[nfbcs] Linux kernel Patch for Speakup
John G. Heim
jheim at math.wisc.edu
Mon Mar 3 15:45:30 UTC 2014
Okay, if you're going to call the code "really, really messy", then
would you please explain to me what the code should do instead?
That, of course, is a loaded question. The reason is that while
everybody agrees that the way speakup deals with the serial port is
wrong, nobody seems to know what the right way would be. So I'd like you
to tell me, jim, what would be the right way to talk at boot time to a
speech synth connected to the serial port.
From my conversations with people on the linux kernel developers list,
it appears there is a standard for how to *not* talk to the serial port
at boot time but no standard for how to do it. If there is a standard
for that, none of the code I've seen uses it. I'd love to have someone
point me to a piece of linux kernel code that does it the right way.
On 03/03/14 00:20, Jim Barbour wrote:
> So, keep in mind that the program we're talking about here is speakup; which is a screen reader for Linux.
>
> This really isn't about retrofitting bolted on accessibility, it's about who owns doing some really messy kernel programming to make speakup work really early in the linux boot process.
>
> Jim
>
> On Sun, Mar 02, 2014 at 10:40:32PM -0600, Larry Wayland wrote:
>> Jim, you are right about the retro fixing part. It's never easy to retro fit
>> and that never works like it should anyway. But! If they had written in the
>> accessibility while the program was on the drawing board it would not have
>> been near as difficult and would have worked much better. I think we are
>> losing ground on educating the developers and this is a real shame.
>> We also need to be educating the companies who are developing the
>> development tools.
>> Larry
>>
>>
>> -----Original Message-----
>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Jim Barbour
>> Sent: Sunday, March 02, 2014 8:02 PM
>> To: NFB in Computer Science Mailing List
>> Subject: [nfbcs] Linux kernel Patch for Speakup
>>
>> John, I looked at your request to kernelhub and I'm surprised at your
>> characterization of "they just don't get it." I think we you mean is
>> "they're not going to fix my problem, and it's anoying me."
>>
>> It's true that they aren't jumping on this, but they are letting you know
>> what the challenges are and what will need to be done to fix the patch. Of
>> course it's difficult, if it were easy it would have been done already.
>>
>> For interested parties, the request is here...
>>
>> http://www.kernelhub.org/?p=2&msg=27537
>>
>> To simplfy and summarize the thread, early in the linux boot process,
>> speakup is trying to get access to the first serial port so that it can send
>> speech to it. The original code is a bit buggy, resulting in the boot
>> process hanging when speakup tries to gain control of the serial port. John
>> is asking that this bug be fixed.
>>
>> The kernel folks are saying that the entire method that speakup is using to
>> gain access to the serial port is completely dead and should be rewritten.
>> The problem is that there isn't a good, existing service for speakup to talk
>> to. The only real solution is the rather significant heavy lifting of
>> rewriting speakup to basically turn itself into a serial console without
>> benefit of a kernal's serial layer.
>>
>> I think this is the problem with putting things like speakup into the kernel
>> at all without a clear indication that some entity is willing to provide
>> long term support. The code needs to be rewritten. John is a user of the
>> technology and doesn't want to do the work. The kernel developers have no
>> resources or insentive to fix it, and there's no real product support group
>> for speakup.
>>
>> I'd be curous to hear others opinions on this.
>>
>> Jim
>>
>> On Sun, Mar 02, 2014 at 07:06:46PM -0600, John G. Heim wrote:
>>> I've been struggling to get a patch into the linux kernel that makes
>> hardware speech synthesizers work at boot time. Some of the kernel
>> developers don't get what the big deal is. But would they release a version
>> of the operating system that didn't have video at boot time? Imagine an
>> operating system where the users had to wait until the computer started up
>> and then type a command to get the monitor to work. That would be insane.
>> Nobody would say they had to put it out that way because it would take too
>> long to fix. It just wouldn't be considered to be done until that problem
>> was addressed.
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Mar 2, 2014, at 12:51 PM, Jim Barbour wrote:
>>>
>>>> <note> I'm starting a new thread about this topic because frankly I
>> don't know if what I'm about to say applies to new versions of MS office. I
>> truly don't know if MS is innovating their user interfaces or if they're
>> just being lazy, and would like for *that* conversation to take place on the
>> other thread.
>>>> </note>
>>>>
>>>> I was having a similar discussion about innovation with Tim Elder this
>> weekend.
>>>>
>>>> It is my rather strong opinion that allowing innovation does, pretty
>> much by definition, require that we allow new ideas to grow and flourish
>> before accessibility can be brought into the picture.
>>>>
>>>> I would say that this applies to most areas of broad usability
>> requirements such as Internationalization, localization, varying color
>> palettes, ergonomic menu placement, etc., etc.
>>>>
>>>> My favorite example of innovation that would have been stopped cold by
>> accessibility standards is the whole AJAX/web 2.0 model that Google
>> introduced with their gmail product in 2004.
>>>>
>>>> Google could not have released gmail in 2004 if they had been required
>> to make it accessible. They couldn't have, even if they'd wanted to. The
>> technology needed to make AJAX accessible, what we now call the ARIA roles,
>> hadn't been invented yet.
>>>>
>>>> What basically had to happen was a rather long set of conversations
>> amongst web browser developers, screen reader developers, and web app
>> developers to figure out how to communicate the necessary information
>> through the necessary channels so that screen readers could get the right
>> information at the right time.
>>>>
>>>> Within the general web usability community, AJAX/web 2.0 is pretty much
>> the big step for web apps becoming truly user friendly web apps, rather than
>> clunky, text based, app like web pages.
>>>>
>>>> Therefore, it's pretty easy to argue that AJAX/web 2.0 was a very
>> necessary usability step for the general user community, that the blind
>> were (sometimes still are) hurt by this usability improvement, and that
>> screen reading technology has to catch up.
>>>>
>>>> This is pretty much the same dynamic that we've seen over and over
>> again. When DOS become windows 3.0, UNIX started using X-windows and
>> session managers, phones switched from keypads to touch screens, and
>> probably a few examples I missed.
>>>>
>>>> So, where is the point at which a webapp stops being experimental and
>> starts being an entity that is expected to provide reasonable accomodations
>> for disabled users?
>>>>
>>>> Is it when you have to acknowledge license agreement or terms of service
>> page? Is it when you are expected to give them a credit card number? Is it
>> when the site is made available to the general public, as apposed to a
>> limited beta? I can argue for and against each of those suggestions.
>>>>
>>>> I do know that the Department of Justice is wrestling with this
>> question, along with other questions about how does a web site provider know
>> for sure that they've made their website accessible.
>>>>
>>>> I'm pretty sure that however the rules come down, they're going to
>> seriously hamper webapp providers and in turn give us relief from much of
>> the inaccessible web content we as blind people have to deal with on a very
>> regular basis. I hope we keep both sides of this in mind when the DOJ
>> developes, releases, and begins enforcing these regulations.
>>>>
>>>> In other words, I'm hoping we can figure out a way to go after Google,
>> facebook, United airlines, and the US government to get their web sites more
>> accessible; without also harassing tech startups, non-profits, and my high
>> cusin who just put up a really cool visualization tool for how he and his
>> friends listen to music.
>>>>
>>>> Take Care All,
>>>>
>>>> JIm
>>>>
>>>> On Mon, Feb 24, 2014 at 12:26:25PM -0600, Gary Wunder wrote:
>>>>> Dear Mike:
>>>>>
>>>>> Perhaps one of the things we need to address is whether or not
>>>>> making things screen reader compatible truly does limit innovation.
>>>>> Is screen reader technology so far behind state-of-the-art
>>>>> technology that this is the case, or are we talking about the
>>>>> failure to add a few lines of code in this new technology that
>>>>> makes it play well with the assistive technology we need? I lack
>>>>> the expertise to answer this question, but it seems to me to be all
>>>>> important. We go to Congress each year with the message that
>>>>> accessibility is easy and doable. I have never heard the software
>>>>> companies argue to the contrary. What I do often hear from software
>>>>> developers is that it is too costly to go back and modify their
>>>>> legacy code but that new development will certainly incorporate
>>>>> accessibility. Only recently have I heard the idea that demanding
>>>>> accessibility threatens innovation. Can someone with some expertise
>>>>> in state-of-the-art coding and state-of-the-art screen reader
>>>>> technology set me straight. It seems to me that this argument, if
>>>>> true, changes where we need to place our emphasis. If it is false,
>>>>> it needs to be revealed as such. If it is true, then we need to place
>> more emphasis on bringing the screen readers into the second decade of the
>> twenty-first century.
>>>>>
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Mike
>>>>> Freeman
>>>>> Sent: Monday, February 24, 2014 11:00 AM
>>>>> To: 'NFB in Computer Science Mailing List'
>>>>> Subject: Re: [nfbcs] Office 2013: Very Much A Work In Progress
>>>>>
>>>>> Gary:
>>>>>
>>>>> I confess that I am beginning to think we're running up against
>>>>> something we ought to be familiar with, being Federationists, but
>>>>> that we don't want to face. That "something" is that we, the blind,
>>>>> are a minority. This is something we're going to come up against
>>>>> more and more as the general universe seeks bling more than
>> information.
>>>>>
>>>>> I confess that I'm beginning to suspect that unless and/or until we
>>>>> come up with *absolutely* iron-clad legislation that, in effect,
>>>>> limits what software vendors are allowed to do to those things
>>>>> wherein we can guarantee accessibility -- in effect, limiting
>>>>> innovation -- something which I obviously know won't happen --
>>>>> we're going to be behind the eight ball even with vendors who claim to
>> put accessibility first.
>>>>>
>>>>> I think more and more we will find ourselves forced to old,
>>>>> tried-and-true but much-forgotten and much-maligned strategies --
>>>>> such as -- gasp -- use of readers.
>>>>>
>>>>> I believe Deborah Kent-Stein and I talked about this a while back
>>>>> and *she* thinks we'll eventually have to come round to a
>>>>> TapTapSee-like app that allows us to point a camera at
>>>>> indecipherable screens and have someone tell us what's going on. I
>>>>> don't think even that would work as corporations would frown on
>>>>> their networks being used for such things and might balk at the
>> possibilities of theft of corporate secrets or intellectual property.
>>>>>
>>>>> Mike Freeman
>>>>>
>>>>>
>>>>> -----Original Message-----
>>>>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Gary
>>>>> Wunder
>>>>> Sent: Monday, February 24, 2014 7:09 AM
>>>>> To: 'NFB in Computer Science Mailing List'
>>>>> Subject: Re: [nfbcs] Office 2013: Very Much A Work In Progress
>>>>>
>>>>> I still respectfully suggest that we put Microsoft on the agenda
>>>>> and try to find out why accessibility always seems to be at the
>>>>> bottom of their priority list. Did screen reader developers have a
>>>>> look at this before it came on the market? Why is it that we were
>>>>> still wrestling with problems in Outlook 2007 when Outlook 2010 hit
>>>>> the market? Is there any kind of consistency between the statement
>>>>> "computing for all" and the kind of release strategy we see from
>> Microsoft?
>>>>>
>>>>> Gary
>>>>>
>>>>> -----Original Message-----
>>>>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of
>>>>> Elizabeth Campbell
>>>>> Sent: Monday, February 24, 2014 8:08 AM
>>>>> To: 'NFB in Computer Science Mailing List'
>>>>> Subject: Re: [nfbcs] Office 2013: Very Much A Work In Progress
>>>>>
>>>>> Hello Curtis,
>>>>>
>>>>> I am using Office 13 at home on my laptop running Windows 8. I must
>>>>> confess that I did not have the installation headaches as I
>>>>> purchased my system and software from Bestbuy. The store in my area
>>>>> has a very helpful geek squad, and I explained that I wanted to put
>>>>> the computer through its paces using JAWS and so forth before I
>> purchased it.
>>>>> The Best Buy folks took care of all of the installations for me as
>>>>> I purchased a year of tech support for my devices.
>>>>> I primarily use office 13 for Outlook and Word. I am a fan of
>>>>> outlook, so I was very disappointed to see that it often crashes,
>>>>> sometimes while I'm reading or writing a message then mysteriously
>>>>> restarts. IN Word, I haven't used the return address features since
>>>>> I'm creating documents for use at home or at work, and I send 99
>> percent of my correspondences via email.
>>>>> However, I've had a lot of frustration accessing documents that are
>>>>> protected.
>>>>> JAWS will start reading the file and then stop. I believe what
>>>>> happens is that Word shuts down and then restarts because I get a
>>>>> prompt about recovering files which I can never find.
>>>>> Interestingly enough, I ran in to this problem last week when
>>>>> accessing some documents for a Newsline seminar.
>>>>> I believe there is a way to unprotect files, but I haven't found it
>> yet.
>>>>> Curtis, I agree that Office 13 is very much a work in progress, and
>>>>> I hope Microsoft does come out with a service pack that will repair
>>>>> these bugs which make it almost impossible to use Office reliably.
>>>>> At work, I an using Windows 7 and Office 2007,and I haven't had the
>>>>> same frustrations.
>>>>>
>>>>> Best regards.
>>>>>
>>>>> Liz Campbell
>>>>>
>>>>> -----Original Message-----
>>>>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Curtis
>>>>> Chong
>>>>> Sent: Friday, February 21, 2014 11:06 AM
>>>>> To: nfbcs at nfbnet.org
>>>>> Subject: [nfbcs] Office 2013: Very Much A Work In Progress
>>>>>
>>>>> Greetings and felicitations:
>>>>>
>>>>> Early this month, I took the rather bold step of upgrading from
>>>>> Microsoft Office 2010 to Microsoft Office 2013. I am running the
>>>>> 64-bit version of Windows 7 Professional.
>>>>>
>>>>> I am interested in hearing from anyone out there who has taken a
>>>>> similar journey. Permit me to provide a brief summary of my experiences
>> so far.
>>>>>
>>>>> To begin with, the upgrade was not at all a trouble-free
>>>>> experience. The first thing that Office 2013 wanted me to do was to
>>>>> link to either an existing or new Microsoft account. There appears no
>> way to avoid this step.
>>>>> Since I had a Microsoft account (which I had never used for years
>>>>> and years) I had to spend considerable time trying to get my
>>>>> password back. This was only the first problem.
>>>>>
>>>>> Then, Office wanted to set up Sky Drive on my computer, which I
>>>>> allowed at first and have since removed.
>>>>>
>>>>> After the install was finished (hours of work), I tried starting
>>>>> Word. Right away, I received a message (which was not spoken by
>>>>> JAWS for Windows) indicating that the program had stopped working.
>>>>> There seemed to be no way around this problem. In the end, I had to
>>>>> contact Microsoft Support over the telephone so that someone could
>>>>> remote into my computer and run some kind of a repair.
>>>>>
>>>>> While I am now using Microsoft Office 2013 to do real work, I must
>>>>> point out that using this software is not without its problems. For
>>>>> one thing, there are many situations during which JAWS goes silent
>>>>> and during which one simply has to wait for something to happen.
>>>>> For another, there are frequent instances when either Word or
>>>>> Outlook will crash and then recover--all in complete silence (from a
>> nonvisual access standpoint).
>>>>>
>>>>> I don't know about the rest of you, but one strategy which I often
>>>>> use is to open a master document from Windows Explorer, bringing it
>>>>> into Word, then save the document under a different name so that I
>>>>> can work on it. On my system right now, there is no way to do this
>>>>> anymore. As soon as I hit F12 to invoke the "Save As..." dialog, Word
>> will immediately crash.
>>>>> Interestingly, this does not happen on the Office 2013 system I am
>>>>> using at work. Go figure.
>>>>>
>>>>> There are two other problems worth mentioning. First, in Word, the
>>>>> return and delivery address edit boxes in the Envelopes dialog are
>>>>> not accessible with any screen access program. You simply cannot
>>>>> read the text that may (or may not) be in these boxes. Secondly, in
>>>>> Outlook 2013, the Signature dialog's edit box is just as
>>>>> inaccessible to a nonvisual user as the Envelopes edit boxes in Word.
>>>>>
>>>>> These days, for new users, it is just about impossible to acquire
>>>>> Office 2010. This is most unfortunate inasmuch as I consider Office
>>>>> 2013 to be very much a work in progress. I very much am looking
>>>>> forward to a service pack on this from Microsoft.
>>>>>
>>>>> Cordially,
>>>>>
>>>>> Curtis Chong
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nfbcs mailing list
>>>>> nfbcs at nfbnet.org
>>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/batescampbell%40
>>>>> charter.n
>>>>> et
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nfbcs mailing list
>>>>> nfbcs at nfbnet.org
>>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/gwunder%40earthl
>>>>> ink.net
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nfbcs mailing list
>>>>> nfbcs at nfbnet.org
>>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/k7uij%40panix.co
>>>>> m
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nfbcs mailing list
>>>>> nfbcs at nfbnet.org
>>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/gwunder%40earthl
>>>>> ink.net
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> nfbcs mailing list
>>>>> nfbcs at nfbnet.org
>>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jbar%40barcore.c
>>>>> om
>>>>>
>>>>
>>>> _______________________________________________
>>>> nfbcs mailing list
>>>> nfbcs at nfbnet.org
>>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jheim%40math.wisc
>>>> .edu
>>>>
>>>
>>> _______________________________________________
>>> nfbcs mailing list
>>> nfbcs at nfbnet.org
>>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>>> To unsubscribe, change your list options or get your account info for
>> nfbcs:
>>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jbar%40barcore.com
>>>
>>
>> _______________________________________________
>> nfbcs mailing list
>> nfbcs at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>> To unsubscribe, change your list options or get your account info for nfbcs:
>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/lhwayland%40sbcglobal.net
>>
>>
>> _______________________________________________
>> nfbcs mailing list
>> nfbcs at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>> To unsubscribe, change your list options or get your account info for nfbcs:
>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jbar%40barcore.com
>>
>
> _______________________________________________
> nfbcs mailing list
> nfbcs at nfbnet.org
> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> To unsubscribe, change your list options or get your account info for nfbcs:
> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jheim%40math.wisc.edu
>
--
---
John G. Heim, 608-263-4189, jheim at math.wisc.edu
More information about the NFBCS
mailing list