[Blindmath] MathJax Accessibility: math/STEM universal access/design

Tue Nov 8 17:22:35 UTC 2011

Steve
(the following are merely my thoughts, if some of them are inaccurate
I'm free to be corrected)

first the negatives:
True, but this is also been the message from DSI for 5 or 8 years, and
it hasn't worked either.
I think there are very few blind people who pursue math or STEM
subjects (well, I know this for a fact), if we further fragment this
group by screen reader used, then expect them to try and figure out
the MathType vs MathPlayer setup, understand the inherent issues with
braille support and the interplay between the braille display, the
screen reader vendor, the browser, and MathPlayer, and then give
constructive feedback to screen reader vendors, I just don't think
it'll ever happen, way too few users to have an impact, and most
people either just learn LaTeX and get by with it .. because it is
linear, it is supported by braille devices, you can use the same
language for reading and writing math, you can use any text editor in
theory.
I, as a blind user, have major issues with not being able to even get
to the MathPlayer menu in IE without using some Jaws cursor and
routing keys. If this is an  application intended for me as a blind
user, why do I need to go the extra mile to accessits menus, by all
standards that would be defined as inaccessible. As far as I know,
there are also no blind users involved with internal development or
testing of MathPlayer,, though the blind community is willing, and
working with Neil is always a pleasure.
Of course once we have public releases of MathPlayer we, as a
community, can help out, and I sure intend to do my best in that
regard.

The other issue is that the one source, core accessibility environment
for creating accessible math is Microsoft Word with MathType, but
accessing math in that environment itself is not even accessible to
me. I have to choose between a variety of partial solutions to read
the output, such as to emboss via other software (if I have an
embosser), get MathPlayer feedback by exporting to IE and lose braille
support but get voice reading (at least once I adjust the security
options in IE and get it to work), or I can revert back to LaTeX so I
can work within the document itself (toggle TeX).
It's hard to sell a product that works if and only if (insert work
arounds here to get things to work). Also, as a user, I feel a bit
weird asking my screen reader vendor to make a product by some other
company accessible.
However technically valid the standard design science answer is, it is
frustrating to get the "I know this doesn't work, talk to the other
company and ask them to get our product to work" response all the
time.
Just try that if you had, say, a browser that didn't work, but you ask
users who are unhappy with it to talk to Microsoft, because it's
really their fault. It may technically be true, but the user stil
feels short changed.

Due to the reasons noted above, I don't think we'll ever see a
significant influx of new users with the current situation. When math
accessibility is this hard and this fragmented, there is no reason why
we should see more students pursuing STeM subjects this year or next
year, than we've seen in the last 5 years. Something has to change in
order for them to come in and succeed.

the good:
Fortunately there are changes coming:
What is happening that is positive?
With the update to MathPlayer and its compatibility to all pages
created with MathJax, the amount of material that it makes available
in an accessible format has the potential to increase by hundreds or
thousands of percentage points, bringing something truly new,
innovative and impressive to the blind community interested in math.
(build it and they will come, if you will).
I am excited to see MathPlayer 3, and test it out with pages made by
MathJax. If we can get it to work with all MathJax created pages, we
truly have a game changer in terms of the amount of material available
to users. This will increase math interest in the group, I believe.
I also know that MathPlayer within Word should become accessible (at
least according to the ETS grant that was posted on the DSI blog a few
weeks back) though it is still a couple of years away, which is a very
long time in the computer world. What is still missing from this
picture is accessible MathType input, which I really think will
impress, but, as far as I understand, that is not on the table. This
means blind users will still have to learn LaTeX if they want to write
math, and once they've learnt that, why not just use it for reading
math as well, due to the advantages listed above and the difficulty of
getting any meaningful refreshable braille in any other format.

MathML is becoming a core technology in HTML5 (I believe the official
"release" of HTML5 is scheduled for 2014or 2015, so it's still a few
years away, but this time around there is a comprihensive set of
browser compatibility tests, which should imply that browser support
for MathML, given that they pass the HTML5 tests, will be more uniform
and widespread. MathML is also a core technology of EPUB3, and
hopefully that will convince device manufacturers such as Apple (iPad)
and Amazon (Kindle) to stop relying on graphics for math and, instead,
create client software for parsing MathML and generating the graphics
themselves .. which incidentally makes the information available to
software that wants to convert MathML into something else (sound,
braille, anything really). Add to that on-going research into turning
touch screens into braille input devices and braille or tactile
feedback through touchscreens, that really has the possibility of
revolutionizing math accessibility, by giving users an idea of the
spatial layout of math for the first time, granted this is probably
still a few years off at best.

The other idea, which probably is not going to happen in the short
term, is to utilize the open source resources out there. There is
NVdA, which still does not work with MathPlayer (may be us users can
push for it), there is LibLouis with MathML to Nemeth translation
capabilities, that has to be integrated (again it seems like NVDA is a
logical choice here, seeing as its braille support is built on
LibLouis already).
Given what I have read about how little it would take to add MP
support to NVDA, and to add LibLouis braille support to MathPlayer, it
seems these could be relatively small steps with relatively large
payoffs (both of these things are mentioned in articles on
www.access2science.com, though no technical details are given).
Once you can point to a free and open source, high quality solution
that provide free and superior math access, you really have something
in hand to "bother" the screen reader vendors with, and NVDA can offer
a unique functionality that will make it a clear choice over the
traditional screen readers for many universities and other academic
settings, which may improve their standing and their cashflow.

What would, of course, be best, is if MathPlayer was open source, or
we had a similar open source solution to push. With open source code,
it's easier to get users involved, to get research and acaademic
funding and support, and to migrate solutions for application or
platform x to other areas. I think MathJax being open source was
brilliant, and the benefits of it are becoming more and more obvious.

Obviously there are downsides too. I have seen very badly supervised
and executed open source projects that fizzled out completely, it
really takes a good central planning and organization to make open
source work, and a large enough community of testers and developers,
and this might be difficult with mathematics. Though I would hope that
university researchers and students could be involved .. after all
there is a lot of people in computer science, engineering and math,
that are looking for big projects for their degree completions. I
think working around a commercial product to increase its user base is
not something that you can easily do as an academic, or as part of an
academic project, at least your funding resources will be severely
limited. Working on an open source project however, such as how to
best turn MathML into speech, definitely has the sound of a successful
PHD project to it.

I am not suggesting that this solution is even on the table, or that
there aren't a myriad of arguments against this. Design Science owns
MathPlayer, and they are, of course,  free to do whatever they want
with it. Neil has worked hard on it, and really made a huge difference
for anyone interested in math as a blind student. My posthere is
simply to underline that it could be taken even further, not to
criticize what has been done, because this product has revolutionized
access to mathematics for blind students, no questions about that.

I just really think it needs the leverage other open source
developments, and that would be most easily done if it were open
source itself.
It would require smart and dedicated community of blind developers to
help it along. I know I don't have the time, or very little time,
though I would help as I could.
I'd imagine the same applies to other people on this list.

I hope that MathPlayer3 and the accessible MathPlayer Word integration
will provide significant upgrades, and that hopefully NVDA integration
can happen, and then we could see a connect between NVDA, MathPlayer
and LibLouis to provide better braille support, both through
refreshable braille, and through the Braille Blaster project for a
free and open source embossing of mathematics.
That would be the next step in the revolution, and would change this
status quo, that really has gone on for too long.
Cheers
-B

On 11/8/11, Noble,Stephen L. <steve.noble at louisville.edu> wrote:
> Of course, one has to remember that the original concept behind MathPlayer
> was to turn math into speech so that screen reader applications (like JAWS)
> could read the math. The only way to turn math symbols into speech is to use
> words to convey that information. The only way to provide any additional
> layout or spatial information in speech is to use more words, for instance
> to announce the column and row location of a data element in a table...but
> again, all of that is done with words.
>
> Now, I will add that getting Nemeth support in refreshable braille out of
> MathML content has been "potentially" possible for many years, but the
> vendors who implement this technology have not bothered to invest the
> development time into their applications. If you want to see this happen,
> then lots of customers need to start complaining to places like Freedom
> Scientific (just an example since I mentioned JAWS) and tell them this is a
> priority. Otherwise, we'll still be having this same discussion five years
> from now.
>
> --Steve Noble
> steve.noble at louisville.edu
> 502-969-3088
>
> -----Original Message-----
> From: blindmath-bounces at nfbnet.org [mailto:blindmath-bounces at nfbnet.org] On
> Behalf Of Kevin Chao
> Sent: Monday, November 07, 2011 5:06 PM
> To: Blind Math list for those interested in mathematics
> Subject: Re: [Blindmath] MathJax Accessibility: math/STEM universal
> access/design
>
> Will MathPlayer 3 address the fact that all the math/STEM symbols/operators
> in 2.2 are converted into words, all spatial layout/structure/relations not
> being conveyed due to all information being converted into words, and there
> not being proper braille support (Nemeth)? Each and every one of these three
> key points are absolutely essential and critical to proper and truly
> accessible math/STEM readable content.
>
> I think that the work that Design Science, MathPlayer has done is a great
> start, but it doesn't provide a great experience, especially when it comes
> to being able to have an overview, drill down, and navigate/review. Not to
> also mention the inability to create/edit work in an integrated/imbedded
> way, such as homework, notes, and exams within the same environment that is
> already being used by education or work, such as a LMS/CMS of some kind.
> Also, the very restrictive system requirement of MathPlayer 2.2 plug-in only
> working with IE, what about Windows Firefox and Chrome or Mac Safari,
> Webkit, Chrome, etc.? This is where I think it would be great if MathJax has
> a standard specifying accessibility, which would work across platforms and
> browsers.
>
> Thanks,
>
> Kevin
>
> On 11/6/11, Neil Soiffer <NeilS at dessci.com> wrote:
>> MathJax works with MathPlayer to provide accessibility if you use IE.
>> There are some issues with MathJax and MathPlayer 2 if people didn't
>> configure their site to allow users to choose the renderer, but most
>> of those have been resolved with MathPlayer 3 (which might be
>> available next week).
>>
>> MathJax is a great solution which has already greatly increased the
>> amount of math that is accessible on the web.  Design Science is one
>> of the founding sponsers for MathJax work and we are working to make
>> sure the math rendered by it is accessible.
>>
>>  Neil Soiffer
>> Senior Scientist
>> Design Science, Inc.
>> www.dessci.com
>> ~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, Equation Editor
>> ~
>>
>>
>> On Sun, Nov 6, 2011 at 3:19 PM, Kevin Chao <kevinchao89 at gmail.com> wrote:
>>
>>> MathJax is an open source JavaScript display engine for mathematics
>>> that works in all modern browsers.
>>> No more setup for readers. No more browser plugins. No more font
>>> installations... It just works.
>>> Accept following sources:
>>> LaTeX
>>> MathML
>>>
>>> http://www.mathjax.org/
>>>
>>> Unfortunately, there's not an MathJax accessibility API, which web
>>> browsers can use, which will allow assistive technologies, such as
>>> screen readers to work with beautiful math in all browsers. As far as
>>> I'm aware, there's no MathJax accessibility in Windows IE9/8, Chrome
>>> 15-17, Firefox 4-10; Mac OS X Safari, Webkit, Lightning, and/or
>>> Chrome; iOS 5 Safari; Android Mobile Accessibility Web browser or
>>> Ideal Web Reader.
>>>
>>> MathJax has the potential to make math/STEM accessibility truly
>>> universal and not speicific to MathML and IE8, MathPlayer 2.2, and
>>> JAWS 12.
>>>
>>> It has the potential to completely revolutionize math/STEM
>>> accessibility for all, but I would like to know what's the best
>>> approach in MathJax including universal access/design, and it working
>>> for all? Perhaps, it should be a community effort, where we all
>>> contact MathJax, create a petition, or some other campaign effort?
>>>
>>> Thanks,
>>>
>>> Kevin
>>>
>>> _______________________________________________
>>> Blindmath mailing list
>>> Blindmath at nfbnet.org
>>> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
>>> To unsubscribe, change your list options or get your account info for
>>> Blindmath:
>>> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/neils%40dessci
>>> .com
>>>
>>
>
> _______________________________________________
> Blindmath mailing list
> Blindmath at nfbnet.org
> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
> To unsubscribe, change your list options or get your account info for
> Blindmath:
> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/steve.noble%40louisville.edu
>
>
>
> _______________________________________________
> Blindmath mailing list
> Blindmath at nfbnet.org
> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
> To unsubscribe, change your list options or get your account info for
> Blindmath:
> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/birkir.gunnarsson%40gmail.com
>