[Blindmath] Making Mathtype documents in Word accessible

Mon Dec 5 01:14:52 UTC 2011

Just a quick thing on what you say about navigation and what to show in Braille, I would possibly say that the full equation (or as much as you can) is shown at all times and like visually a cursor moves through it. Possibly the cursor would be under the entire element you are on. I possibly would really say this sort of thing should be done by the screen reader, the application should say "this is what is focused, assistive technology you choose how the user wants it presented, may it be speech, Braille, or something else".

However I get the feeling you won't get that sort of help from screen readers so you may be stuck with doing the best you can on your own. I don't know whether you would be permitted to do as part of the work make contributions to NVDA which may help in getting the best solution.

Michael Whapples
On 4 Dec 2011, at 07:34, Neil Soiffer wrote:

> We are working on navigation as part of a grant that we received to improve
> math accessibility.  My plans for braille are that as one navigates the
> math (not the words nor the braille), that MathPlayer will generate the
> appropriate speech and also the appropriate math braille code that
> corresponds to it.  Eg, if you navigate into the numerator of a fraction
> and the numerator is spoken, the math braille code corresponding to that
> numerator will be what is available for the screen reader to put on the
> braille display.
> 
> There are lots of details to be worked out though.  Eg, navigating into the
> numerator will likely generate a speech string like "In numerator, x
> squared plus y squared".  I don't know what, if anything, we will generate
> for braille that corresponds to "in numerator".  Once we get to the point
> of detailed design, I'll be asking the list for suggestions as to what
> should happen.
> 
> Neil Soiffer
> Senior Scientist
> Design Science, Inc.
> www.dessci.com
> ~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, Equation Editor ~
> 
> 
> On Sat, Dec 3, 2011 at 3:38 PM, Michael Whapples <mwhapples at aim.com> wrote:
> 
>> Hello,
>> One thing I would question of how it would work, if NVDA were to produce
>> Braille from MathML, how would cursoring work? Currently as one cursors
>> they move through the words of the speech, as Braille shows that it can
>> match the cursor position. If using liblouisxml then Braille is coming from
>> a different source (the original MathML rather than the text which is used
>> for speech (I think what NVDA finds from MSAA). So even if liblouisxml can
>> synchronise the cursor between MathML and Braille (not sure either way on
>> this) it would still be stuck about synchronising it with the speech/MSAA
>> words. Also when cursoring, which would you be cursoring through (eg. one
>> press of the cursor relates to a Braille cell or does it relate to
>> something in speech or may be best something to do with the maths and
>> probably found from the original MathML).
>> 
>> May be a different approach, however needing support from MathPlayer, may
>> be there should be a simple character based notation (eg. like triangle or
>> lambda). The only thing I wonder about is whether it would be possible to
>> get good speech and Braille from such a notation. I would guess the Braille
>> would be possible as its character based you could probably even do it in
>> liblouis (so not requiring NVDA to depend on liblouisxml as well).
>> 
>> Finally, yes I feel that as you say NVDA is probably the best hope and
>> also if NVDA can offer things above commercial offerings then it might make
>> them sit up and listen.
>> 
>> Michael Whapples
>> On 3 Dec 2011, at 22:02, Birkir R. Gunnarsson wrote:
>> 
>>> Hi all
>>> 
>>> Neil, the instructions work well on the web page you pointd to (or did
>>> for me, for Ben too I assume). One thing that bothers me slightly
>>> about them, is that they claim that a blind user needs sighted
>>> assistance the first time around. That is not correct, this can easily
>>> be done completely independently (which is great). Indicating one
>>> needs a sighted colleague to help the first time just always strikes
>>> me as very patronizing.
>>> I'd suggest changing this to (if you have problems with anyof these
>>> steps, please consult a sighted colleague).
>>> This is knit picky ;) and doesn't really matter, just thought I'd
>>> throw it out there.
>>> You can basically do 4 things with Word + MathType in terms of
>>> accessible output.
>>> 1. Convert all equations in word document to LaTeX and then try to
>>> read through it (Ben, I agree, it is a heck of a job trying to work
>>> with it, it is doable though).
>>> The advantage is that you can use refreshable braille display, for
>>> what it's worth, you can also print the text directly witha braille
>>> embosser and any software (PEF for instance).
>>> 
>>> 2. You can use TSS from ViewPlus or DBT from Duxbury to emboss the
>>> Word with MathType equations document to Nemeth or UeB math (or
>>> whatever braille codes they support, I believe additional DBT supports
>>> the French braille code).
>>> 
>>> 3. You can export the Word with Mathtype equation via MathPage to a
>>> webpage (.xht file) and then read it with IE + MathPlayer (missing
>>> refreshable braille support,other than just showing the strings that
>>> are spoken, does not really allow much in terms of working with it
>>> otherwise, be it copying and pasting, analyzing etc, but good for
>>> reading math).
>>> 
>>> 4. Use the MathDaisy and save As daisy plug ins to make a Daisy MathML
>>> document that REadHere, EasyReader, and a few other math aware Daisy
>>> players can read. I thought REadHere/GH Player had some sort of
>>> refreshable braille support even, but I never got it to work. Really
>>> all that happens is that the math in the Word document is converted to
>>> spech, so if you are using speech only, you should have no problem
>>> grabbing that Daisy file and listening to it on VR Stream or
>>> compatible Daisy player .. I haven't tried it, I will tonight
>>> actually.
>>> 
>>> None of these options are perfect, but together they allow you some
>>> flexibility, based on what you need to do exactly.
>>> 
>>> A more readable TeX output of MathType equations would be great. I
>>> thought HR-TeX was more widely used and there were interpreters out
>>> there to translate from MathML to HR-TeX, but on more Google search,
>>> it appears the format is mostly theoretical and only supported by
>>> Infty, at least I found no open source interpreter or anything of the
>>> kind for it, though my Google searching abilities may have let me down
>>> this once.
>>> Once MathPlayer supports Word natively, that will be a huge step
>>> forward and Neil and Co are working on it, dilligently I trust *grin*.
>>> It's still not a minor project, so nothing we'll see this year, at least.
>>> A refreshable braille code support of read math equations with
>>> MathPlayer would be another cool feature. and with NvDA finally
>>> support MathPlayer 3, I am hoping that perhaps there is an avenue to
>>> look into this, if it were possible to use the work of the LiblouisXML
>>> team to produce refreshable Nemeth/Nubs math on your display whilst
>>> MathPlayer reads the math questions concurrently to you, I think that
>>> would be a fantastic scenario.
>>> I believe this is nearly impossible with other screen readers, since
>>> we'd have to beg them to do it, and convincing these manufacturers to
>>> put development efforst into math has not been particularly successful
>>> (though I am by no means suggesting we should stop pestering them,
>>> those of us who do, but if we can get together and make a superior
>>> solution via free and open source software, it might jolt them into
>>> life).
>>> 
>>> I am also hopeful that the EPUB3 support and rendering of MathML may
>>> cause a revolution in math accessibility, if we can harness it, it's
>>> just a bit too early for me to understand exactly how this works, and
>>> it depends how those who manufacture devices compatible with EPUB3
>>> implement the MathML part of the specs.
>>> But, that's way off topic and subject for its own discussion. I just
>>> wanted to point out the 4 different uses of Word with MathType, in
>>> hope it may clarify some things for some of you.
>>> And if I am wrong, or missing something and need a clarification
>>> myself, then by all means, post it to this list and it will be duely
>>> noted.
>>> 
>>> 
>>> On 12/3/11, Neil Soiffer <NeilS at dessci.com> wrote:
>>>> Yes, it is horrible.  There is a feature request to improve it and that
>> is
>>>> being looked at for MathType 6.8, but no decision has been made yet as
>> to
>>>> whether it will make it into 6.8 (there are lots of competing requests).
>>>> 
>>>> Neil Soiffer
>>>> Senior Scientist
>>>> Design Science, Inc.
>>>> www.dessci.com
>>>> ~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, Equation Editor ~
>>>> 
>>>> 
>>>> 
>>>> On Sat, Dec 3, 2011 at 10:03 AM, Andrew Stacey
>>>> <andrew.stacey at math.ntnu.no>wrote:
>>>> 
>>>>> The first thing that I want to say is that is the most awful LaTeX
>> that I
>>>>> have
>>>>> ever seen!  If that is what the process produces as it's LaTeX "output"
>>>>> then
>>>>> I'm not surprised that you are looking for another system.
>>>>> 
>>>>> That said, this does look like the sort of thing where my system might
>> be
>>>>> able
>>>>> to help as the idea behind it is to convert LaTeX to "something else"
>> by
>>>>> using
>>>>> LaTeX itself to do the conversion.  So all that would need doing is for
>>>>> someone to tell LaTeX that you want, say, \int to be written as
>> "Integral"
>>>>> and
>>>>> then you could run LaTeX on the equations to convert them to something
>>>>> suitable.
>>>>> 
>>>>> This would take a bit of work to get to work properly, though, as, for
>>>>> example, LaTeX likes to run on whole documents, not snippets.  Also, it
>>>>> would
>>>>> need considerable configuring at the start.
>>>>> 
>>>>> As Neil's suggested another way that is working for you then I don't
>>>>> recommend
>>>>> this.  I mention it in case there are others for whom Neil's method
>>>>> wouldn't
>>>>> work.
>>>>> 
>>>>> Andrew
>>>>> 
>>>>> _______________________________________________
>>>>> 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/birkir.gunnarsson%40gmail.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/mwhapples%40aim.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/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/mwhapples%40aim.com