some time. How can it be made to use some of Plan 9's
> features (/dev/*ctl, /srv, text-based commands, etc.)? I was thinking
> either something like
> 	echo say (voice) (ipa-pronounciation) > /dev/speech
> 	echo sayword (voice) (word) > /dev/speech
> then use such a device to build a screen reader.
>

Better to have something like:
	echo 'voice AmericanMale' > /dev/speechctl
	echo 'values of β may give rise to dom!' > /dev/speech
This way you do all setup in the ctl file and only send the things you
want said to /dev/speech

For a simplest first implementation, it may be best to try interfacing
with a Festival TTS server on a UNIX box rather than developing a
text-to-speech application for Plan 9--Nemo, isn't that what you have
at lsub?

John

Re: [9fans] Are there any blind users of Plan 9?

[Dave Eckhardt]

The bad news is that the existing interface is very
mouse-oriented and there is no text-to-speech support.

The good news is that the system, and each part of it,
is small, so if you don't like something it can be
replaced.  Unlike Windows or Unix, where you can't
do much about the windowing system, in Plan 9 you
really can replace it and it's not that much code.

The bad news is that support for whizzy AJAX browsers,
etc., is unlikely soon.

Is there a university near you which has a CHI/HCI
(Computer-Human/Human-Computer Interaction) degree?
It might be possible to find a student or two looking
for a way-out thesis project and help them design
something interesting.  Where are you located?

Dave Eckhardt

Re: [9fans] Are there any blind users of Plan 9?

[Chris Brannon]

Dave Eckhardt writes:

> The good news is that the system, and each part of it,
> is small, so if you don't like something it can be
> replaced.  Unlike Windows or Unix, where you can't
> do much about the windowing system, in Plan 9 you
> really can replace it and it's not that much code.

Quite a few people have mentioned the heavy use of the mouse, and that
is definitely a drawback.  Replacing parts of the system does sound like
a good approach.

> Is there a university near you which has a CHI/HCI
> (Computer-Human/Human-Computer Interaction) degree?

I doubt it, but it might be worth investigating.  This is really a hobby
interest, rather than a necessity.  I stumbled upon the Plan 9 papers
while pursuing my own MS degree in CS.

An earlier poster mentioned Inferno, and I'm looking into that right
now.

Thanks to everyone for the informative replies.

-- Chris

Re: [9fans] Are there any blind users of Plan 9?

[Noah Evans]

Hey Chris,

Can you describe how you interact with the system now, and how you'd
do it if everything was perfect? Do you do much copying and pasting?
If so how do you determine the boundaries? Do you use a braille
display or text to speech stuff?

Noah

On Mon, Oct 20, 2008 at 11:06 AM, Chris Brannon <cmbrannon@cox.net> wrote:
> Dave Eckhardt writes:
>
>> The good news is that the system, and each part of it,
>> is small, so if you don't like something it can be
>> replaced.  Unlike Windows or Unix, where you can't
>> do much about the windowing system, in Plan 9 you
>> really can replace it and it's not that much code.
>
> Quite a few people have mentioned the heavy use of the mouse, and that
> is definitely a drawback.  Replacing parts of the system does sound like
> a good approach.
>
>> Is there a university near you which has a CHI/HCI
>> (Computer-Human/Human-Computer Interaction) degree?
>
> I doubt it, but it might be worth investigating.  This is really a hobby
> interest, rather than a necessity.  I stumbled upon the Plan 9 papers
> while pursuing my own MS degree in CS.
>
> An earlier poster mentioned Inferno, and I'm looking into that right
> now.
>
> Thanks to everyone for the informative replies.
>
> -- Chris
>
>

Re: [9fans] Are there any blind users of Plan 9?

[Federico G. Benavento]

charles ported rsynth some time ago, here's the link

http://www.terzarima.net/plan9/dist/rsynth.tgz

On Mon, Oct 20, 2008 at 9:34 AM, Pietro Gagliardi <pietro10@mac.com> wrote:
> I'm blind in only one eye and have low vision in the other, so I run Plan 9
> in a virtual machine with an enlarged screen using Mac OS X's Universal
> Access.
> The concept of a Text-to-Speech program for Plan 9 has been floating in my
> head for some time. How can it be made to use some of Plan 9's features
> (/dev/*ctl, /srv, text-based commands, etc.)? I was thinking either
> something like
> echo say (voice) (ipa-pronounciation) > /dev/speech
> echo sayword (voice) (word) > /dev/speech
> then use such a device to build a screen reader.
> PS -
> On Oct 20, 2008, at 6:15 AM, Jeff R. Allen wrote:
>
> PS: Heh, just found this: http://lsub.org/magic/man2html/1/emacs
>
> That's in the PDF. In fact, it's on the GNU Humor page now, too. I wonder
> how vi is related anymore, though.
>



--
Federico G. Benavento

Re: [9fans] Are there any blind users of Plan 9?

[Pietro Gagliardi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Charles sent me an email about rsynth. I'll get 9vx and take this up
(QEMU doesn't support Plan 9's audio system).

However, I do have some more ideas that would help:

1) A *blind or visually impaired* option in the installer
2) An option in rio to automatically open a new window
3) A window switching program for rio (think alt+tab in Windows)

It's a good thing the default resolution is 640x480!

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)

iEYEARECAAYFAkj89OEACgkQuv7AVNQDs+w8JQCeIIvEdoV7kbty7QUZylaCgc18
B8oAn3FP/o7qQQZNSIYEfDNdOLuOX8+z
=0sYF
-----END PGP SIGNATURE-----

Re: [9fans] Are there any blind users of Plan 9?

[Lyndon Nerenberg]

> 2) An option in rio to automatically open a new window
> 3) A window switching program for rio (think alt+tab in Windows)

For this particular case it's probably better to ditch rio completely and
write something dedicated to the task at hand.

As for editors, all you need is ed(1).  And believe me, it works. I've
used it many times to "blind"ly edit config files over bodged up tty
links.

--lyndon

Re: [9fans] Are there any blind users of Plan 9?

[Uriel]

sam -d is also very powerful, and might be worth learning even for
unix use, I'm totally ignorant about specialized editors for the
blind, but if I ever lost my sight, I think sam -d would be all I
would use.

(The tutorial on the sam language is most enlightening for everyone,
even lowly acme users like me:
http://doc.cat-v.org/bell_labs/sam_lang_tutorial/ )

uriel

On Tue, Oct 21, 2008 at 7:00 AM, Lyndon Nerenberg <lyndon@orthanc.ca> wrote:
>> 2) An option in rio to automatically open a new window
>> 3) A window switching program for rio (think alt+tab in Windows)
>
> For this particular case it's probably better to ditch rio completely and
> write something dedicated to the task at hand.
>
> As for editors, all you need is ed(1).  And believe me, it works. I've used
> it many times to "blind"ly edit config files over bodged up tty links.
>
> --lyndon
>
>

[9fans] How to go about doing screen reading

[Pietro Gagliardi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

In pseudocode:

	when mouse has not been moved for at least 1 second
		find cursor position
		if cursor has moved
			stop
		find window where cursor is
		if  cursor has moved or no text window underneath
			stop
		find line of text in the window device where the cursor is
		say the text
	when F1 is hit
		read out line already typed at open rio window

Any technical problems with this approach?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)

iEYEARECAAYFAkj9sOAACgkQuv7AVNQDs+y/aQCeIuuTrAdNar5BTXXa7oZQNMu7
9XMAn0qmnR5jup2t7cO8GeczROTOYu70
=r7hM
-----END PGP SIGNATURE-----

Re: [9fans] How to go about doing screen reading

[Chris Brannon]

pietro10@mac.com (Pietro Gagliardi) writes:

> In pseudocode:
>
> 	when mouse has not been moved for at least 1 second
> 		find cursor position
> 		if cursor has moved
*SNIP*
> Any technical problems with this approach?

I think there is an easier way to do screen reading within Acme.  Here's
a quote from the Acme paper:

<quote>
The last file, event, is the most unusual. A program reading a
window's eventfile is notified of all changes to the text of the
window, and is asked to interpret all middle- and right-button
actions. The data passed to the program is fixed-format and reports
the source of the action (keyboard, mouse, external program, etc.),
its location (what was pointed at or modified), and its nature
(change, search, execution, etc.). This message, for example,

MI15 19 0 4 time

reports that actions of the mouse (M) inserted in the body (capital
I) the 4 characters of timeat character positions 15 through 19;
the zero is a flag word. Programs may apply their own
interpretations of searching and execution, or may simply reflect
the events back to Acme, by writing them back to the eventfile, to
have the default interpretation applied. Some examples of these
ideas in action are presented below.
</quote>

A screenreader can obtain information about changes to the state of the
system by reading the event file.  Do other parts of the window system
provide this sort of interface?

-- Chris

[9fans] bitsy anyone?

[Benjamin Huntsman]

[-- Attachment #1: Type: text/plain, Size: 512 bytes --]

Does anyone happen to have a working bitsy kernel and paqdisk?

I've tried both compiling myself and the kernel/paqdisk that used to be on nemo's contrib, both to no avail.

Either way, I get a kernel panic on the iPaq.  The various instructions around on the wiki and such have the user partition too small for the paqdisk that gets built "out of the box" from a fresh Plan 9 install... so I did increase the partition sizes accordingly.  I'm hoping that's not the problem...

Thanks in advance!

-Ben

[-- Attachment #2: winmail.dat --]
[-- Type: application/ms-tnef, Size: 2614 bytes --]

Re: [9fans] bitsy anyone?

[Francisco J Ballesteros]

I can dig in the dump to find out our last kernel/paqdisk.

But we don´t use that anymore.

On Tue, Oct 21, 2008 at 7:21 PM, Benjamin Huntsman
<BHuntsman@mail2.cu-portland.edu> wrote:
> Does anyone happen to have a working bitsy kernel and paqdisk?
>
> I've tried both compiling myself and the kernel/paqdisk that used to be on nemo's contrib, both to no avail.
>
> Either way, I get a kernel panic on the iPaq.  The various instructions around on the wiki and such have the user partition too small for the paqdisk that gets built "out of the box" from a fresh Plan 9 install... so I did increase the partition sizes accordingly.  I'm hoping that's not the problem...
>
> Thanks in advance!
>
> -Ben
>

Re: [9fans] bitsy anyone?

[David du Colombier]

> Does anyone happen to have a working bitsy kernel and paqdisk?
>
> I've tried both compiling myself and the kernel/paqdisk that used to be on nemo's contrib, both to no avail.
>
> Either way, I get a kernel panic on the iPaq.  The various instructions around on the wiki and such have the user partition too small for the paqdisk that gets built "out of the box" from a fresh Plan 9 install... so I did increase the partition sizes accordingly.  I'm hoping that's not the problem...

Hello,

I am currently using Plan 9 on a Compaq iPaq h3650.

I compiled the kernel and the paqdisk in May 2008 from a
current Plan 9. It worked really fine, except with the Wi-Fi.

However, I had some problems when uploading the paqdisk
on the iPaq when the image file is larger than 4 MB. The image
seems to be corrupted during the upload on the iPaq and Plan 9
crash during boot. This problem does not appear with a smaller paqdisk.

Also, as you said, If you leave the armpaqproto configuration file as it is,
the image file created will be larger than the partition, so you must modify it.

In fact, I read in a documentation you cannot easily resize
the partition since it is apparently hard coded in the kernel.

One of the kernel and paqdisk I made is available on my website [1].
I also put the concerning armpaqproto file online.
It was compiled in May 2008.

I also tried from old Nemo's and John's [2], and it worked fine too.
But they are really old.

[1] http://www.9grid.fr/misc/plan9/ipaq_h3650/work-mine
[2] http://www.9grid.fr/misc/plan9/ipaq_h3650/work-other

--
David du Colombier

Re: [9fans] bitsy anyone?

[Eoghan Sherry]

> However, I had some problems when uploading the paqdisk
> on the iPaq when the image file is larger than 4 MB. The image
> seems to be corrupted during the upload on the iPaq and Plan 9
> crash during boot. This problem does not appear with a smaller paqdisk.

I don't have a Plan 9 system or a bitsy at hand but I remember
running into this. The hard coded partition definition is fparts in
/sys/src/9/boot/paq.c. The ramdisk partition is defined to be 4 MB although the
various documents imply it should be 6 MB.

I recall simply changing the ramdisk entry to,
        "add ramdisk    0x0200000 0x0800000",
allowed me to use larger images.

Hope this helps,
eoghan

Re: [9fans] bitsy anyone?

[Benjamin Huntsman]

[-- Attachment #1: Type: text/plain, Size: 1176 bytes --]

>"add ramdisk    0x0200000 0x0800000",

I believe the command in question is:
partition define ramdisk 0x200000 0x600000 0

I had changed it to 0x800000 too, but still got the kernel panic...
I'll take a look in paq.c and see if I can fix the hard-coded sizes.

Thanks!!
-Ben


-----Original Message-----
From: 9fans-bounces@9fans.net on behalf of Eoghan Sherry
Sent: Tue 10/21/2008 2:09 PM
To: Fans of the OS Plan 9 from Bell Labs
Subject: Re: [9fans] bitsy anyone?
 
> However, I had some problems when uploading the paqdisk
> on the iPaq when the image file is larger than 4 MB. The image
> seems to be corrupted during the upload on the iPaq and Plan 9
> crash during boot. This problem does not appear with a smaller paqdisk.

I don't have a Plan 9 system or a bitsy at hand but I remember
running into this. The hard coded partition definition is fparts in
/sys/src/9/boot/paq.c. The ramdisk partition is defined to be 4 MB although the
various documents imply it should be 6 MB.

I recall simply changing the ramdisk entry to,
        "add ramdisk    0x0200000 0x0800000",
allowed me to use larger images.

Hope this helps,
eoghan



[-- Attachment #2: winmail.dat --]
[-- Type: application/ms-tnef, Size: 3250 bytes --]

Re: [9fans] bitsy anyone?

[Eoghan Sherry]

2008/10/21 Benjamin Huntsman <BHuntsman@mail2.cu-portland.edu>:
>>"add ramdisk    0x0200000 0x0800000",
>
> I believe the command in question is:
> partition define ramdisk 0x200000 0x600000 0
>
> I had changed it to 0x800000 too, but still got the kernel panic...
> I'll take a look in paq.c and see if I can fix the hard-coded sizes.

The boot loader command,
        partition define ramdisk 0x200000 0x600000 0
is correct. The change is to paq.c.

The boot loader uses start+size while Plan 9 uses
[start,end) to define partitions.
The instructions on the wiki describe a 6 MB ramdisk starting
2 MB into the flash.
To the boot loader that is 0x200000+0x600000 but to Plan 9
it is [0x200000,0x800000).
The problem is paq.c defines a [0x200000,0x600000) ramdisk
which is only 4 MB in size.

Comparing the values in paq.c with those on the wiki should make
things clear.

eoghan

Re: [9fans] Are there any blind users of Plan 9?

[matt]

why not just ssh into plan9 and use that ?


btw. no IWP for me :(

bugger bugger bollocks