[Blindmath] Translating documents with MathML into Nemeth

[Birkir R. Gunnarsson]

Neil

Good, I knew as much, but was probably privy to that information, so I
could not say it.
I'll be curious to see how well it works in practice.
After all, if some issues were to come up, Liblouis is open source, so
fixing them is something we can actually do, rather than pray, email
and call to Freedom Scientific (as long as we get Jaws to make the
appropriate calls).


On 7/18/11, Neil Soiffer <NeilS at dessci.com> wrote:
> I want to correct a misunderstanding about MathPlayer.  MathPlayer 3 will
> incorporate liblouis in it and will convert MathML on the fly (in real time)
> to Nemeth code.  No pre-translation is needed.  Screen readers need to ask
> for the braille (currently none do), but the functionality will be in
> MathPlayer 3.
>
> Neil Soiffer
> Senior Scientist
> Design Science, Inc.
> www.dessci.com
> ~ Makers of MathType, MathFlow, MathPlayer, MathDaisy, Equation Editor ~
>
>
>
>
> On Mon, Jul 18, 2011 at 9:16 AM, Susan Jolly <easjolly at ix.netcom.com> wrote:
>
>> Let me make sure we aren't talking apples and oranges here.
>>
>> There are two ways to translate print to braille, which I'll call realtime
>> processing and batch processing.
>>
>> Realtime processing is used only with braille displays and requires
>> software that can translate print electronic documents such as web pages
>> in
>> realtime. Screenreaders often have this capability as do some models of
>> braille displays. This realtime processing essentially works line-by-line
>> so
>> it does not (to my knowledge) produce proper document formatting or handle
>> planar layouts.
>>
>> However, the realtime translators I am aware of only handle narrative or
>> literary text. I do not know of any software than can provide realtime
>> translation of electronic mathematical expressions (whatever format) to
>> Nemeth (or other) braille math.  I'm not sure this would even make sense
>> in
>> all situations.
>>
>> My understanding is that it is the intent of apps like MathPlayer to
>> pre-translate math expressions from MathML to Nemeth with the idea that a
>> properly configured screenreader could pass this pre-translated braille
>> math
>> directly to a braille display while at the same time providing the correct
>> realtime translation of any surrounding non-mathematical text.  While this
>> would likely be readable, it would not necessarily follow all of the
>> Nemeth
>> rules including but not limited to the rules for linebreaking which depend
>> on both semantics and the cells per line in the braille display.
>>
>> Batch processing is used to generate electronic braille documents which
>> are
>> treated as paged media where the formatting depends on the specified cells
>> per line. The generated electronic braille document can be either
>> displayed
>> on a braille display or embossed.  (Of course, if one is reading
>> electronic
>> braille on a braille display, it is essential to ensure that this is done
>> in
>> translate off mode since it is already translated!) My understanding is
>> that
>> there can be problems if one uses a braille display to read an electronic
>> braille document that has been formatted for a line length different from
>> that of the display but I could be wrong.  I suspect this would be more of
>> a
>> problem for mathematical expressions than for text.
>>
>> So now the question is whether there are batch processors that can
>> translate from LaTeX or from HTML+MathML to Nemeth or other braille math
>> codes?  There are a number of processors including DBT, TSS, and liblouis
>> which can do this to some extent.  However, I don't think there is any
>> current translation software that can produce perfect braille math
>> automatically. In fact, there isn't any current software that can produce
>> perfect contracted American English braille.
>>
>> I'm getting ahead of myself here but I was going to suggest that the
>> members of this list set up some well-designed test input for these
>> applications. Good test input is preferably short which means that a lot
>> of
>> separate input files are needed to provide complete coverage.
>>
>> Susan
>>
>> ______________________________**_________________
>> Blindmath mailing list
>> Blindmath at nfbnet.org
>> http://www.nfbnet.org/mailman/**listinfo/blindmath_nfbnet.org<http://www.nfbnet.org/mailman/listinfo/blindmath_nfbnet.org>
>> To unsubscribe, change your list options or get your account info for
>> Blindmath:
>> http://www.nfbnet.org/mailman/**options/blindmath_nfbnet.org/**
>> neils%40dessci.com<http://www.nfbnet.org/mailman/options/blindmath_nfbnet.org/neils%40dessci.com>
>>
> _______________________________________________
> Blindmath mailing list
> Blindmath at nfbnet.org
> http://www.nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
> To unsubscribe, change your list options or get your account info for
> Blindmath:
> http://www.nfbnet.org/mailman/options/blindmath_nfbnet.org/birkir.gunnarsson%40gmail.com
>